Challenges and opportunities in the management of severe pneumonia. Key concepts from the Seventh Annual Meeting of Spanish Experts 2025 Authors: Cristobal Rodríguez-Leal, Berta Laguna-Fonseca, Francisco Javier Candel, Montserrat Rodríguez-Aguirregabiria, Juan Manuel García-Lechuz, Fátima Galán-Sánchez, Cruz Soriano-Cuesta, Ángel Estella Published: 2026 ABSTRACT AbstractPneumonia is the leading cause of death from infection in the developed world. In recent years, researchers and healthcare professionals have worked tirelessly to address this issue. Multiresistance is a significant problem in severe pneumonia, so it is crucial to select the appropriate antibiotic at the correct dose and via the appropriate route of administration to ensure effective concentrations in the lung parenchyma. The diagnosis of pneumonia can also be challenging, so it is vital that imaging techniques, biomarkers, and microbiological methods are employed properly. An appropriate therapeutic approach is vital for managing severe community-acquired pneumonia, as it can be complex and is key to achieving optimal clinical outcomes. Nosocomial pneumonia also requires optimal management to avoid therapeutic failure and to improve outcomes, particularly in post-surgery patients. Not only bacteria, but also viruses and fungi can cause severe pneumonia, and appropriate diagnosis is essential for proper treatment. Finally, the correct use of new antimicrobial molecules can help to improve the clinical outcomes of these patients. To provide an update on these issues, a multidisciplinary team of Spanish experts met at the Seventh Annual Pneumonia Day Meeting, which was sponsored by the Grupo de Estudio de Infección en el Paciente Crítico (GEIPC) of the Sociedad Española de Enfermedades Infecciosas y Microbiología Clínica (SEIMC). This paper summarises the discussions held at the meeting and offers the latest insights on these topics, providing guidance on overcoming the challenges associated with pneumonia management.Keywords: Community acquired pneumonia; Aetiology; Management; Therapeutic failure; Healthcare-associated pneumonia; Ventilator-associated pneumonia, epidemiology, treatment Introduction Pulmonary infections are a significant cause of mortality and morbidity on a global scale. Community-acquired pneumonia (CAP) is the infectious disease with the highest mortality rate in industrialised countries. The term hospital-acquired pneumonia (HAP) refers to a pulmonary inflammatory and infectious process that begins more than 48 hours after hospital admission and was not present or incubating at the time of admission. Ventilator-associated pneumonia (VAP) is a significant subset of HAP that occurs in patients with an artificial airway after two to three days of tracheal intubation [1]. The objective of this paper is to provide a comprehensive and in-depth analysis of pneumonia, drawing upon a rigorous, plural, multidisciplinary, and up-to-date approach. The management of CAP, HAP and VAP is challenging and requires collaboration between experts to improve clinical outcomes. Despite the progress made in recent years in terms of knowledge and the management of these clinical entities, pneumonia remains a significant cause of morbidity, disability, economic burden, and mortality. Individualised management is imperative for severe clinical forms. When assessing the risk, it is essential to consider comorbidities, the presence of immunosuppression, and exposure to multiresistant microorganisms. Viruses, beyond SARS-CoV-2 and influenza, as well as fungi, represent other pathogens for which a high index of suspicion is required to facilitate a timely diagnosis and the initiation of appropriate therapy. Fortunately, new diagnostic procedures and antimicrobial molecules are now available to assist in this fight [2]. Since 2019, Spain has hosted an annual meeting that brings together major medical societies involved in the diagnosis and treatment of pneumonia. The seventh meeting was convened on 11-12 November 2025 [3]. Experts from leading medical specialties involved in its management presented the latest advances in their respective fields and addressed the challenges and opportunities raised in the previous paragraph. The purpose of this article is to provide a concise overview of the key concepts presented during the meeting, in the context of the scientific programme. Material and methods Design. The Study Group on Infection in Critically Ill Patients of the Spanish Society of Infectious Diseases and Clinical Microbiology (GEIPC-SEIMC) invited experts from various Spanish medical societies (listed in the affiliations of this document) involved in the diagnosis, treatment and prevention of CAP, HAP, and VAP, caused by bacteria, fungi or viruses. These experts were tasked with providing a narrative review of their respective fields of expertise and presenting their conclusions at different workshops during the Annual Meeting on Pneumonia. Search strategy. Between July and November 2025, the experts performed a bibliographic search of their corresponding topics in PubMed ((http://www.ncbi.nlm.nih.gov/pubmed/ , accessed on 1 November 2025), Embase (http://www.elsevier.com/online-tools/embase/ , accessed on 1 November 2025) and Scopus (http://www.elsevier.com/onlinetools/scopus , accessed on 1 November 2025). They selected the most relevant and recent articles for each topic to prepare a presentation for the meeting. Drafting. On 11 and 12 November 2025, two medical writers (CMRL and BLF) attended the meeting and subsequently drafted a manuscript summarising the main points discussed between November and December. Revision. Between January and February 2026, all the experts had the opportunity to read the complete text and propose objections and changes. Results Pharmacokinetic and pharmacodynamic (PK/PD) strategies in the treatment of multiresistant Gram-negative bacteria (MDR-GNB) Bacterial resistance is a global health problem that causes 700,000 deaths every year. It is a major source of healthcare expenditure, estimated at €1,500 million in Europe, and is predicted to become the leading cause of death by 2050 [4]. Antimicrobial stewardship programmes (PROA in Spanish) have four main objectives: to improve clinical outcomes, to reduce adverse reactions, to diminish resistance development and to optimize the cost-effectiveness relationship [5]. To achieve these objectives, it is essential to apply knowledge of pharmacokinetic (PK) and pharmacodynamic (PD) parameters appropriately. Pharmacokinetics can be defined as the temporal evolution of a drug and its metabolites in the body and can be summarized with the acronym LADME: liberation, administration, distribution, metabolism, and excretion. Key properties of antimicrobial drugs are [6,7]: Physicochemical properties. Hydrophilic drugs, such as beta-lactams, aminoglycosides, and glycopeptides, tend to have a small volume of distribution, to stay in the extracellular space, and to be more affected by augmented renal clearance (ARC). On the other hand, lipophilic drugs, such as quinolones, macrolides, rifampicin and linezolid, have a large distribution model and penetrate more in tissues. Physicochemical properties. Hydrophilic drugs, such as beta-lactams, aminoglycosides, and glycopeptides, tend to have a small volume of distribution, to stay in the extracellular space, and to be more affected by augmented renal clearance (ARC). On the other hand, lipophilic drugs, such as quinolones, macrolides, rifampicin and linezolid, have a large distribution model and penetrate more in tissues. Distribution volume. It is related to the loading dose. The larger the distribution volume, the greater the needed loading dose to achieve an adequate concentration. Distribution volume. It is related to the loading dose. The larger the distribution volume, the greater the needed loading dose to achieve an adequate concentration. Clearance. It is related to the maintenance doses. The faster the clearance, the greater the maintenance dose. Clearance. It is related to the maintenance doses. The faster the clearance, the greater the maintenance dose. Exposition. It is related to the bacterial susceptibility (PD parameter) and drug penetration into the target tissue. Exposition. It is related to the bacterial susceptibility (PD parameter) and drug penetration into the target tissue. Pharmacodynamics is defined as the effect of a drug on the body. This effect can be desired (therapeutic effect) or undesired (toxicity). For antimicrobial drugs, this effect is determined by the susceptibility of the microorganism, which is related to the minimal inhibitory concentration (MIC). Antibiotics can be classified into three families, as can be seen in Table 1 [8,9]. The MIC is not a fixed value, but rather a probability distribution. Consequently, the current ISO standards for MIC determination allow a variability of one dilution for its communication. Additionally, there are limitations in its determination, related to assay variability, inoculum size and incubation time. Therefore, current recommendations suggest that, when the isolate has a MIC below the epidemiological cut-off value (ECOFF), the target attainment should correspond to the ECOFF; however, when the MIC found is over the ECOFF, the target attainment should be set at the MIC plus two-fold dilutions [10]. When the MIC is unknown, treatment decisions must be based on the most probable microorganism and the local ecology, but it is not recommended to consider the worst scenario [11]. Table 1.Antibiotic families according to their PK/PD properties. Adapted from Pereira et al. and Onita et al. [8,9]Therapeutic effectKey parameterAntibacterial effectOther propertiesAntibioticsConcentration dependentCmax/MICRelated to Cmax, which must be above the MICProlonged PAEAminoglycosides, quinolonesConcentration and time dependentAUC/MICRelated to concentration AUC over time. The larger the AUC over the MIC, the better-Glycopeptides, linezolidTime dependent%T>MICRelated to the time that the antibiotic concentration is above the MIC. The longer, the betterMinimum or moderate PAEBeta-lactamsCmax: maximum concentration; MIC: minimum inhibitory concentration; PAE: post-antibiotic effect; AUC: area under the curve; %T: percentage of the time. Antibiotic families according to their PK/PD properties. Adapted from Pereira et al. and Onita et al. [8,9] Cmax: maximum concentration; MIC: minimum inhibitory concentration; PAE: post-antibiotic effect; AUC: area under the curve; %T: percentage of the time. In critical ill patients, there is considerable variability, both interindividual and intraindividual. The main characteristics to consider in these patients for treatment adjustment are [12,13]: Extreme body weight: obesity, anorexia, cachexia. Extreme body weight: obesity, anorexia, cachexia. Extreme ages: children and older adults. Extreme ages: children and older adults. Critical situations: ARC, renal replacement therapies (RRT), etc. Critical situations: ARC, renal replacement therapies (RRT), etc. Polypharmacy and derived interactions. Polypharmacy and derived interactions. Underrepresentation of critically ill patients in clinical trials. Underrepresentation of critically ill patients in clinical trials. In Table 2 these factors are grouped into categories related to the risk of underdosing or overdosing. Therefore, traditional formulas for calculating creatinine clearance are not reliable in critically ill patients, and it is recommended to determine the current creatinine clearance through its measurement in the patient’s blood and urine. In Figure 1, the formula for its calculation is shown, and a urine sample collected over a two-hour period is sufficient [14,15]. This approach allows the identification of ARC; however, underdiagnosis still occurs even with this method. ARC is defined as a creatinine clearance (ClCr) greater than 130 mL/min/1.73 m2. It is a prevalent situation (14–80%) among patients in the intensive care unit (ICU), and it is more frequent in younger patients, polytraumatized patients, patients with neurological conditions such as cranioencephalic trauma and subarachnoid haemorrhage, males, and patients with less severe disease [16]. Moreover, it is a dynamic condition, more frequent during the first week of ICU stay. It is a recognised situation for beta-lactam underdosing [17]. Conversely, patients with acute renal failure and those receiving RRT are also at risk for over or underdosing, so surveillance of drug levels is an important need for severe ill patients [18]. Table 2.Critical illness situations that impact antimicrobial pharmacokineticsSituationPharmacokinetic effectConsequenceCorrectionSeptic shock, pleural effusion, mediastinitis, ascites (third spacing)Increased volume of distributionUnderdosing riskTo increase loading doseIndwelling surgical drains, ECMO, RRTHypoalbuminemiaARCMajor burns, hypoalbuminemia, haemodynamically active drugsIncreased clearanceUnderdosing riskTo increase maintenance doses or to give doses more oftenAKIDecreased clearanceOverdosing riskTo decrease maintenance doses or to give doses less oftenRecovering AKIIncreased clearanceUnderdosing riskSurveillance of renal function recovery to avoid unintentional underdosingECMO: extracorporeal membrane oxygenation; RRT: renal replacement treatment; ARC: augmented renal clearance; AKI: acute kidney failure.Adapted from William et al. [13]. Critical illness situations that impact antimicrobial pharmacokinetics ECMO: extracorporeal membrane oxygenation; RRT: renal replacement treatment; ARC: augmented renal clearance; AKI: acute kidney failure. Adapted from William et al. [13]. Formula for creatinine clearance (ClCr) calculation. Adapted from Herrera-Gutiérrez et al. [14] Vol: volume; Cr: creatinine. Consequently, optimisation strategies are needed for these patients. These strategies are summarized in Figure 2. The first one relates to the PD characteristics displayed in Table 1. Particularly, in the case of beta-lactams, the therapeutic target is to maintain serum concentration above the MIC as much time as possible, as they are time-dependent antibiotics. For critically ill patients, the current recommendations are to maintain serum concentration above the MIC throughout the dosing interval [16,17,19,20]. This is easier to achieve by an intravenous (iv) loading dose followed by a continuous iv infusion, rather than by intermittent iv boluses [9,21]. Most of the beta-lactams are suitable for extended infusion, but carbapenems, and particularly imipenem, are not so appropriate for it, as can be seen in Table 3 [22]. There is substantial evidence supporting the use of continuous iv infusion of beta-lactams for patients with severe pneumonia, as it can improve their outcomes. Extended infusion of beta-lactams reduces mortality, increases clinical cure rates, and does not increase adverse event rates. It should be implemented in critically ill patients, in patients with sepsis or septic shock, and in those with severe pneumonia caused by Gram-negative bacteria. It may also be considered in cases of elevated MIC of the isolated microorganism, ARC, and in severe ill children. The evidence is less robust for non-critically ill patients and obese patients [17,21–28]. Optimisation strategies for severe pneumonia treatment Developed by the authors. Table 3.Stability and characteristics of different beta-lactams for extended infusionAntibioticMaximum safe perfusion durationCommentsPiperacillin-tazobactam, aztreonam, cefazolin, cefepime, ceftolozane, cefiderocolUntil 24 hoursVery appropriate for extended infusionCeftazidimeUntil 16 hoursCaution: it releases pyridineMeropenem4–8 hoursStability depending on concentration. Not ideal for prolonged continuous infusionImipenemLess than 4 hoursBad solubility and fast degradation. Not suitable for continuous infusionAdapted from Jenkins A, et al. [22]. Stability and characteristics of different beta-lactams for extended infusion Adapted from Jenkins A, et al. [22]. In the early initial phase (0 – 24 hours), it is recommended to administer a high iv loading dose before continuous iv infusion. In the early maintenance phase (24–48 hours), a continuous or prolonged infusion is advised, considering possible ARC and measuring CrCl using blood and urine samples. In the late maintenance phase (more than 48 hours), the clinician must adjust drug dosing according to renal function and RRT effluent rate, if applicable. In both the early and late maintenance phases, therapeutic drug monitoring (TDM) is highly valuable [7]. The second optimisation strategy is nebulised therapy. This approach enables higher concentrations of antibiotics to be delivered to the lung parenchyma without causing systemic toxicity. Compared to intravenous administration, it can improve PK/PD targets and has demonstrated higher microbiological eradication. Some evidence suggests that it is beneficial in preventing VAP. However, there is insufficient evidence to suggest a benefit in terms of clinical cure or higher survival rates. It is associated with a higher risk of bronchospasm, and the drug is distributed unevenly throughout the lung parenchyma. Furthermore, it lacks standardisation and intravenous formulations are often used off-label for this method of administration, which is associated with a higher incidence of bronchospasm. Therefore, routine use is discouraged. It may be considered as a rescue therapy for Gram-negative bacteria susceptible only to polymyxins or aminoglycosides, in combination with intravenous therapy. It is advisable to administer it in the inspiratory branch of the mechanical ventilation system, by a vibrating mesh, and to use an expiratory filter, which must be changed after each nebulisation. The development of bronchospasm must be closely surveyed [29–35]. The third optimisation strategy is a combination of antimicrobials. A summary of the justification, clinical evidence, useful combinations, and target microorganisms is shown in Table 4 [36–39]. Table 4.Combination therapy for severe pneumoniaJustificationClinical evidenceRecommended associationsTarget microorganisms- Risk of extreme resistance.- Increased coverage probability.- To avoid delay in effective treatment that can lead to a higher bacterial burden.- To overcome resistance mechanisms.- To overcome the limitation of lung PK/PD.Reduction of clinical failure.- In vitro synergia.- Without survival difference compared to monotherapy.- Meropenem plus colistin or fosfomycin.- Colistin plus fosfomycin or rifampicin.- Ceftazidime/avibactam plus aztreonam.Multiresistant Gram-negative bacteria:- K. pneumoniae.- A. baumanii.- Pseudomonas spp.Adapted from Scudeller et al., Farooq et al., Huralska et al., and Tamma et al. [36–39]. Combination therapy for severe pneumonia Adapted from Scudeller et al., Farooq et al., Huralska et al., and Tamma et al. [36–39]. Finally, the fourth optimisation strategy is TDM. It is completely justified as previously mentioned: there is considerable variability in PK in critically ill patients, which entails a high risk of underdosing and overdosing, and measuring serum antimicrobial concentrations can help to achieve PK/PD targets. Moreover, TDM is associated with higher rates of clinical cure, lower rates of treatment failure, higher rates of microbiological eradication, and reduced toxicity. Despite these advantages, there are barriers to its implementation, due to the lack of enough resources (laboratory, software, available population models, professional training), operative definitions (PK/PD targets, toxicity, MIC target) and evidence (benefit in terms of survival, cost-effectiveness studies). Better clinical evidence for TDM can be obtained by improving the protocols used to define PK/PD targets and by selecting patients at risk of inadequate antimicrobial dosing, such as those with ARC or RRT. Earlier access to TDM and better candidate selection are also important. Potential indications are: Critically ill patients. Critically ill patients. Patients with altered kidney function (ARC or RRT). Patients with altered kidney function (ARC or RRT). Microorganisms with a high MIC, such as Pseudomonas spp. Microorganisms with a high MIC, such as Pseudomonas spp. Infections in tissues with bad drug penetration, such as VAP or meningitis. Infections in tissues with bad drug penetration, such as VAP or meningitis. Need for high doses. Need for high doses. Risk of toxicity. Risk of toxicity. With all these elements in mind, the recommended target serum concentration for time-dependent antibiotics such as beta-lactams is to maintain concentrations above the MIC for 100% of the dosing interval, as previously mentioned. A more aggressive approach, with serum concentration between four and eight times the MIC, may optimise the therapeutic effectiveness, prevent the selection of resistant mutants, and provide a safe margin, as CMI is frequently underestimated in critically ill patients. Finally, new mathematical software based on Bayesian Methodology, the integration of pharmacogenomic data, and the integration of point-of-care sensors with information and communication technologies (TIC) and artificial intelligence (AI) represent new opportunities for personalised medicine for these complex patients [7,13,16,18,20,21,40–45]. Update on pneumonia diagnosis Diagnostic imaging for pneumonia in critically ill patients. Classical definitions of HAP and VAP are based on imaging, clinical, and microbiological criteria. In these definitions, it is compulsory that the patient has a new lung infiltrate [46,47]. Apart from Ventilator-Associated Event (VAE) protocol [48], all other guidelines maintain this imaging criterion, along with clinical and microbiological criteria, time from onset of ventilation in the case of VAP and gasometrical criteria [49]. However, in a systematic review of VAP, the authors stated that almost half of the studies included patients without clear diagnostic criteria, and among those with diagnostic criteria, there was a great heterogeneity between them, with 38 different combinations of diagnostic criteria identified [50]. In fact, chest radiography is an established but limited technique in the context of HAP and VAP, even in combination with other diagnostic criteria, with suboptimal sensitivity and specificity [51,52]. Furthermore, the sensitivity and specificity of interpreting portable chest radiography images is comparable when performed by an intensivist versus a radiologist [53]. Due to the limited accuracy of chest radiography in VAP, current guidelines recognise a new pathogenic entity, ventilator-associated tracheobronchitis (VAT), defined as signs of respiratory infection (e.g. fever, leucocytosis, purulent secretions, positive tracheal aspirate culture, etc.) in the absence of radiological lung infiltrates [54,55]. Although the Infectious Diseases Society of America (IDSA) guidelines are ambiguous regarding the necessity of treatment of VAT [55], current evidence advocates for antibiotic treatment, as it improves outcomes [56–61]. Apart from chest radiography, other techniques have been demonstrated useful for the diagnosis of HAP and VAP [2,62–66], as shown in Table 5. The application of artificial intelligence to the interpretation of chest radiographs does not appear to improve diagnostic accuracy compared to a radiologist’s interpretation [67]. Although thoracic computerised tomography (CT) is the gold standard for diagnosing HAP and VAP, intrahospital transport of critically ill patients is associated with side effects, so it should only be performed in selected cases [68,69]. Table 5.Comparison of diagnostic techniques for HAP-VAPAdvantagesDisadvantages Portable chest radiography - Universally available.- Point-of-care technique.- It allows serial follow-up.- Low radiation dose.- Low sensitivity (60 – 70%). Limited specificity.- False images due to the supine position, catheters, wires, etc.- Difficult differential diagnosis of atelectasis, pleural effusion and consolidation.- Poor performance for some lung locations such as the basal and retrocardiac segments. Lung ultrasound - A portable, point-of-care technique.- Repeatable and radiation-free.- High sensitivity (85 – 90%) and excellent NPV.- It detects consolidations, dynamic air bronchograms, focal B-lines and pleural effusions.- Ideal for diagnosing VAP and for unstable patients who are not suitable for transfer.- Operator-dependent.- Moderate specificity (70 – 80%), which is reduced in cases of oedema or atelectasis.- Lower accuracy in VAP than in CAP or HAP.- Training and standardisation of criteria are required. Computerized tomography - Anatomical gold-standard, with a sensitivity of over 95% and high specificity.- It allows differential diagnosis of conditions such as atelectasis, pulmonary thromboembolism, abscess, ARDS, etc.- It can quantify extension and complications: empyema, cavitation, etc.- There is a risk of adverse events during transfer of critically ill patients.- High dose of radiation.- It is more expensive and less available than other techniques.- It is not suitable for frequent follow-up.CAP: community-acquired pneumonia; HAP: hospital-acquired pneumonia; VAP: ventilation-associated pneumonia; ARDS: acute respiratory distress syndrome. NPV: negative predictive value.Adapted from Sistani et al., Orso et al., Haaksma et al., Patrado et al., Strom et al., Rodríguez-Leal et al., Murata et al., and Pagliarini et al. [2,62–66,68,69]. Comparison of diagnostic techniques for HAP-VAP CAP: community-acquired pneumonia; HAP: hospital-acquired pneumonia; VAP: ventilation-associated pneumonia; ARDS: acute respiratory distress syndrome. NPV: negative predictive value. Adapted from Sistani et al., Orso et al., Haaksma et al., Patrado et al., Strom et al., Rodríguez-Leal et al., Murata et al., and Pagliarini et al. [2,62–66,68,69]. Nowadays, updates to current guidelines are underway [70,71]. An operational definition of VAP may help to improve standardisation and surveillance [48]. Finally, as a practical guide, we propose the diagnostic algorithm for VAP shown in Figure 3. Diagnostic algorithm for VAP VAP: ventilation-associated pneumonia; CRP: c-reactive protein; PCT: procalcitonin; VAT: ventilation-acquired tracheobronchitis; VAI: ventilation-acquired respiratory infection; BAS: bronchial aspirate; BAL: bronchoalveolar lavage; PAFI: ratio that compares the partial pressure of oxygen in arterial blood (PaO2) to the fraction of oxygen a patient breathes (FiO2); LUS: lung ultrasound. Developed by the authors. Implementation of biomarkers for pneumonia in critically ill patients. Biomarkers in pneumonia can provide information on diagnosis, prognosis, monitoring of disease progression, and response to antibiotics [2]. For diagnosis, both procalcitonin (PCT) and mid-regional proadrenomedullin (MR-proADM) can give valuable information. MR-proADM showed an earlier increase in patients with burns who developed sepsis compared to PCT, providing a warning of clinical deterioration 24 hours in advance [72]. In the emergency department (ED), PCT is useful to guide the initiation of antibiotics in patients with suspected lower respiratory tract infections, acute exacerbation of asthma or chronic obstructive pulmonary disease, who are likely to be admitted to the hospital [73]. In patients with severe illness in the ICU, PCT can also inform the probability of a bacterial aetiology. However, the prescription of antibiotics is usually more dependent on clinical severity [74]. In critically ill immunocompromised patients, PCT is also helpful. For a cut-off value of 0.5 ng/mL, it has a sensitivity for bacterial infection near to 100%, but with a moderate specificity (63%). A low level of PCT in critically ill and immunocompromised patients can help to rule out bacterial infection or, at least, raise suspicion of non-bacterial processes [75]. A recent systematic review showed that PCT lacks the accuracy required for diagnosing VAP patients [76]. Several biomarkers obtained from samples of the lower respiratory tract in patients with VAP also lack sufficient accuracy for diagnosis [77]. The monocyte volume distribution width (MDW) helps in sepsis identification, but its elevation does not distinguish between bacterial aetiology and other pathogenic agents, such as viruses or fungi [78]. In summary, biomarkers can aid the diagnosis of HAP. While all biomarkers are valuable, they lack sufficient accuracy when used as independent tools, with PCT being the most valuable [79]. In terms of prognosis, MR-proADM is the most effective biomarker. It has been shown to be more effective than lactate or the National Early Warning Score (NEWS) in identifying patients at risk of death [80]. It is also superior to the Sequential Organ Failure Assessment (SOFA) score in patients with sepsis [81]. In another recent study, MR-proADM was compared to other biomarkers (PCT, lactate, C-reactive protein [CRP]) and prognosis scales (Systemic Inflammatory Response Syndrome [SIRS], quick SOFA [qSOFA], and NEWS). MR-proADM performed better than any other comparator. Its ability to identify patients at risk of ICU admission or death was not enhanced by combining it with other biomarkers or prognostic scales, even for immunosuppressed patients [82,83]. Finally, for critically ill patients in the ICU, maintaining a high serum concentration of MR-proADM is associated with poorer outcomes, even in patients whose serum concentrations of PCT are decreasing [81]. Low levels of procalcitonin (PCT) should prompt clinicians to consider alternative diagnoses to infection when guiding antibiotic prescriptions. However, in cases of pneumonia, it is advisable to prescribe antibiotics, even with low serum levels of PCT. Decreasing levels of PCT can help to guide the discontinuation of antibiotics [74,76]. In conclusion, PCT can help rule out bacterial infections and guide the length of antibiotic therapy. MR-proADM is the most reliable biomarker for obtaining prognostic information. At present, no alternative biomarkers have demonstrated sufficiently consistent or reproducible evidence to support their clinical implementation. Nevertheless, emerging observational studies and pilot experiences suggest potential utility, although these findings remain hypothesis-generating rather than practice-changing [84,85]. To make the best possible decisions for our patients, it is necessary to integrate information from clinical data and several biomarkers. Finally, information derived from biomarkers must complement clinical assessment and must not replace it [86–91]. Classic microbiological diagnosis for pneumonia in critically ill patients. The microbiological diagnosis of pneumonia in critically ill patients is based on the appropriate selection of samples, the combination of direct and indirect techniques, and contextualised clinical interpretation. The aetiologies vary according to the clinical scenario: Streptococcus pneumoniae or Haemophilus influenzae in CAP; Pseudomonas aeruginosa, Acinetobacter baumannii, or Staphylococcus aureus in VAP; and a wide spectrum of opportunistic pathogens in immunocompromised patients. The correct choice of sample and technique is essential, as it determines diagnostic sensitivity and the possibility of prescribing targeted antibiotic therapy [92–97]. For conventional bacteriological studies, sample quality is essential. Accepted criteria for sputum evaluation (>25 leukocytes and <10 squamous epithelial cells per field at 100×) remain valid; the leukocyte-to-epithelial cell ratio further improves assessment in mixed samples. Culture remains the reference technique, allowing pathogen identification and susceptibility testing within 24–48 hours, and facilitating targeted therapy or de-escalation, particularly when methicillin-resistant S. aureus (MRSA) or multidrug-resistant Gram-negative bacilli are not isolated. Blood cultures are indicated for severe pneumonias or for patients with risk factors for MRSA or multidrug-resistant Gram-negative bacilli, but their sensitivity is very low. The sensitivity of blood cultures for filamentous fungi is also very low [98,99]. The main classical microbiological techniques used in critically ill patients, together with their principal advantages and limitations, are summarised in Table 6. Table 6.Classical microbiological diagnostic techniques in critically ill patientsTechniqueMain utilityAdvantagesLimitations Gram stain Rapid etiological orientationImmediate results; high specificity and high negative predictive value in VAPLimited sensitivity; low yield in fungal infections Bacterial culture Pathogen identification and antimicrobial susceptibility testingReference method; enables targeted therapy and antimicrobial de-escalationTurnaround time of 24–48 hours; dependent on sample quality Fungal culture Diagnosis of proven invasive fungal infectionHigh specificityLow sensitivity in BAL; slow growth Legionella culture Species identification and epidemiological investigationDetection of non–L. pneumophila serogroup 1 speciesSlow growth; variable sensitivity Urinary antigen tests Rapid diagnosis of pneumonia caused by S. pneumoniae and L. pneumophilaHigh specificity; rapid turnaround timeLimited serogroup coverage; possible false-positive results Galactomannan / β-D-glucan Diagnosis of invasive aspergillosis and invasive fungal infectionHigh diagnostic performance in high-risk patientsDoes not identify fungal species; false-positive resultsBAL: bronchoalveolar lavage; IFI: invasive fungal infection; VAP: ventilator-associated pneumonia. Classical microbiological diagnostic techniques in critically ill patients BAL: bronchoalveolar lavage; IFI: invasive fungal infection; VAP: ventilator-associated pneumonia. In severe CAP and VAP, international guidelines differ on the recommended use of invasive techniques. The IDSA advocates non-invasive sampling methods, such as tracheal aspirates, whereas ERS/ESICM guidelines advise bronchoscopy with bronchoalveolar lavage (BAL) in uncertain cases or among immunocompromised patients [54,55,100–102]. BAL also enables non-culture-based assays, including galactomannan and β-D-glucan, which enhance the detection of filamentous fungi. Nonetheless, overdiagnosis may occur if samples are obtained or cultured without appropriate indication [92,100,103]. Gram staining retains its value in VAP, showing high specificity and clinically actionable information for early antibiotic adjustment [98,101,102,104–108]. Direct microscopy for fungi and mycobacteria provides rapid results but has limited sensitivity, requiring cautious interpretation of negative findings [106,109,110]. Fungal culture remains indispensable for diagnosing proven invasive fungal infection, whereas culture for Legionella preserves its epidemiological relevance despite slow growth and variable sensitivity. Urinary antigen tests, with high specificity, support rapid diagnosis of pneumococcal disease and Legionella pneumophila serogroup 1, although their limitations necessitate complementary techniques when clinical suspicion is high [111–119]. Rapid antigenic tests from respiratory samples for viral diagnosis have a high specificity but a suboptimal sensitivity compared to molecular techniques. Their use is beneficial for the patient, because they allow an adequate use of antivirals and help to avoid overuse of antibiotics [117,120–122]. In aspergillosis, serum and BAL biomarkers—particularly galactomannan and β-D-glucan—play a central role, exhibiting good diagnostic performance in high-risk populations [123–126]. Serology for Legionella spp. can aid diagnosis in selected scenarios but has limited usefulness during the acute phase due to delayed antibody responses [111]. Syndromic molecular panels in pneumonia management. What is new?. The microbiological diagnosis of pneumonia continues to represent a clinical challenge, especially in HAP and VAP, where the need to initiate early antibiotic treatment often precedes aetiological identification. In this context, syndromic molecular panels have emerged as a tool capable of significantly shortening diagnostic times, allowing the simultaneous detection of respiratory pathogens and resistance determinants directly from clinical respiratory samples. One of the main advances introduced by platforms such as FilmArray Pneumonia Plus (FA-PP) is the provision of semi-quantitative results expressed in genomic load ranges. High loads (≥106–107 copies/mL) are generally associated with an aetiological role of the detected microorganism, while lower values may still be clinically relevant in certain pathogens, particularly in patients undergoing active antibiotic treatment. This quantitative approach contributes to a more refined interpretation of the molecular result compared to the classic binary concept of presence or absence [127,128]. Several studies have indicated that incorporating these techniques into the diagnostic algorithm can contribute to improved initial empirical treatment, as well as subsequent optimisation of antimicrobial therapy, provided that their interpretation is carried out in conjunction with clinical and radiological data. In this regard, the experience published by Candel et al. supports the value of syndromic panels as complementary tools within antimicrobial use optimisation programmes (PROA), especially in severe respiratory infections [127,129]. The rapid availability of microbiological results using syndromic molecular panels allows for early and appropriate adjustment of empirical antibiotic treatment, facilitating both therapeutic de-escalation and targeted antimicrobial selection according to the resistance mechanisms detected (Table 7). The clinical experience presented showed improved treatment adequacy and a significant reduction in the total duration of antibiotic therapy following the implementation of FA-PP, especially when these tools are integrated into PROA [127]. Table 7.Early adjustment of empirical antimicrobial therapy based on the results of the syndromic molecular panelSyndromic panel resultClinical interpretationRecommended antimicrobial actionMRSA not detectedLow likelihood of MRSA involvementDiscontinue MRSA coverageMRSA detectedHigh probability of MRSA infectionMaintain or initiate anti-MRSA therapyP. aeruginosa not detected*Low probability of P. aeruginosa involvementDiscontinue anti-Pseudomonas coverageP. aeruginosa detected (non-MBL producer)Likely susceptible P. aeruginosa infectionDe-escalate to ceftolozane–tazobactamP. aeruginosa detected (MBL producer)Multidrug-resistant pathogenAdminister cefiderocol or aztreonam-avibactamESBL-producing Enterobacterales detectedProbable ESBL infectionDe-escalate empirical therapy to ertapenemKPC-producing Enterobacterales detectedCarbapenemase-producing EnterobacteralesAdminister ceftazidime–avibactam, meropenem–vaborbactam or cefiderocolOXA-48–producing Enterobacterales detectedOXA-48 carbapenemase detectedAdminister ceftazidime–avibactamAcinetobacter baumannii detectedHigh-risk multidrug-resistant pathogenAdminister cefiderocol or tigecycline, alone or in combinationNo MDR pathogens detectedLow probability of multidrug-resistant infectionReassess need for broad-spectrum empirical therapy and consider de-escalationMRSA: methicillin-resistant S. aureus; MBL: metallo-β-lactamase; ESBL: extended spectrum beta-lactamase; MDR: multi-drug resistant.*Consider maintaining empirical broad-spectrum therapy if infection by multidrug-resistant microorganisms not included in the syndromic panel is clinically suspected.Adapted from Candel et al. [127]. Early adjustment of empirical antimicrobial therapy based on the results of the syndromic molecular panel MRSA: methicillin-resistant S. aureus; MBL: metallo-β-lactamase; ESBL: extended spectrum beta-lactamase; MDR: multi-drug resistant. Consider maintaining empirical broad-spectrum therapy if infection by multidrug-resistant microorganisms not included in the syndromic panel is clinically suspected. Adapted from Candel et al. [127]. Beyond the initial diagnosis, the potential of syndromic panels in the ongoing monitoring of respiratory infection must also be assessed. An observational study in ventilated patients demonstrated that sequential testing using FA-PP allows for the evaluation of bacterial load dynamics and their relationship with the clinical response to antibiotic treatment [130]. One widely debated aspect was the concept of molecular colonisation, understood as the detection of microbial genetic material without a clear clinical correlation. Recent studies warn that the high sensitivity of molecular techniques can generate results that are difficult to interpret if the clinical context and pre-test probability are not considered [131–133]. Syndromic molecular panels are a highly useful diagnostic tool in the management of pneumonia, especially in critically ill patients. The available evidence indicates that their appropriate use improves diagnostic accuracy, facilitates early adjustment of antibiotic treatment, and helps reduce the overall duration of therapy. However, their maximum benefit is achieved when they are used in a targeted manner, integrated into PROA programmes and interpreted by clinical experts [127,133–137]. Update on community-acquired pneumonia From empirical treatment to the pathogenesis of the most virulent bacteria in community acquired pneumonia. CAP remains one of the leading causes of infectious morbidity and mortality in adults, resulting in a significant healthcare burden and notable clinical heterogeneity. Despite advances in diagnosis, aetiological identification remains limited, which restricts the widespread use of empirical antibiotic treatment. Recent epidemiological studies have shown that, even when contemporary microbiological and molecular techniques are used, the causative pathogen is not identified in a significant proportion of hospitalised patients. There is a relevant contribution of respiratory viruses, and negative results are common [94,138]. This scenario has prompted strategies aimed at optimising diagnostic performance and rationalising the use of antibiotics. International guidelines recommend a selective microbiological approach, guided by clinical severity and the presence of risk factors, avoiding low-performance tests in patients with non-severe CAP. In this context, the usefulness of biomarkers such as procalcitonin is limited, as their sensitivity for bacterial infection is variable and does not allow reliable discrimination between viral and bacterial aetiology at the time of clinical presentation [139]. The current understanding of CAP also incorporates the role of the respiratory microbiome in modulating inflammation and susceptibility to infection. The lung harbours a dynamic microbial community that is influenced by oropharyngeal microaspiration, the immune status of the host, and previous exposure to antibiotics. Respiratory dysbiosis has been linked to more severe clinical presentations and poorer outcomes, while certain commensal species may offer protection against respiratory pathogens [140–142]. From a pathophysiological point of view, specific bacterial virulence mechanisms can explain the progression to severe forms despite adequate antibiotic treatment. In S. pneumoniae, the transition from nasopharyngeal colonisation to pulmonary invasion involves coordinated processes of adhesion, immune evasion, and release of toxic factors [143]. The polysaccharide capsule and pneumolysin play a central role in altering the host immune response, tissue damage, and the development of systemic complications [144–147]. Similarly, infection with Legionella spp. represents a model of severe CAP mediated by intracellular survival of the pathogen and immune dysregulation. The formation of the vacuole containing Legionella spp. and the action of the type IV secretion system allow bacterial replication and condition a biphasic inflammatory response that can progress to severe respiratory failure and multiple organ failure. The main virulence mechanisms of those microorganisms and S. aureus are outlined in Table 8 [148–150]. Table 8.The virulence mechanisms of S. pneumoniae, Legionella spp. and S. aureusPathogenVirulence mechanismPathophysiological consequence Streptococcus pneumoniae Polysaccharide capsuleComplement evasion and reduced opsonisationPneumolysin (PLY)Epithelial injury, inflammation, ARDSInteraction with mannose receptor (MRC-1)Modulation of host inflammatory response Legionella spp. Legionella-containing vacuole (LCV)Intracellular survival and replicationDot/Icm type IV secretion systemTranslocation of effector proteins Staphylococcus aureus Biofilm formation and quorum sensingPersistence, tissue invasionARDS: acute respiratory distress syndrome; LCV: Legionella-containing vacuole.Adapted from Weiser JN et al., Almeida ST et al., Hentrich K et al., Pereira JM et al., Anderson R et al., Iliadi V et al., Pivard M et al., and Rello J et al. [143–150]. The virulence mechanisms of S. pneumoniae, Legionella spp. and S. aureus ARDS: acute respiratory distress syndrome; LCV: Legionella-containing vacuole. Adapted from Weiser JN et al., Almeida ST et al., Hentrich K et al., Pereira JM et al., Anderson R et al., Iliadi V et al., Pivard M et al., and Rello J et al. [143–150]. The recognition that part of the clinical damage in severe CAP does not depend exclusively on the bacterial load has driven the development of adjuvant therapies targeting virulence factors. These include immunomodulatory and antitoxin strategies, such as liposomes capable of sequestering bacterial toxins, as well as new therapeutic approaches combined with conventional antibiotics [151–154]. In parallel, protein vaccines, antimicrobial peptides, and approaches targeting molecular determinants of pneumococcal virulence are being evaluated [155–160]. Those emerging adjunctive and pathogen-targeted therapeutic strategies are summarised in Table 9. Table 9.Emerging prevention and therapeutic strategiesStrategyMechanism of actionLevel of evidenceMacrolides / ketolidesImmunomodulation; inhibition of cytokine releaseClinical and observational studiesCAL02Neutralization of pore-forming toxins via liposomesPhase IIb randomised trialAnti-pneumolysin peptidesDirect toxin neutralisationExperimental modelsProtein-based vaccinesTargeting virulence factorsPhase I–II clinical studiesAntimicrobial peptidesBactericidal activity and antibiotic synergyPreclinical studiesAdapted from Rombauts A et al., Laterre PF et al., Subramanian K et al. 2019, Subramanian K et al. 2020, Sasagawa K et al., Wang Y et al., Ahmad I et al., Cavallazzi R et al., Wunderink RG et al., and Lhommet C et al. [151–160]. Emerging prevention and therapeutic strategies Adapted from Rombauts A et al., Laterre PF et al., Subramanian K et al. 2019, Subramanian K et al. 2020, Sasagawa K et al., Wang Y et al., Ahmad I et al., Cavallazzi R et al., Wunderink RG et al., and Lhommet C et al. [151–160]. Challenges and opportunities in the management of community acquired pneumonia. Contributions from the latest clinical guidelines. CAP continues to be one of the leading causes of infectious morbidity and mortality, both in hospitalised patients and in those with severe symptoms, representing a significant public health problem. Early appropriate antimicrobial treatment is one of the main prognostic determinants, as delay in administering the appropriate antibiotic is associated with a significant increase in mortality, especially in critically ill patients [161–163]. One of the main challenges currently facing the management of CAP is early aetiological diagnosis. Conventional microbiological methods have significant limitations, both in terms of their low sensitivity and the time required to obtain results. In this context, the development of molecular techniques has enabled progress towards the direct diagnosis of pathogens from clinical samples, significantly reducing the time required for microbiological identification [164]. Current recommendations support the use of rapid molecular tests in acute respiratory infections, especially in hospitalised patients or those with severe disease [114,165]. Syndromic molecular platforms have brought a significant change in the diagnostic approach, allowing the simultaneous detection of multiple pathogens and resistance genes in a few hours [166]. However, their high sensitivity can lead to problems of overdiagnosis, detection of colonisation, or genotype-phenotype discrepancies, making it essential to integrate them into structured diagnostic stewardship and antimicrobial stewardship programmes. Several studies have shown that the correct implementation of these strategies can promote early therapeutic optimisation and reduce the unnecessary use of broad-spectrum antibiotics [165–168]. From a clinical-practice point of view, various studies have evaluated the impact of these diagnostic tools in patients with severe infections. Recent studies have shown a high concordance between rapid molecular techniques and conventional microbiological methods, as well as a high rate of therapeutic adjustment based on the early detection of pathogens and resistance mechanisms [97,127,169,170]. Clinical trials and contemporary reviews have confirmed that the use of syndromic panels in CAP can improve the adequacy of antimicrobial treatment and represents a paradigm shift from traditional microbiology to molecular frontiers (Table 10), although its direct impact on mortality remains variable [171–174]. Table 10.The advantages of a precision approach for treating pneumonia, compared to an empirical approachEmpirical approachPrecision approachDelayed etiological diagnosisRapid molecular diagnosisConventional culturesSyndromic molecular panels (multiplex PCR)Broad-spectrum antibioticsTargeted antimicrobial therapyLimited resistance informationEarly detection of resistance genesHigh selective pressureAntimicrobial stewardship strategiesUniform treatmentIndividualised patient-based therapyMinimal host-response evaluationImmunological phenotype assessmentHigher risk of treatment failureImproved clinical outcomes The advantages of a precision approach for treating pneumonia, compared to an empirical approach International clinical guidelines reflect this paradigm shift (Table 11). The ATS/IDSA 2025 guidelines, developed using the GRADE methodology, advocate greater individualisation of CAP management in immunocompetent adults, incorporating the value of early microbiological diagnosis, the possibility of using lung ultrasound in experienced centres, reducing the duration of antibiotic treatment once clinical stability is achieved, and limiting the use of corticosteroids to cases of severe CAP, excluding influenza pneumonia. In parallel, the European ERS 2023 guidelines, focused on severe CAP, introduce conceptual changes with a significant clinical impact. Notably, these guidelines definitively abandon the concept of healthcare-associated pneumonia (HCAP), replacing it with individual stratification for multidrug-resistant pathogens. This approach promotes more rational antimicrobial selection, reinforcing the role of combined beta-lactam and macrolide treatment in severe CAP and discouraging systematic treatment against anaerobes in aspiration pneumonia [139,175,176]. Table 11.Key differences between ATS 2025 and ERS 2023 guidelinesAspectATS / IDSA 2025ERS 2023Target populationImmunocompetent adults with CAPAdults with sCAPMethodologyGRADE and Evidence-to-Decision frameworkPICO questions with GRADEHCAP conceptNot includedDefinitive abandonmentDiagnostic imagingLung ultrasound accepted in experienced settingsEarly imaging strongly recommendedMicrobiological diagnosisEarly etiological assessment when indicatedStrong emphasis on sCAPAntimicrobial strategyIndividualised empirical therapyRisk-based MDR stratificationDuration of antibiotics≤ 5 days after clinical stabilityIndividualized according to severityCorticosteroidsOnly in sCAP; exclude InfluenzaSelected sCAP casesAspiration pneumoniaNo routine anaerobic coverageNo routine anaerobic coveragePreferred combination therapyβ-lactam + macrolideβ-lactam + macrolideATS/IDSA: American Thoracic Society/Infectious Diseases Society of America; ERS: European Respiratory Society; CAP: community acquired pneumonia; sCAP: severe community acquired pneumonia; GRADE: Grading of Recommendations Assessment, Development and Evaluation; PICO: patient, intervention, comparison, outcome; HCAP: healthcare-associated pneumonia; MDR: multidrug-resistance.Adapted from Jones B et al., and Martin-Loeches I et al. [139,175]. Key differences between ATS 2025 and ERS 2023 guidelines ATS/IDSA: American Thoracic Society/Infectious Diseases Society of America; ERS: European Respiratory Society; CAP: community acquired pneumonia; sCAP: severe community acquired pneumonia; GRADE: Grading of Recommendations Assessment, Development and Evaluation; PICO: patient, intervention, comparison, outcome; HCAP: healthcare-associated pneumonia; MDR: multidrug-resistance. Adapted from Jones B et al., and Martin-Loeches I et al. [139,175]. The management of immunosuppressed patients represents one of the most complex scenarios. In this group, there is higher mortality, a more heterogeneous aetiology, and a higher frequency of opportunistic pathogens, such as Pneumocystis jirovecii or Aspergillus spp., coexisting with common bacterial microorganisms [177–180]. Recent multicentre studies have shown that immunosuppression is an independent factor for poor prognosis, especially in patients admitted to intensive care units [181]. Finally, advances in understanding the immune response in CAP have led to the identification of different inflammatory phenotypes, ranging from states of hyperinflammation to situations of immunoparalysis. This knowledge has driven the development of new immunomodulation strategies and a growing interest in integrating artificial intelligence to combine clinical, microbiological and immunological data. The aim is to move towards a precision medicine approach to pneumonia [176,182]. Should adult inpatients with community acquired pneumonia be treated with corticosteroids?. There is contradictory evidence regarding the usefulness of corticosteroids in CAP. Recommendations from the main guidelines, the results of randomised clinical trials (RCTs), and real-world evidence (RWE) studies disagree on their utility [139,183,184]. These contradictory recommendations are derived from several studies with different conclusions. A recent RCT demonstrated the benefits of hydrocortisone for severe CAP [185], whereas other recent RCTs have found no difference compared to a placebo, or even a lower survival rate [186,187]. Even RCTs conducted by the same research group have reached different conclusions about the effect of hydrocortisone in severe CAP [187,188]. This disagreement can be partly explained by differences in study design, as shown in Table 12 [189]. As a result, interpretation of systematic reviews and meta-analyses is difficult, as they amplify these contradictions [139,190–194]. The reader must know that meta-analyses have inherent limitations, derived from characteristics of the selected studies and from the employed methodology, as it is shown in Table 13 [195]. Table 12.Problems and potential solutions for the design of studies about corticosteroid effect in severe community-acquired pneumoniaProblems in study designAetiology misclassificationCorticosteroid regimen variabilityPatient heterogeneityDiagnosis uncertainty- Lack of rigorous identification.- Potential misclassification.- Different molecules.- Dose variations.- Difference in baseline severity.- Variability in age and sex.- No systematic microbiological testing.- Bacterial vs viral infections. Possible solutions - Stratify the patients most likely to benefit from treatment.- Determine the optimal therapeutic window.- Choose the right molecule, dose and treatment duration.Adapted from Helms et al. [189]. Problems and potential solutions for the design of studies about corticosteroid effect in severe community-acquired pneumonia Adapted from Helms et al. [189]. Table 13.Limitations of meta-analyses about corticosteroid treatment in severe community-acquired pneumoniaCharacteristics of the selected studiesMeta-analysis methodology- Differences in population properties.- Selection bias.- Differences in chosen corticosteroid: type, dose, timing.- Increase in antibiotic resistance.- Differences in organ support.- Using I2 for heterogeneity quantification can be inadequate.- Increasing the number of selected studies increases the number of individuals to achieve statistical significance.- Including studies with a reduced sample size.Adapted from Luo et al. [195]. Limitations of meta-analyses about corticosteroid treatment in severe community-acquired pneumonia Adapted from Luo et al. [195]. Patients’ phenotypes are important for tailoring corticosteroid treatment. Patients with septic shock and severe CAP clearly benefit from this treatment [196]. This favourable effect may be mediated by the presence of shock rather than by severe CAP itself. These differences between patients with CAP can be clinical and immunological [197,198]. Recognising these phenotypes can help individualise treatment, since they respond differently to the administration of corticosteroids. In a recent study that classified patients according to metagenomic characteristics, patients who tested positive for Streptococcus spp. and had higher levels of inflammatory markers had a poorer response to corticosteroids [199]. These phenotypes are also important in viral pneumonia, but they must be adapted to the characteristics of the host and the aetiology [200,201]. Finally, another recent meta-analysis suggests that patients with a serum concentration CRP higher than 150-200 mg/L, respiratory failure, and inflammatory phenotypes may benefit from corticosteroid treatment. However, controversy still exists regarding the use of corticosteroids in patients with Influenza co-infection. The most appropriate timing and type of corticosteroid also remain to be determined. Therefore, an individualised approach is essential when prescribing corticosteroids for severe CAP, and many questions remain unanswered. Reasons for combined treatment in cases of severe community-acquired pneumonia. The long-term outlook for severe CAP is poor, with a mortality rate of almost 50% after one year, a cardiovascular event rate of around 13.9%, and an increased risk of cognitive impairment [202]. To improve these outcomes, combined antibiotic therapy is recommended [203]. The aetiology of hospitalised CAP is diverse, and more than 65% of cases remain undiagnosed. The rate of microbiological diagnosis is better in severe CAP that requires ICU admission, and it is around 77%. The most frequent isolated microorganism is S. pneumoniae in both environments. However, given the diversity of bacteria that can cause CAP, an antibiotic combination increases the odds of adequate coverage. For patients with risk of antibiotic resistance, this combination must be adapted according to their characteristics, the ecology, and the results of rapid diagnostic devices [204]. Evidence of the benefits of combined antibiotic therapy for CAP began to emerge at the end of the last century, first with retrospective studies and then with observational prospective studies. The benefit is more evident for severely ill patients [205–209]. In terms of choice of the antibiotic combination, the mortality rate is lower for ventilated ICU patients receiving a beta-lactam plus macrolide combination than in those receiving a beta-lactam and quinolone combination [210]. These better outcomes do not seem related to susceptibility to macrolides of the isolated microorganisms or to the broader spectrum [211,212]. In addition to their antibacterial activity, macrolides have immunomodulatory properties. They promote the differentiation of macrophages into the M2 phenotype and increase their phagocytic and efferocytic activity. They also affect neutrophil activity, decreasing oxidative stress and modulate T-cell activity and secretion of proinflammatory cytokines, such as IL-10, IL-8 and IL-6 [213,214] All of these favourable properties are also associated with better long-term outcomes for ICU-admitted patients with CAP, with lower mortality rates at 6 and 12 months for patients who received a macrolide-based antibiotic combination [215]. However, for milder patients, these benefits are not so clear. Several recent retrospective observational studies failed to demonstrate differences between antibiotic regimens for nonsevere CAP [216,217]. Despite this, a meta-analysis published in 2023 found that regimens with macrolides were associated with better outcomes in terms of CAP resolution and mortality at 30 days [218]. Evidence derived from RCTs for nonsevere CAP failed to demonstrate better mortality outcomes for macrolide combinations. However, a lower 30-day readmission rate was observed for this combination [219–221]. The ACCESS RCT provided valuable insights into this treatment strategy for patients with CAP and systemic inflammatory response syndrome (SIRS), with a Sequential Organ Failure Assessment (SOFA) score greater than one and serum PCT levels of at least 0.25 ng/mL. The macrolide used was clarithromycin. Patients receiving combined treatment were more likely to improve, as measured by a composite outcome of clinical response and decreased systemic inflammation. Moreover, patients on the combination regimen were at lower risk of developing sepsis at 30 days. No severe adverse effects were reported [222]. Based on this evidence, the current guidelines recommend prescribing a combination of a beta-lactam and a macrolide rather than a fluoroquinolone for severe CAP [175]. Update on hospital-acquired pneumonia Empirical therapeutic strategy for ventilator-associated pneumonia in multidrug-resistant settings. VAP is one of the main healthcare-associated infections in ICUs, associated with a significant increase in morbidity and mortality, length of hospital stay, and healthcare resource utilisation. Despite advances in prevention, diagnosis and treatment, VAP continues to present a clinical challenge, particularly in settings with a high prevalence of multidrug-resistant (MDR) bacteria. Several meta-analyses have shown that VAP is associated with significant attributable mortality and a notable economic impact on healthcare systems [223,224]. Observational studies have also identified an increase in long-term mortality in patients who develop VAP during their ICU stay [225]. Recent data from the European ENIRRIs study confirm the high clinical severity of these infections, with rates of septic shock between 26 and 40% and failure of initial antibiotic treatment in up to 46% of cases [226]. This scenario is accentuated by the progressive increase in antimicrobial resistance described in national and European surveillance programs [93,227,228], as well as by the circulation of carbapenem resistance mechanisms (e.g. carbapenem-producing Enterobacterales -CPE-) with geographical variability [229]. The appropriateness of initial empirical antibiotic treatment is one of the main prognostic determinants in VAP, as well as in other serious infections. Multicentre studies have shown that early administration of appropriate antibiotic treatment is associated with a significant reduction in hospital mortality [230]. Moreover, multidrug resistance and inappropriate initial treatment are associated with a worse prognosis in severe sepsis and septic shock caused by Gram-negative bacilli [231]. Conversely, delays in initiating appropriate antibiotic therapy independently increase the risk of clinical failure and death [232], with special relevance in infections caused by CPE [233]. The burden of disease remains high in contemporary series of nosocomial bacteraemic pneumonia caused by Gram-negative bacilli [234]. Previous colonisation by multidrug-resistant microorganisms is a key risk factor for subsequent infection. Population studies and studies in critically ill patients show that colonization by resistant pathogens is associated with an increased risk of subsequent infection and poorer clinical outcomes, including higher 90-day mortality rates in certain groups [235–237]. International guidelines recommend that the choice of empirical treatment for VAP be based on an individualised assessment that includes clinical severity, previous antibiotic exposure, length of hospital stay, known colonisation, and local ecology, as can be seen in Table 14 [54]. Risk factors and patient profiles for multidrug-resistant Gram-negative infection have been described in reviews and observational studies [238–242] and support the need to adjust empirical coverage to individual risk, as it is shown in Table 15. Table 14.Stepwise algorithm for empirical antibiotic therapy in ventilator-associated pneumonia in multidrug-resistant settingsStepClinical assessmentTherapeutic decision1. Clinical suspicion of VAPFever, purulent respiratory secretions, worsening oxygenation, new or progressive pulmonary infiltratesInitiate diagnostic work-up and evaluate severity2. Severity assessmentPresence of septic shock, multiorgan failure, or need for vasopressor supportImmediate initiation of empirical antibiotic therapy3. Assessment of MDR riskPrevious colonization with MDR pathogens; antibiotic exposure within the last 90 days; ICU stay > 5–7 days; high local antimicrobial pressureStratify patients according to low or high risk of MDR pathogens4. Empirical antibiotic selectionLow MDR risk and clinical stabilityStandard empirical therapy according to local guidelinesHigh MDR risk and/or severe clinical presentationBroad-spectrum empirical therapy; consider novel β-lactam/β-lactamase inhibitor combinations5. PK/PD optimizationCritically ill patients, altered volume of distribution, augmented renal clearanceUse extended or continuous infusion strategies and optimized dosing6. Early microbiological diagnosisRespiratory quantitative cultures; blood cultures; syndromic PCR panels when availableAdjust therapy as soon as microbiological data become available7. Reassessment at 48–72 hoursClinical evolution and microbiological resultsDe-escalate therapy, optimize dosing, or discontinue antibiotics if infection is not confirmed8. Ongoing follow-upClinical response and microbiological evolutionEnsure antimicrobial stewardship and prevent emergence of resistanceICU: intensive care unit; MDR: multidrug resistance; PK/PD: pharmacokinetics-pharmacodynamics; PCR: polymerase chain reaction.Adapted from Torres A, et al. [54]. Stepwise algorithm for empirical antibiotic therapy in ventilator-associated pneumonia in multidrug-resistant settings ICU: intensive care unit; MDR: multidrug resistance; PK/PD: pharmacokinetics-pharmacodynamics; PCR: polymerase chain reaction. Adapted from Torres A, et al. [54]. Table 15.Key determinants for the selection of empirical antibiotic therapy in ventilator-associated pneumoniaClinical or epidemiological factorTherapeutic implicationSeptic shock or multiorgan failureImmediate initiation of broad-spectrum empirical antibiotic therapyPrevious colonization with MDR pathogensEmpirical coverage tailored to documented colonizationHigh prior antibiotic exposureIncreased risk of MDR pathogens; consider novel β-lactam/β-lactamase inhibitorsProlonged ICU stayHigher probability of late-onset VAP due to MDR organismsLocal ecology and resistance patternsSelection of empirical therapy adapted to local epidemiologyRapid microbiological diagnosisEarly optimization and antimicrobial de-escalationPK/PD optimizationImproved clinical cure and reduced emergence of resistanceMDR: multidrug resistance; ICU: intensive care unit; VAP: ventilator-associated pneumonia; PK/PD: pharmacokinetics-pharmacodynamics.Adapted from Martin-Loeches I et al., Falcone M. et al., Blagojevic C et al., McConville TH et al., Sim JK et al., Torres A et al., Timsit JF et al., EPINE study, EARS-Net, ENVIN-HELICS, Gatti et al., Papazian L et al., Berteau F et al. [54,93,226–228,233,235–237,242,243,245,246]. Key determinants for the selection of empirical antibiotic therapy in ventilator-associated pneumonia MDR: multidrug resistance; ICU: intensive care unit; VAP: ventilator-associated pneumonia; PK/PD: pharmacokinetics-pharmacodynamics. Adapted from Martin-Loeches I et al., Falcone M. et al., Blagojevic C et al., McConville TH et al., Sim JK et al., Torres A et al., Timsit JF et al., EPINE study, EARS-Net, ENVIN-HELICS, Gatti et al., Papazian L et al., Berteau F et al. [54,93,226–228,233,235–237,242,243,245,246]. The development of new beta-lactam antibiotics combined with novel beta-lactamase inhibitors has expanded the therapeutic options for multidrug-resistant Gram-negative pathogens and may be particularly relevant in patients at high risk of specific resistance mechanisms [229,240,243]. PK/PD optimization through extended or continuous dosing strategies and the achievement of more demanding PK/PD targets have been associated with improved clinical outcomes and reduced selective pressure for resistance [20]. In parallel, early diagnosis and treatment strategies for multidrug-resistant Gram-negative infections reinforce the approach of early appropriate therapy followed by de-escalation [244]. Likewise, the incorporation of rapid diagnostic techniques, such as syndromic PCR panels, makes it possible to shorten the time to targeted treatment, improve antibiotic appropriateness, and promote therapeutic de-escalation [245,246]. In settings with a high incidence of CPE and P. aeruginosa infections, contemporary epidemiological surveillance provides essential information for decision-making [247]. Finally, recent updates focusing on the therapeutic management of healthcare-associated pneumonia provide a framework applicable to clinical practice [248]. Refractoriness and therapeutic failure in hospital-acquired pneumonia: management strategies. The concept of refractory pneumonia dates back to 1943 [249]. Nowadays, it is defined as a lack of significant clinical improvement or worsening of a pulmonary infection after at least 48–72 hours of adequate antibiotic treatment. Regular re-evaluation of the clinical situation, laboratory results, microbiological results, and imaging studies is key to diagnosis. The diagnosis must be suspected in patients with persistent clinical signs, radiological progression, worsening of organ failure, elevated infection-biomarkers, isolation of a new microorganism on the third day, and the most important, no improvement in the relationship between partial arterial oxygen pressure and inhaled oxygen fraction (PAFI). Its incidence among patients with severe pneumonia is between 20-60%, with a mortality rate over 40%. Despite its severity, the literature lacks detailed epidemiological data and clear management recommendations [97,250–261]. The causes of refractory pneumonia are listed in Table 16. The most common causes are suboptimal focus control and a subinhibitory antibiotic concentration [97,250,254,262–264]. Predictive factors for refractory pneumonia are summarized in Figure 4. Although resistant microorganisms are the main predictor of treatment failure, they are not always easy to identify. A large quantity of bacteria is associated with heteroresistance. Sometimes, only a small number of mutant bacteria carrying genes related to antibiotic resistance are present. Once the antibiotic treatment has destroyed the susceptible bacteria, those with resistance mechanisms can proliferate. In other cases, bacteria tolerance may emerge. This mechanism involves bacteria that can survive without proliferating in the presence of antibiotics. Finally, some species of bacteria, such as P. aeruginosa and Enterobacter spp., can reactivate chromosomal genes that encode beta-lactamases [97,250,254,255,257,263,265–271]. Table 16.Causes of refractory pneumoniaNo infectiousInfectiousRelated with the infection- Atelectasis- Lung oedema- Pulmonary embolism- Drug-induced pneumonitis- ARDS, diffuse alveolar damage- Pulmonary contusion- Pulmonary haemorrhage- Bronchiolitis obliterans- Acute eosinophilic pneumonia- Heart failure- Cancer- Drug fever- Outside the lungs: infectious process in another location Eg.: mechanical ventilation-associated sinusitis, vascular catheter-related infection, etc.- In the lungs, uncommon pathogens:- Fungi- Viruses- Mycobacteria- Complications- Parapneumonic pleural effusion- Empyema- Lung abscess- Necrotizing pneumonia- Inadequate treatment:- Dose- Lack of drug penetration into the tissue- Resistant microorganismsARDS: acute respiratory distress syndrome.Adapted from Povoa et al., Rodríguez-Leal et al., Candel et al., Modi et al., Schuierer et al., and Grief et al. [97,250,254,262–264]. Causes of refractory pneumonia ARDS: acute respiratory distress syndrome. Adapted from Povoa et al., Rodríguez-Leal et al., Candel et al., Modi et al., Schuierer et al., and Grief et al. [97,250,254,262–264]. Predictive factors of refractory pneumonia HF: heart failure. DM: diabetes mellitus; COPD: chronic obstructive pulmonary disease; CKF: chronic kidney failure; APACHE: acute physiology and chronic health evaluation; SOFA: sequential organ failure assessment; ICU: intensive care unit; MV: mechanical ventilation; MRSA: methicillin-resistant S. aureus; NFGNB: non-fermenting Gram-negative bacteria; PK/PD: pharmacokinetic and pharmacodynamic; ARC: augmented renal clearance; PAFI: relationship between partial arterial oxygen pressure and inhaled oxygen fraction; PCT: procalcitonin; MR-proADM: mid-regional proadrenomedullin; CPIS: Clinical Pulmonary Infection Score. Adapted from Povoa et al., Wunderick et al., Kollef et al., Bouglé et al., Gursel et al., Albin et al., Modi et al, Cilloniz et al., Candel et al., Zaragoza et al, Grief et al, Huu et al., and Ranzani et al. [97,250,254,255,257,263,265–271]. Finally, in Figure 5, an algorithm for the management of refractory pneumonia is proposed. Invasive techniques, such as bronchoalveolar lavage and, in selected cases, lung biopsy, are key to achieving an accurate diagnosis [272]. Radiological diagnostic techniques must be used appropriately. Plain thorax radiography cannot detect up to 30% of cases of pneumonia, whereas lung echography is almost 90% sensitive and highly specific. A dynamic air bronchogram is considered pathognomonic. Computer tomography is the gold standard for diagnosing pneumonia [273]. Management of refractory pneumonia BAL: bronchoalveolar lavage; NAAT: nucleic acid amplification test; CT: computerised tomography; PK/PD: pharmacokinetics and pharmacodynamics; ARC: augmented renal clearance. A tailored antibiotic treatment is crucial to achieve favourable outcomes, and PK/PD parameters must be taken into account. ARC is a frequent problem in critically ill patients, specifically among patients with brain injuries, so a correct dosing regimen must be prescribed, with dose escalation when necessary [270,274]. Prognostic factors in postoperative pneumonia. Postoperative pneumonia is one of the most significant infectious complications in surgical patients, with a significant impact on morbidity and mortality, quality of life, and healthcare resource consumption. Surgery is an essential component of healthcare systems worldwide, with more than 300 million procedures performed annually, making postoperative complications a priority public health issue [275]. Population studies have shown that surgical mortality remains high even in countries with developed healthcare systems, and that a substantial proportion of these deaths are related to preventable complications [276]. Pneumonia consistently ranks among the leading postoperative complications, independently increasing the risk of both early and late mortality. Data from European cohorts have shown that patients who develop pneumonia after surgery are more likely to die and to have prolonged hospital stays [276]. Furthermore, the clinical impact of these complications extends beyond the immediate period, with a significant deterioration in quality of life up to 12 months after surgery [277]. The onset of postoperative pneumonia is the result of a complex interaction between host-dependent factors, the surgical procedure, and perioperative management, as can be seen in Table 17. Patient-related factors include advanced age, the presence of chronic comorbidities—especially chronic obstructive pulmonary disease, diabetes mellitus, renal failure, and heart disease—and multimorbidity, which has been established as one of the main prognostic determinants after surgery. Recent studies have shown that the coexistence of chronic diseases significantly influences clinical outcomes and the development of infectious complications [278]. Table 17.Risk factors associated with postoperative pneumoniaCategoryRelevant factorsPatient-relatedAdvanced age, COPD, diabetes mellitus, chronic kidney disease, cardiovascular disease, multimorbidityFunctional statusFrailty, sarcopenia, baseline dependency, malnutritionSurgical factorsMajor surgery, emergency surgery, prolonged operative timePerioperative factorsProlonged mechanical ventilation, blood transfusion, ICU admissionCOPD: chronic obstructive pulmonary disease; ICU: intensive care unit.Adapted from Fowler et al. [278]. Risk factors associated with postoperative pneumonia COPD: chronic obstructive pulmonary disease; ICU: intensive care unit. Adapted from Fowler et al. [278]. The type of surgery also plays a decisive role. The incidence of pneumonia is particularly high after cardiothoracic procedures, major abdominal surgery, and orthopaedic surgery in elderly patients, particularly in the context of hip fractures, where a significant increase in mortality associated with postoperative infection has been reported. In head and neck surgery, the duration of intubation, the need for tracheostomy, and admission to intensive care units are associated with an increased risk of pneumonia [279,280]. From a pathophysiological perspective, postoperative pneumonia is favoured by impaired respiratory defence mechanisms, hypoventilation due to pain, aspiration of secretions, and the colonisation of the lower respiratory tract. Mechanical ventilation is a key factor, and its prolongation has been associated with worse clinical outcomes, even in the absence of risk factors for VAP [281]. In this context, HAP not associated with mechanical ventilation has emerged as a relevant entity, with partially modifiable risk factors [282], as shown in Table 17. In recent years, frailty and sarcopenia have taken on a central role in surgical risk stratification. Recent studies and meta-analyses have shown that sarcopenia is consistently associated with an increased risk of postoperative pneumonia, especially in elderly patients, and is also associated with poorer functional recovery and higher mortality [283,284]. Preventive strategies have become a priority in addressing this complication. Structured prevention programmes have been shown to consistently reduce the incidence of pneumonia in surgical wards [285]. In addition, multimodal prehabilitation, which integrates physical exercise, nutritional optimisation, and control of comorbidities, has been shown to reduce postoperative complications and improve functional capacity, especially in colorectal cancer surgery [286,287]. These approaches are currently integrated into the principles of Enhanced Recovery After Surgery (ERAS), aimed at optimising clinical outcomes [288] (Table 18). Table 18.The main preventive and perioperative strategies currently recommended to reduce postoperative pneumoniaInterventionClinical objectiveMultimodal prehabilitationImprove functional capacity and reduce postoperative complicationsNutritional optimizationReduce frailty and sarcopeniaEarly mobilizationDecrease respiratory complicationsOptimal pain controlFacilitate effective ventilation and airway clearanceStructured prevention programsReduce the incidence of postoperative pneumoniaAdapted from Wright RM et al. [288]. The main preventive and perioperative strategies currently recommended to reduce postoperative pneumonia Adapted from Wright RM et al. [288]. Update on viral pneumonia in critically ill patients: beyond influenza and COVID-19 Global impact of respiratory syncytial virus in critically ill patients. Management strategies. Respiratory syncytial virus (RSV) is a recognized major source of morbidity, ICU admission, and death among paediatric population. However, its importance among the elderly population is under-recognised. Recent studies show that its impact on hospital and ICU admissions, disability, and death is significant among older adults [289,290]. In Spain, the RSV epidemic usually begins at the end of autumn, preceding the influenza epidemic [291]. About 6.4% of hospitalised patients with RSV are admitted to the ICU. Advanced age is a risk factor for mortality, as inpatients aged over 80 have a death rate of around 9.5% [292]. The rate of ICU admission is approximately one in four for hospitalised patients in other countries and it is higher for RSV than for Influenza [293]. The severity of RSV infection in hospitalised patients is similar to that in unvaccinated patients with SARS-CoV-2 or influenza [294,295]. In fact, these three viruses share common risk factors for severe illness, such as comorbidities, immunosuppression, and advanced age [290,296]. It is more important to consider related frailty and comorbidities than advanced age, as chronic illnesses are prone to decompensate with RSV infection. Patients with chronic heart failure, in particular, are more susceptible to complications arising from an RSV infection [297]. The development of RSV infection depends on the patient’s underlying characteristics. The risk factors for mortality are age over 65 years, immunosuppression, ICU admission, and respiratory failure. Three clusters of patients can be identified [290,298]: Cluster 1. Patients with severe immunosuppression, such as haematological patients. Their mortality rate is around 20%, which is higher than that of the average hospitalised patient with RSV (10%). These patients require ICU admission more frequently and often have pneumonia or acute respiratory failure. Bacterial coinfection is frequent. Their prognosis is worse compared to patients with influenza. Cluster 1. Patients with severe immunosuppression, such as haematological patients. Their mortality rate is around 20%, which is higher than that of the average hospitalised patient with RSV (10%). These patients require ICU admission more frequently and often have pneumonia or acute respiratory failure. Bacterial coinfection is frequent. Their prognosis is worse compared to patients with influenza. Cluster 2. Patients with advanced age and comorbidities, such as chronic heart failure. These patients often have chronic respiratory disease, hypoxia, dyspnoea, or congestive heart failure. They have a low incidence of pneumonia, and the prognosis is comparable to unvaccinated patients with Influenza. Bacterial coinfection rate is comparable to Influenza. Laboratory-confirmed viral-bacterial coinfection is associated with a higher mortality rate [299,300]. Although opportunistic fungal infections are not well documented in the literature, their incidence can reach 32% among severely ill patients [301]. Cluster 2. Patients with advanced age and comorbidities, such as chronic heart failure. These patients often have chronic respiratory disease, hypoxia, dyspnoea, or congestive heart failure. They have a low incidence of pneumonia, and the prognosis is comparable to unvaccinated patients with Influenza. Bacterial coinfection rate is comparable to Influenza. Laboratory-confirmed viral-bacterial coinfection is associated with a higher mortality rate [299,300]. Although opportunistic fungal infections are not well documented in the literature, their incidence can reach 32% among severely ill patients [301]. Cluster 3. Patients with mild immunosuppression, such as those who have undergone solid organ transplantation. Cluster 3. Patients with mild immunosuppression, such as those who have undergone solid organ transplantation. Moreover, RSV infection is a trigger for acute cardiac events. It is associated with a cardiac event in about 22% of hospitalised older patients. The risk is higher in patients with known cardiovascular disease (33%), but patients without documented underlying cardiovascular disease are also at risk (9%). RSV infection worsens prognosis in these patients. Congestive heart failure is more frequent than with other respiratory viral infections (influenza, SARS-CoV-2), whereas similar rates are observed for ischemic events [302]. Among the immunosuppressed population, patients who have undergone lung or haematopoietic transplant have a higher incidence of infection, and higher incidence of bacterial or fungal coinfection [303]. In a recent publication, RSV infection in patients with haematological malignancies was associated with a higher mortality rate than influenza and COVID-19 [304]. For risk stratification in these patients, the Immunodeficiency Scoring Index (ISI) can be used, as patients with high ISI scores have worse prognosis than patients with intermediate or low ISI score (Table 19) [305,306]. Viral detection in critically ill haematology patients is also associated with worse outcomes. Patients with RSV infection have higher rates of acute respiratory failure and death, and around 17% are initially asymptomatic for respiratory viral infection [307]. Table 19.Immunodeficiency Scoring Index (ISI)FeatureCut-off valuePointsAge> 40 years2 Absolute neutrophil count<500/mcL3 Absolute lymphocyte count<200/mcL3 GVHDAcute2Chronic1 Myeloablative conditioning regimenYes1 Corticosteroids>1 mg/kg/day of prednisone2 Onset of RSV infection since transplant<31 days231–100 days1>100 days0 Interpretation 0–2 pointsLow risk 3–6 pointsIntermediate riskGVHD: graft-versus-host disease; RSV: respiratory syncytial virus.Adapted from Shah et al. and Micó-Cerdà et al. [305,306]. Immunodeficiency Scoring Index (ISI) GVHD: graft-versus-host disease; RSV: respiratory syncytial virus. Adapted from Shah et al. and Micó-Cerdà et al. [305,306]. As we have seen, the clinical burden of RSV infection is high, so controlling the infection is key to preventing related complications. Non-pharmacological measures are the first line of defence against transmission. Basic measures such as hand hygiene, staying at home if unwell, wearing face masks, and ensuring proper ventilation of shared indoor spaces are key to prevent respiratory viral transmission [308]. In recent years, monoclonal antibodies, such as nirsevimab, and vaccines have arisen as effective pharmacological measures to prevent adverse outcomes in vulnerable populations [290]. Nirsevimab has changed RSV epidemiology among infants, and vaccines have demonstrated their utility in older patients [309–311]. In Spain, the Ministry of Health has published updated recommendations for RSV vaccination. These include immunosuppressed patients aged over 18, as well as patients aged over 60 who live in long-term care facilities [312]. For people with severe immunodeficiency, monoclonal antibodies may be an alternative, but more research is needed [290]. There are current limited options for RSV treatment. Moreover, damage derived from the host response, as well as virus-derived damage, must be considered when managing this illness [289]. Ribavirin has been used for selected patients. Oral and aerosolised ribavirin are similarly effective, and the oral route is cheaper and associated with fewer adverse effects. It significantly improves survival in patients with haematological diseases, but there is insufficient evidence to support its use in lung transplant recipients and healthy infants. No severe adverse events were reported in a recent meta-analysis. Currently, ribavirin treatment is recommended for high risk haematological patients with a risk of progression [313–315]. Other antivirals are under investigation, but they are not yet clinically available [316–319]. Other therapies, such as non-specific immunoglobulins in patients with hypogammaglobulinemia and decreasing corticosteroids dose, if used, are recommended [315]. Palivizumab, a monoclonal antibody targeting the F protein, could be useful for immunocompetent patients. However, further research is required [320]. An approach to the reactivation of herpesviruses and cytomegalovirus in critically ill immunocompetent patients. Herpes simplex virus type 1 (HSV-1) and cytomegalovirus (CMV) are both members of the Herpesviridae family. HSV-1 can cause an asymptomatic primary infection, or one with malaise and the characteristic mucocutaneous lesions. HSV-1 can cause herpetic encephalitis and other infections in immunosuppressed patients. CMV can also cause an asymptomatic primary infection or a non-specific viral syndrome, and may lead to severe illness in newborns and in patients with weakened immune systems. Quantitative polymerase chain reaction is useful for diagnosis and monitoring. Following primoinfection, both viruses survive in a latent state in different cells and can reactivate in response to various stimuli or immunosuppression [321,322]. HSV-1 reactivation in critically ill patients is associated with poorer outcomes. Patients with viral reactivation require invasive mechanical ventilation for longer, have a longer ICU stay and are at greater risk of VAP. Reactivation is also associated with corticosteroid treatment [323]. These associations are also evident in patients with pneumonia caused by SARS-CoV-2, and mortality is higher in patients with a positive histology test result [324–326]. However, these studies lacked clinical signs of infection, and they relied only on the molecular detection of HSV-1. In a clinical trial of acyclovir for HSV-1 reactivation in patients with IMV, these patients also lacked clinical signs of HSV-1 infection. While the administration of acyclovir (5 mg/kg/8 hours) was well tolerated, it did not demonstrate efficacy in reducing the length of IMV or the death rate [327]. Nevertheless, an observational study demonstrated a lower mortality rate for patients with molecular detection of HSV-1 who received a higher dose of acyclovir (10 mg/kg/8 hours) [328]. Treatment with acyclovir is also linked to better outcomes in cases of extrapulmonary illness, such as hepatitis, due to HSV-1 reactivation [329]. CMV reactivation in critically ill patients is also associated with worse outcomes [330]. However, a pre-emptive strategy with ganciclovir treatment of seropositive critically ill patients did not demonstrate an improvement of those outcomes or IL-6 production, but patients with ganciclovir prophylaxis had a lower rate of CMV reactivation [331]. Another randomised clinical trial with a similar pre-emptive strategy (prophylaxis with valacyclovir or valganciclovir) demonstrated a benefit in terms of lower CMV reactivation. However, the trial had to be stopped early due to a higher mortality rate in the treatment group [332]. Molecular detection of CMV is associated with higher mortality in critically ill patients, but in patients with severe COVID-19, although is associated with thrombocytopenia and bacterial coinfection, is not associated with a higher mortality rate [333,334]. Patients who are seropositive or have molecular CMV detection have an impaired immune response. The normal function of dendritic cells and leukocytes is affected, and interleukin production shifts towards a less pro-inflammatory state [335,336]. Moreover, the expression of the ACE-2 receptor is increased in infected respiratory tract epithelial cells, facilitating SARS-CoV-2 superinfection [337]. Furthermore, seropositive healthy individuals exhibit an impaired response to SARS-CoV-2 vaccination [338]. Critical illness can reactivate latent CMV infection, leading to immunosuppression and worse outcomes for patients. However, there is no clear management strategy [334]. Nevertheless, patients with clinical signs of CMV pneumonitis may benefit from ganciclovir treatment [339]. Management of pneumonia in the intensive care unit due to other respiratory viral infections (rhinovirus, adenovirus, etc.). Impact on antimicrobial therapy. Respiratory viruses are increasingly recognised as major pathogens in adult community-acquired pneumonia, with a clinical impact that extends beyond seasonal influenza [340]. The widespread implementation of multiplex molecular diagnostic techniques has revealed a substantial burden of non-influenza respiratory viruses among hospitalised adults, particularly in elderly patients and those with underlying comorbidities [341,342]. These findings have reshaped the understanding of viral pneumonia in adults, highlighting its relevance as a cause of severe respiratory illness. Large observational cohorts and prospective studies have demonstrated that pneumonia caused by respiratory viruses is associated with significant morbidity, frequent hospitalisation, and rates of intensive care unit admission and mortality comparable to those observed in influenza-associated pneumonia [341–345]. In this context, human metapneumovirus has emerged as a clinically relevant pathogen, with recent multicentre data showing a substantial burden of severe disease and in-hospital mortality among adult inpatients [346]. The main respiratory viruses associated with severe pneumonia in adults and their key clinical features are summarised in Table 20. Table 20.Emergent respiratory viruses associated with severe pneumonia in adultsVirusKey clinical features Human metapneumovirus Severe pneumonia in elderly patients; frequently ICU admission Adenovirus ARDS and multiorgan involvement; high mortality in immunocompromised hostsRhinovirus and EnterovirusFrequent detection in severe respiratory failure; prolonged sheddingRespiratory syncytial virusSevere disease in elderly and comorbid adultsParainfluenza virusLower respiratory tract infection; severe disease in vulnerable patientsICU: intensive care unit; ARDS: acute respiratory distress syndrome.Adapted from Philippot Q et al., Lee N et al., Zhou F et al., Piralla A et al., van Someren Gréve et al., and van der Bie et al. [341–346]. Emergent respiratory viruses associated with severe pneumonia in adults ICU: intensive care unit; ARDS: acute respiratory distress syndrome. Adapted from Philippot Q et al., Lee N et al., Zhou F et al., Piralla A et al., van Someren Gréve et al., and van der Bie et al. [341–346]. The pathophysiology of viral pneumonia is characterised by a complex interaction between direct viral cytopathic effects and host immune responses. Respiratory viruses exhibit distinct epithelial tropism, leading to epithelial injury, impaired mucociliary clearance, and disruption of barrier integrity [347,348]. These mechanisms are accompanied by immune dysregulation, with excessive inflammatory responses contributing to diffuse alveolar damage and respiratory failure [347,349]. Histopathological studies have provided important insights into these processes, demonstrating necrotising bronchopneumonia, interstitial inflammation, alveolar haemorrhage, and the presence of multinucleated inflammatory cells in severe viral infections [350]. In human metapneumovirus infection, lymphocytic interstitial infiltrates and epithelial inclusions have been described, supporting the concept of direct pulmonary injury and immune-mediated damage [351,352]. Epidemiological studies conducted in hospital wards and intensive care units have confirmed high detection rates of respiratory viruses in patients with severe pneumonia. Multicentre ICU cohorts have reported viral identification in a substantial proportion of mechanically ventilated patients, with some viruses detected exclusively in lower respiratory tract samples, highlighting the limitations of upper airway testing in severe disease [343,344]. Viral–bacterial coinfection represents a frequent and clinically relevant phenomenon. Coinfection has been associated with increased disease severity, higher requirements for organ support, and worse clinical outcomes in patients with severe community-acquired pneumonia [353,354]. In addition, cases of pure viral sepsis have been described, highlighting that respiratory viruses alone may trigger systemic inflammatory responses and multiorgan dysfunction [355]. Diagnostic interpretation remains challenging in clinical practice, as can be seen in Table 21. Although multiplex PCR has markedly improved viral detection, uncertainty persists regarding the clinical significance of positive results, particularly in the setting of prolonged viral shedding or asymptomatic carriage [349]. Cycle threshold values have been explored as potential surrogates of viral load; however, available evidence does not support their routine use for prognostic assessment [356,357]. Biomarkers such as procalcitonin may assist in distinguishing viral from bacterial infection, although their performance is imperfect and should be interpreted within the overall clinical context [358]. Table 21.Key diagnostic and prognostic challenges in viral pneumoniaFeatureClinical implicationsMultiplex PCR positivityDifficulty distinguishing infection from colonizationUpper vs lower respiratory samplesLower tract samples increase diagnostic yieldViral–bacterial coinfectionAssociated with increased severity and mortalityCycle threshold values in PCRLimited value for predicting outcomesViraemiaMarker of severity in selected infectionsBiomarkers (procalcitonin)Supportive but insufficient to rule out coinfectionPCR: polymerase chain reaction. Key diagnostic and prognostic challenges in viral pneumonia PCR: polymerase chain reaction. Adenovirus infection has been associated with a high risk of acute respiratory distress syndrome even in immunocompetent adults, highlighting its potential for severe pulmonary injury [359]. In addition, viraemia has emerged as a potential marker of disease severity in selected viral infections. Systematic reviews have shown an association between viraemia and worse outcomes in respiratory viral infections, particularly in adenovirus disease [360]. In immunocompromised patients, serum PCR has demonstrated value in predicting disseminated infection, which is associated with high mortality [361,362]. Therapeutic options for viral pneumonia remain limited. Current management in the intensive care unit is largely supportive, as highlighted in recent comprehensive reviews of viral infections in critically ill patients [363]. Specific antiviral therapy is available only for selected pathogens and patient populations, particularly in the context of adenovirus infection in transplant recipients [361,364]. Consequently, supportive care remains the cornerstone of management. Despite confirmed viral aetiology, antibiotic use remains common due to concern for bacterial coinfection, contributing to antimicrobial overuse [353]. The role of adjunctive therapies, including corticosteroids, remains controversial and may depend on individual patient characteristics and inflammatory phenotypes [365]. Update on fungal pneumonia Current epidemiology of fungal pneumonia in critically ill patients: New settings. Fungal pneumonia is an emerging complication in ICUs and is associated with high morbidity and mortality. While invasive fungal infections were traditionally specific to patients with prolonged neutropenia or haematological diseases, multiple studies have demonstrated a significant epidemiological change, with a sustained increase in cases among critically ill non-neutropenic patients [366,367]. This phenomenon is related to the increased survival of immunocompromised patients, the widespread use of corticosteroids and immunomodulatory drugs, and the emergence of severe viral infections such as influenza and COVID-19 [368–371]. Aspergillus fumigatus is the main aetiological agent involved in invasive pulmonary aspergillosis (IPA), due to its environmental ubiquity, ability to grow at 37 °C, and ease of reaching the alveolar space [366]. Classic studies conducted in ICUs have already shown that the true incidence of IPA is underdiagnosed and is higher in studies that specifically address this issue. In these initial cohorts, the incidence ranged from 5% to 8%, with mortality rates exceeding 50%, especially in patients with chronic obstructive pulmonary disease, liver failure, or prolonged exposure to corticosteroids [367]. The influenza A (H1N1) pandemic marked a turning point in the understanding of IPA. Multicentre studies demonstrated a high frequency of influenza-associated pulmonary aspergillosis (IAPA) in critically ill patients, with incidences close to 19% and a significant prognostic impact [93,369,372]. The retrospective study by the Dutch-Belgian group confirmed these findings and established IAPA as a distinct clinical entity, associated with higher mortality and worse respiratory outcomes [371]. Subsequently, national registries, such as the ENVIN study, have corroborated the relevance of Aspergillus spp. as an emerging pathogen in Spanish ICUs [373]. During the SARS-CoV-2 pandemic, COVID-19-associated pulmonary aspergillosis (CAPA) was described, with its incidence showing marked variability between studies, ranging from 0.7% to 34% depending on the diagnostic strategy used [374,375]. The pathophysiological mechanisms include virus-induced pulmonary epithelial damage, alteration of innate immunity, and widespread use of corticosteroids and immunomodulatory treatments [376]. Both for CAPA and IAPA, bronchoinvasive and angioinvasive forms have been described, representing a continuum of clinical severity [374,377]. The diagnosis of API in critically ill patients remains one of the main clinical challenges. The classic EORTC/MSG criteria have limited sensitivity in non-neutropenic patients, which led to the development of specific algorithms such as AspICU [378]. The culture of respiratory samples has a sensitivity of less than 50% and a poor ability to differentiate colonisation from infection [367]. The determination of galactomannan in bronchoalveolar lavage and molecular techniques have demonstrated greater diagnostic performance, especially when integrated into structured clinical algorithms [379]. The most recent consensus definitions propose criteria adapted for critically ill patients, including IAPA and CAPA [380,381]. The number of immunocompromised patients admitted to the ICU has increased progressively in recent decades. This group includes patients with haematological malignancies, haematopoietic stem cell transplantation, or prolonged immunosuppressive treatments, which significantly increase the risk of IPA. In this context, risk stratification and early recognition of infection are essential, as mortality remains high even with early antifungal treatment [382–385]. Table 22 summarises the main epidemiological features and diagnostic procedures for fungal pneumonia in critically ill patients. Table 22.Main fungal pneumonia entities in critically ill patientsEntityMain risk factorsTypical populationKey diagnostic toolsReported mortalityIPAImmunosuppression, prolonged corticosteroid therapy, COPDHematologic malignancies, solid organ or stem cell transplantation, ICU patientsBAL culture, galactomannan, PCR40–60%IAPASevere influenza, epithelial lung damageCritically ill patients with severe influenzaBAL galactomannan, PCR>50%CAPASevere COVID-19, corticosteroids, immunomodulatory therapyMechanically ventilated COVID-19 patientsBAL galactomannan, PCR, ICU-specific diagnostic algorithms40–60%Pulmonary mucormycosisDiabetes mellitus, corticosteroids, metabolic acidosisCOVID-19 patients and other immunocompromised hostsMucorales PCR, histopathology>60%IPA: invasive pulmonary aspergillosis; IAPA: influenza-associated pulmonary aspergillosis; CAPA: COVID-19 associated pulmonary aspergillosis; COPD: chronic obstructive pulmonary disease; BAL: bronchoalveolar lavage; PCR: polymerase chain reaction; ICU: intensive care units. Main fungal pneumonia entities in critically ill patients IPA: invasive pulmonary aspergillosis; IAPA: influenza-associated pulmonary aspergillosis; CAPA: COVID-19 associated pulmonary aspergillosis; COPD: chronic obstructive pulmonary disease; BAL: bronchoalveolar lavage; PCR: polymerase chain reaction; ICU: intensive care units. At the same time, the COVID-19 pandemic highlighted the emergence of mucormycosis associated with viral infection. Studies conducted in India demonstrated a strong association with diabetes mellitus, corticosteroid use, and uncontrolled hyperglycaemia [386–388]. In Europe, series reporting the detection of Mucor spp. in respiratory samples from critically ill patients have also been described, highlighting the usefulness of PCR techniques for early diagnosis [389]. International guidelines recommend an aggressive diagnostic approach and early initiation of liposomal amphotericin B upon clinical suspicion [390,391]. Treatment of invasive pulmonary aspergillosis in the critically ill patient. IPA is one of the most severe infectious complications affecting critically ill patients in ICUs. Although it is traditionally associated with profound immunosuppression, IPA is increasingly recognised in non-neutropenic critically ill patients, particularly those with acute respiratory failure requiring mechanical ventilation. Despite improvements in diagnostic tools and antifungal therapy, crude mortality in ICU patients with IPA remains unacceptably high, ranging from 46% to 82% [392,393]. The global burden of severe fungal diseases continues to rise, with Aspergillus spp. representing a major contributor to morbidity and mortality worldwide [393]. Epidemiological studies have revealed that IPA is frequently underdiagnosed in ICUs, primarily due to non-specific clinical symptoms and the limitations of traditional diagnostic methods [372,392]. Delayed diagnosis directly results in delayed antifungal therapy, which is one of the strongest predictors of a poor outcome [394]. From a microbiological perspective, the epidemiology of Aspergillus infections has evolved considerably. Advances in molecular techniques have revealed cryptic species within the Aspergillus complex that cannot be distinguished by conventional morphological characteristics, but which may exhibit intrinsic or acquired antifungal resistance [395]. Population-based surveillance studies in Europe have demonstrated significant rates of resistance to amphotericin B and triazoles, raising concerns regarding empirical therapy [396]. Resistance mechanisms include mutations in target enzymes, overexpression of efflux pumps, and complex transcriptional networks, which may also contribute to cross-resistance among antifungal classes [397]. When selecting an antifungal for critically ill patients, it is important to consider not only the in vitro susceptibility profile, but also the drug’s spectrum, resistance patterns, pharmacokinetics, pharmacodynamics, potential drug–drug interactions, and safety, as seen in Table 23. Triazoles remain the cornerstone of first-line therapy for IPA [372,398]; however, their use is complicated by extensive metabolism through cytochrome P450 enzymes. Clinically relevant interactions, such as those observed with metamizole or other commonly used ICU medications, can significantly reduce exposure to the antifungal agent and compromise its efficacy [399,400]. Furthermore, pharmacovigilance studies have highlighted variability in toxicity profiles, including sex-related differences in adverse events [401]. Table 23.Key factors influencing antifungal treatment in critically ill patients with invasive pulmonary aspergillosisFactorClinical implicationsAntifungal spectrumEmpirical therapy should cover Aspergillus spp. and emerging species according to local epidemiology.Antifungal resistancePresence of cryptic species and resistance to azoles or amphotericin B may lead to therapeutic failure.Drug–drug interactionsCommon ICU medications may significantly reduce antifungal plasma concentrations, particularly with azoles.Safety profileHepatic, renal and neurological toxicity may limit drug selection and treatment duration.Organ dysfunctionRenal or hepatic failure alters drug clearance and requires dose adjustment.Extracorporeal therapiesECMO and renal replacement therapy modify pharmacokinetics and increase the risk of underexposure.Therapeutic drug monitoringAllows optimization of efficacy and minimization of toxicity in complex ICU patients.ICU: intensive care unit; ECMO: extracorporeal membrane oxygenation.Adapted from Brinkman DJ et al., Xu Q et al., Canales-Siguero D et al., Thompson GR et al., and Natesan SK et al. [372,398–401]. Key factors influencing antifungal treatment in critically ill patients with invasive pulmonary aspergillosis ICU: intensive care unit; ECMO: extracorporeal membrane oxygenation. Adapted from Brinkman DJ et al., Xu Q et al., Canales-Siguero D et al., Thompson GR et al., and Natesan SK et al. [372,398–401]. Alterations in pharmacokinetics and pharmacodynamics in critical illness pose a significant challenge to antifungal treatment, as can be seen in Table 24. Increased volume of distribution, hypoalbuminemia, altered hepatic metabolism, renal dysfunction, and extracorporeal therapies frequently lead to subtherapeutic antifungal concentrations when standard dosing regimens are applied. Large, multicenter studies have confirmed that conventional antifungal doses often fail to achieve predefined pharmacokinetic/pharmacodynamic targets in ICU patients [402,403]. Table 24.Major pharmacokinetic and pharmacodynamic alterations of antifungal agents in critically ill patientsAlterationUnderlying mechanismClinical consequenceIncreased volume of distributionSystemic inflammation and third-space fluid accumulationReduced initial plasma concentrationsHypoalbuminemiaDecreased protein bindingIncreased variability in the free drug fractionHepatic dysfunctionImpaired drug metabolismRisk of toxicity or subtherapeutic exposureRenal dysfunctionAltered drug eliminationNeed for dose adjustmentECMODrug sequestration within the extracorporeal circuitSignificant reduction in plasma levelsHemodilutionPriming volume of extracorporeal circuitsInitial decrease in drug concentrationLack of TDMAbsence of pharmacological monitoringHigher risk of therapeutic failureECMO: extracorporeal membrane oxygenation; TDM: therapeutic drug monitoring.Adapted from Dolton MJ, et al., and Roberts JA, et al. [402,403]. Major pharmacokinetic and pharmacodynamic alterations of antifungal agents in critically ill patients Alternative antifungal agents, including isavuconazole and liposomal amphotericin B, represent important therapeutic options in specific clinical scenarios, such as intolerance, drug–drug interactions or refractory disease [398,404]. Real-world studies and experimental data have demonstrated the relevance of tissue penetration and safety profiles when selecting these agents in critically ill patients [404–406]. Extracorporeal membrane oxygenation (ECMO) can interfere with antifungal therapy by causing drugs to be trapped within the circuit, by reducing the volume of blood in the body due to priming and by altering clearance. Lipophilic agents with high protein binding, such as azoles, are particularly affected by this procedure [407,408]. These factors emphasise the importance of therapeutic drug monitoring and personalized dosing strategies[402,403]. Beyond pharmacology, the timing of antifungal therapy is also critical. Consistent experimental and clinical data demonstrate that delays during the first 96 hours of IPA are associated with rapid angioinvasion, tissue necrosis, and irreversible lung damage [394]. Consequently, early empirical or pre-emptive antifungal therapy, guided by clinical suspicion and available biomarkers, is essential for high-risk ICU patients. New therapeutic evidence in the pneumonia due to P. jirovecii in the critically ill patient. Pneumocystis spp. was first identified as a protozoan at the beginning of the twentieth century but was reclassified as a fungus in 1988. There are six known species, only one of which can infect humans: P. jirovecii. Its life cycle has an asexual stage and a sexual stage [409]. The bond of Pneumocystis to the alveolar epithelium promotes the organism’s transition from the trophic stage to the cyst stage and leads to the production of inflammatory mediators, which ultimately result in lung injury and impaired gas exchange [410]. P. jirovecci is transmitted from human to human, with no animal reservoir. There are asymptomatic carriers and immunosuppressed people can develop clinical illness (pneumonia). Both asymptomatic carriers and patients with pneumonia caused by P. jirovecii (PjP) can transmit the pathogen, even during incubation period, which lasts from three to twelve weeks. In certain circumstances, isolation of these patients may be beneficial [411]. The clinical course of PjP differs between patients with and without HIV infection. Patients with HIV infection usually exhibit a clinical-radiological dissociation and a slower clinical course, whereas people without HIV infection usually experience a fulminant clinical course, with diffuse lung infiltrates and severe lung injury [409]. Diagnosing PjP can be challenging. As culturing of the pathogen is not possible, molecular techniques such as polymerase chain reaction (PCR) are considered the gold standard for diagnosis. However, qualitative PCR does not distinguish between infection and colonisation, so quantitative PCR is required. In these patients, coinfection can also happen, so multiple PCR is desirable. Biomarkers, such as 1,3-β-D-glucan, are also useful for diagnosing and follow-up [410]. The epidemiology of PjP has changed over the last few decades. Apart from classic risk factors such as HIV infection, there are now many emergent risk factors. P. jirovecii infection has a significant global impact and is currently the third most common cause of invasive fungal infection. The causes of this are improved diagnosis approaches and the growing number of immunosuppressed patients worldwide. Currently, 90% of patients with PjP do not have HIV infection. Treatment with corticosteroids is the main risk factor, even with intermittent regimens. Other important emerging risk factors include immunosuppression caused by malignancies and solid organ transplants, and other immunosuppressive treatments such as rituximab and other monoclonal antibodies that target the CD20 antigen, as well as monoclonal antibodies against tumour necrosis factor-alpha (TNFα), mammalian target of rapamycin (mTOR) inhibitors, and calcineurin inhibitors [412,413]. Severe cases of the SARS-CoV-2 infection can lead to PjP due to the resulting immunosuppression, which is caused by the infection itself and by immunosuppressive treatment with corticosteroids and other drugs [414]. Primary immunodeficiencies are also important risk factors for PjP [412]. As previously mentioned, patients without HIV infection have a distinct clinical course. They have non-classic risk factors, are not usually prescribed prophylaxis, and reactivation of CMV has a synergistic effect with P. jirovecii infection. Clinicians have a low degree of suspicion, resulting in frequent late diagnosis [415]. Fortunately, trimethoprim-sulfamethoxazole (TMP-SMX) is highly effective in preventing PjP. For patients who are intolerant or allergic to TMP-SMX, other alternatives are available, including dapsone alone, dapsone plus pyrimethamine and leucovorin, atovaquone, and aerosolised pentamidine. Prophylaxis is indicated for patients at risk of developing PjP due to underlying risk factors and must be continued until these factors have been resolved and immune function is restored [410,412,416,417]. The treatment for P. jirovecii targets critical points in its metabolic process. TMP-SMX and dapsone target folate synthesis; atovaquone and primaquine target the mitochondrial electron transport chain; and echinocandins inhibit the formation of the cell wall. The action mechanisms of other active drugs against this microorganism, such as clindamycin and pentamidine, are unknown. TMP-SMX is the preferred drug for treating PjP. However, the serious potential side effects, which are more common among critically ill patients due to the need for higher doses, and the emergence of drug resistance associated with mutations in dihydropteroate synthase, must be considered. Furthermore, prophylactic breakthroughs and treatment failures associated with TMP-SMX, and second-line therapies have high relapse and recurrence rates [417,418]. A recently published network meta-analysis shows that TMP-SMX and intravenous pentamidine are the most effective drugs against P. jirovecii, with regimens containing TMP-SMX resulting in lower mortality rates, albeit with a higher incidence of side effects. Newer therapeutic approaches are needed to avoid the disadvantages associated with current available regimens [419]. Echinocandins are highly effective at destroying spore cysts by inhibiting the synthesis of 1,3-β-D-glucan. This destruction dramatically impairs the contagiousness of the illness. However, echinocandins do not affect trophic forms, which remain unaltered in animal models of PjP treated with echinocandins. One of the main causes of the inflammatory response that destroys the lungs in PjP is 1,3-β-D-glucan [420]. Caspofungin monotherapy is ineffective for PjP, but when combined with another drug that is active against P. jirovecii, it achieves better survival rates and a lower degree of lung inflammation than classic therapies [421,422]. A case series of 14 patients with PjP, who were HIV-negative and contraindicated for TMP-SMX, were treated with caspofungin and clindamycin. The clinical outcomes were very good: thirteen patients survived and there were no relapses. Although more studies are needed, this combination may be a good alternative to conventional regimens [423]. Other antifungal drugs, such as rezafungin and oteseconazole, are promising alternatives that may constitute therapeutic alternatives in the future [424]. Brief update in antimicrobial therapy Delafloxacin. The world is facing an increasing threat from antimicrobial resistance. This problem is linked to deaths caused by lower respiratory tract infections. In fact, antimicrobial resistance was responsible for more than 1.5 million pneumonia-related deaths worldwide. S. pneumoniae is the fourth most common cause of this problem, after E. coli, S. aureus and K. pneumoniae [425]. This problem becomes more relevant for patients with severe CAP, as failure of the initial treatment can be hazardous. However, when deciding on the best treatment, clinicians must consider not only the severity of the illness but also risk factors for resistant microorganisms and colonisation studies [94]. In fact, empirical anti-MRSA therapy might not be beneficial to most patients hospitalized with pneumonia, as it was not associated with reduced mortality. S. pneumoniae remains the most frequently isolated microorganism in CAP, in both for immunocompetent and immunocompromised patients, and remains an important cause of mortality. Among immunosuppressed patients, MRSA and P. aeruginosa are more frequently isolated. Moreover, HAP has a very high mortality rate, that ranges from 13% to 28%. So, in these complex settings, a tailored therapeutic regimen is essential for improving patient outcomes and newly developed antibiotics can aid to achieve this [138,178,181,182,203,269,426,427]. Fluoroquinolones are a therapeutic alternative for pneumonia treatment. For treatment of severe CAP, levofloxacin combined with a beta-lactam antibiotic is better than other combinations of fluoroquinolones plus a beta-lactam [428]. However, the current official recommendations are to prescribe a macrolide plus a beta-lactam to treat severe CAP instead of fluoroquinolones, which should only be prescribed to patients allergic to beta-lactams [175,429]. This is due to the associated side effects of fluoroquinolone treatment, which include nephrotoxicity, photosensitivity, hepatotoxicity, toxicity to the central nervous system, cardiotoxicity and connective tissue-related events such as tendon rupture and aortic aneurysm [430]. Despite these disadvantages, levofloxacin resistance among S. pneumoniae is rare and it is a good alternative to other oral antibiotics [431,432]. Delafloxacin is an interesting new fluoroquinolone. It is more active in acidic conditions, and its antimicrobial potential is higher than that of older fluoroquinolones, including against Gram-positive bacteria and anaerobes [433]. Delafloxacin inhibits bacterial DNA gyrase and topoisomerase, demonstrating potent activity against both. This dual targeting decreases resistance and prevents the emergence of mutants at lower concentrations than other fluoroquinolones. It exhibits a broad spectrum of in vitro activity against both Gram-positive and Gram-negative bacteria, including MRSA [434–444]. It also has a synergistic interaction with third generation cephalosporins against S. pneumoniae [445]. It also exhibits good activity against P. aeruginosa and S. maltophilia, outperforming ciprofloxacin and levofloxacin, and maintaining its effectiveness in an acidic environment [446,447]. Furthermore, the side-effect profile of delafloxacin is more favourable than that of other fluoroquinolones, as it does not have the potential to cause QT prolongation or phototoxicity [434,448–450]. The DEFINE-CABP clinical trial established the non-inferiority of delafloxacin compared to moxifloxacin for treating CAP, and it was even superior to moxifloxacin for patients with chronic obstructive pulmonary disease (COPD) or asthma. Most microbiological isolates were sensitive to delafloxacin, and 30% of patients had at least one atypical pathogen detected at baseline. The side-effect profile was very similar to that of the comparator [451,452]. Therefore, delafloxacin represents a potential option for monotherapy treatment in CAP, with a broad antimicrobial spectrum. However, real-world evidence about its use is needed. Ceftaroline. Ceftaroline is a fifth-generation cephalosporin with a broad spectrum of antimicrobial activity. It irreversibly inhibits penicillin-binding proteins (PBPs) and promotes lysis and cell death. It is active against MRSA. Although it is primarily indicated for treating CAP and soft tissue infections, it can also be useful for treating HAP. It is effective against a broad variety of Gram-positive and Gram-negative bacteria, but P. aeruginosa and producing extended spectrum beta-lactamase (ESBL) Gram-negative bacteria are not susceptible. Ceftaroline is an effective treatment option for CAP, but dosage adjustments are required for elderly patients with suspected MRSA, penicillin-resistant P. pneumoniae strains, and beta-lactamase-positive strains, as there is a risk of reduced susceptibility in these cases [453–458]. In terms of its pharmacokinetic properties, it penetrates the lung parenchyma well, a property that is enhanced by lung inflammation in porcine models [457]. Ceftaroline is predominantly renally excreted; consequently, in patients with ARC, elevated doses are required or even prolonged infusions. Conversely, patients requiring RRT require maintenance doses that are lower than those typically prescribed [2,458]. As previously mentioned, ceftaroline is clearly indicated for the treatment of CAP and has been shown to be more effective than ceftriaxone in clinical trials. In a multinational, real-world study, ceftaroline was most often prescribed empirically, usually as a rescue therapy. It was prescribed more frequently in ICUs and in combination with other antibiotics. Prescribing it as a first-line agent was associated with better clinical outcomes [2,456]. Although there are no clinical trials evaluating the efficacy of ceftaroline in treating MRSA and HAP, real-world evidence suggests that it is an effective alternative with fewer toxic side effects than vancomycin and linezolid. It is particularly useful in cases where vancomycin or linezolid have failed, in patients with associated bacteraemia or empyema, and in patients at high risk of experiencing adverse effects from other antimicrobial agents. Finally, it is equally safe and effective compared to ceftobiprole [453,455]. Ceftobiprole. Ceftobiprole is also a fifth-generation cephalosporin with chemical properties that give it fast bactericidal activity against both Gram-negative and Gram-positive bacteria. These include ceftazidime-sensitive P. aeruginosa and SAMR, but not ESBL Gram-negative bacteria. It has an excellent activity against S. pneumoniae, as does ceftaroline [459–463]. It is renally eliminated, has very low protein binding, and is highly stable in its preparations. Its pharmacokinetics are linear [464]. The susceptibility of Gram-positive bacteria isolated in European hospitals is almost 100%, whereas the susceptibility of Enterobacterales and P. aeruginosa ranges between 60% and 80% [465]. There is extensive clinical experience with ceftobiprole, and multiple clinical trials have proven its effectiveness against various types of infection, including severe CAP and HAP, both for adults and paediatric patients [182,203,466–472]. This antimicrobial, along with other relatively new antibiotics, provides a wide range of options beyond the traditional choice between third-generation cephalosporins and fluoroquinolones. This allows for a more tailored approach to treating patients with CAP, HAP and VAP. In fact, the PK/PD characteristics of ceftobiprole make it a good choice for treating severe infections in an outpatient setting. However, in Spain, it is currently used primarily as rescue therapy in patients with comorbidities and severe infections. Due to its safety profile, tolerability, and efficacy, this drug is recommended as a first-line treatment for the following conditions: severe community-acquired pneumonia (CAP) without risk factors for multidrug-resistant Gram-negative bacteria; patients with structural pneumopathy; HAP and VAP without risk factors for MDR bacteria (e.g. negative colonisation studies, early ICU pneumonia); pneumonia with bacteraemia; necrotising pneumonia; and lung abscess [473,474]. Aztreonam-avibactam. The choice of empirical treatment for VAP is challenging due to bacterial resistance. In fact, the rate of inappropriate empirical treatment for VAP caused by Gram-negative bacteria is 35–40% [475]. Newer beta-lactams decrease this risk as their spectrum against multiresistant Gram-negative bacteria is broader. Aztreonam is active against metallo- β-lactamases (MBL, Class B of Ambler classification, e.g. NDM) but it is vulnerable to other types of beta-lactamases, such as class A (e.g. KPC) and D (e.g. OXA-48). The combination of aztreonam with avibactam is active against all those types of beta-lactamases [476]. Aztreonam-avibactam is effective against isolates that are resistant to ceftazidime-avibactam and meropenem-vaborbactam, and even against S. maltophilia [477]. The 3:1 ratio of commercial preparations has a favourable PK/PD profile, with a high AUC and excellent penetration of the lung parenchyma. Its PK/PD profile is superior to that of the combination of aztreonam plus ceftazidime-avibactam. However, the latter combination may be useful for certain P. aeruginosa and S. maltophilia isolates, as well as in cases of bacteraemia [478,479]. The clinical trial REVISIT established aztreonam-avibactam as non-inferior to meropenem for treating HAP and VAP caused by Gram-negative bacteria, with a good safety profile [480]. So, aztreonam-avibactam is an effective treatment for HAP-VAP caused by aerobic Gram-negative bacteria when other options are limited. It is also more effective than other alternatives in cases of CPE infection. For patients infected with S. maltophilia, aztreonam-avibactam enables cefiderocol to be saved and is better tolerated than TMP-SMX. Finally, in cases involving multidrug-resistant P. aeruginosa, aztreonam-avibactam helps to avoid the overuse of carbapenems and cefiderocol [481]. Ceftazidime-avibactam. Ceftazidime/avibactam (CAZ-AVI) has established itself as one of the main therapeutic options for severe infections caused by β-lactamase-producing Gram-negative bacilli, particularly KPC- or OXA-48-producing Enterobacterales and multidrug-resistant P. aeruginosa. Evidence from clinical trials, observational studies, and meta-analyses supports its clinical efficacy and safety profile, as well as its role in the context of pneumonia and infection in critically ill patients. A combined analysis of phase 3 clinical trials demonstrated that CAZ-AVI has clinical cure and microbiological eradication rates comparable to or superior to reference treatments in infections caused by β-lactamase-producing Gram-negative pathogens, with a favorable safety profile even in high-risk populations. These results support its use in severe respiratory infections, including ventilator-associated pneumonia [482]. In the case of multidrug-resistant P. aeruginosa, a recent meta-analysis showed that CAZ-AVI is associated with a higher probability of clinical success and lower mortality compared to other active antimicrobials, although without consistent differences in microbiological eradication, highlighting the importance of the clinical context and comprehensive patient management [483]. In line with these findings, the multicentre observational CACTUS study did not show significant differences in mortality between CAZ-AVI and ceftolozane/tazobactam, although it showed variations in the emergence of resistance and in certain clinical subgroups [484]. The comparison with other new β-lactams against CPE has also been the subject of multiple studies. In critically ill patients with KPC-Klebsiella pneumoniae infections, CAZ-AVI showed similar clinical results to meropenem/vaborbactam, with no clear differences in mortality, although with nuances related to baseline severity and the site of infection [485]. Retrospective multicenter studies conducted in different geographical contexts, such as Turkey, reinforce the efficacy of CAZ-AVI compared to alternative therapies in infections caused by OXA-48 or KPC-producing Enterobacterales [486]. Experience in critically ill patients confirms that CAZ-AVI maintains adequate efficacy in complex settings such as the ICU, with acceptable clinical response and survival rates, even in the presence of severe comorbidities and organ dysfunction [487]. However, the selective pressure associated with its use has raised concerns about the emergence of resistance. Recent studies describe a progressive increase in resistance to CAZ-AVI globally, especially in P. aeruginosa and K. pneumoniae, reinforcing the need for strategies for prudent use and therapeutic optimization [488,489]. From a pharmacokinetic and pharmacodynamic point of view, several studies highlight the importance of individualizing CAZ-AVI dosing, especially in critically ill patients with alterations in distribution volume or renal clearance. Population studies and PK/PD models suggest that intensified strategies, including prolonged or continuous infusion, allow more aggressive targets to be achieved and could reduce the risk of treatment failure and the development of resistance [490–492]. In this context, the ZAVICONT trial, currently under development, prospectively evaluates continuous infusion versus intermittent administration in severe infections caused by OXA-48 K. pneumoniae and P. aeruginosa, and will provide future evidence on the best administration strategy [493]. The ecological impact of CAZ-AVI treatment has also been evaluated. Recent data indicate that colonization by other resistant Gram-negative bacilli may occur during its use, highlighting the need for close microbiological surveillance and antimicrobial optimisation programmes [494,495]. Stewardship studies based on aggressive PK/PD targets and multidisciplinary approaches have demonstrated not only better clinical outcomes but also a lower emergence of resistance during treatment [496]. Overall, the available evidence establishes ceftazidime/avibactam as a key component in the treatment of severe infections and pneumonia caused by multidrug-resistant Gram-negative bacteria. To maximise its clinical efficacy and preserve its long-term activity, its use should be integrated into individualised strategies supported by PK/PD principles, microbiological surveillance and antimicrobial optimisation programmes. Cefepime-enmetazobactam. Cefepime/enmetazobactam is the first approved combination of a fourth-generation cephalosporin and a new β-lactamase inhibitor. It is designed to address the increasing clinical and epidemiological challenges posed by multidrug-resistant Enterobacterales infections [497,498]. The combination has been developed with the aim of preserving the usefulness of cefepime against ESBLs and AmpC, offering an alternative to carbapenems in selected scenarios [499–503]. Enmetazobactam (AAI101) is a penamic-type β-lactamase inhibitor, structurally related to tazobactam, but with greater potency against CTX-M-type ESBLs and some plasmid AmpCs, with no activity against carbapenemases [499,500,504]. Its association with cefepime restores antibacterial activity against ESBL- and AmpC-producing Enterobacterales, maintaining the intrinsic broad spectrum of cefepime [497,498,504]. From a microbiological point of view, multiple in vitro studies and experimental models have demonstrated a significant reduction in minimum inhibitory concentrations against ESBL-producing E. coli and K. pneumoniae, even at high bacterial inoculum [504]. Cefepime is a fourth-generation cephalosporin with a broad spectrum of bactericidal activity and high stability against degradation by chromosomal and plasmid-mediated AmpC cephalosporinases, as well as OXA-48–type enzymes. This stability translates into preserved activity against OXA-48–producing Enterobacterales. Enmetazobactam is a novel penicillanic acid sulfone β-lactamase inhibitor, structurally related to tazobactam, with potent activity against ESBLs of the CTX-M, TEM, and SHV types, as well as certain class A β-lactamases, including some KPC variants with Ω-loop mutations. The cefepime/enmetazobactam combination plays a therapeutic role against Enterobacterales with OXA-48–mediated resistance associated with ESBL co-production, as inhibition of these enzymes consistently restores cefepime activity [505,506]. Pharmacokinetic and pharmacodynamic studies show that cefepime/enmetazobactam has profiles comparable to those of cefepime, with good tissue and lung penetration and an adequate free fraction to optimise time above the MIC (T>MIC), especially when prolonged or continuous infusion strategies are used [498,507–509]. These aspects are particularly relevant in critically ill patients and in nosocomial pneumonia. The clinical evidence is based primarily on the ALLIUM trial, which demonstrated non-inferiority and, in some analyses, microbiological superiority compared to piperacillin/tazobactam in complicated urinary tract infections and acute pyelonephritis, including infections caused by ESBL-producing Enterobacterales [510]. Subsequent studies and network meta-analyses have reinforced its clinical efficacy and safety profile in this context [511]. The clinical interest in cefepime/enmetazobactam is part of carbapenem-saving strategies, given the association between its use and the emergence of resistance, alterations in the intestinal microbiome, and selection of resistant non-fermenting pathogens. Several epidemiological and microbiota studies support the negative impact of extensive carbapenem use, reinforcing the role of β-lactam alternatives when microbiologically appropriate [93,509,512–520]. This ecological advantage is partly related to the absence of relevant anaerobic activity, in contrast to carbapenems. Recent guidelines from the Infectious Diseases Society of America state that, in selected scenarios and if there is no carbapenemase production and local susceptibility data is available, cefepime/enmetazobactam can be used to treat ESBL-producing Enterobacteriaceae infections [39]. Therefore, cefepime/enmetazobactam represents a valuable addition to the therapeutic armamentarium against multidrug-resistant Enterobacterales. It is a particularly useful option for ESBL- and AmpC-producing isolates and a therapeutic alternative for selected OXA-48–producing Enterobacterales. Its use may contribute to carbapenem-sparing strategies and antimicrobial stewardship programs by allowing effective treatment while reducing selective pressure associated with carbapenem exposure. Meropenem-vaborbactam. Meropenem/vaborbactam is a combination of a broad-spectrum carbapenem with a boronic acid-derived β-lactamase inhibitor, developed for the treatment of serious infections caused by carbapenem-resistant Enterobacterales (CRE), especially those producing KPC-type carbapenemases. It has been authorised in the United States since 2017 and in the European Union since 2018 for complicated urinary tract infections, complicated intra-abdominal infections, nosocomial pneumonia (including ventilator-associated pneumonia), and associated bacteraemia, as well as for infections caused by aerobic Gram-negative bacilli with limited therapeutic options [521]. Meropenem/vaborbactam potently inhibits class A β-lactamases, particularly KPC, and to a lesser extent class C β-lactamases, with no activity against metallo-β-lactamases or OXA-48-type oxacillinases. This inhibition restores the activity of meropenem against CRE-KPC strains, with sensitivity rates close to 99% and a MIC90 of 1 mg/L in large collections of clinical isolates. The main resistance mechanisms described include mutations in the OmpK35 and OmpK36 porins, which reduce antibiotic entry [522]. From a pharmacokinetic point of view, the combination has a favourable profile. Studies in healthy volunteers have demonstrated adequate concentrations of meropenem and vaborbactam in alveolar epithelial fluid and alveolar macrophages, with lung/plasma penetration ratios of 63% and 65%, respectively, supporting its use in severe pneumonia [523]. These data are consistent with previous studies of other β-lactams and inhibitors in the pulmonary compartment [524]. The TANGO I randomised clinical trial compared meropenem/vaborbactam with piperacillin-tazobactam in complicated urinary tract infections. The combination demonstrated non-inferiority and superiority in clinical and microbiological response, with a lower emergence of resistance during treatment [525]. In the TANGO II trial, which included patients with severe CRE infections (urinary tract infections, intra-abdominal infections, nosocomial pneumonia, and bacteraemia), meropenem/vaborbactam showed a higher clinical response rate and a significant reduction in 28-day mortality compared to the best available therapy, in addition to a more favorable safety profile [526]. Beyond pivotal trials, various observational and compassionate use studies have confirmed the efficacy of meropenem/vaborbactam in KPC-Klebsiella pneumoniae infections, including critically ill patients and strains resistant to ceftazidime/avibactam. In an Italian multicenter study, clinical cure was achieved in more than 75% of cases, even in a context of high clinical severity. A post-hoc analysis showed better results when used as early treatment and in patients without previous antibiotic failure [527–529]. Comparison with ceftazidime/avibactam has been the subject of multiple studies. Retrospective data suggest comparable clinical and mortality outcomes, although with lower recurrence and lower resistance when meropenem/vaborbactam is used as monotherapy versus ceftazidime/avibactam, especially when the latter is used as monotherapy [530]. Large database studies in the US indicate a possible reduction in mortality, ICU length of stay, and the incidence of Clostridioides difficile infection with meropenem/vaborbactam compared to ceftazidime/avibactam, although detailed microbiological data are lacking [531]. Recent results in European ICUs show a better early clinical response with meropenem/vaborbactam, with no significant differences in 30-day mortality [485]. A key aspect in the clinical positioning of meropenem/vaborbactam is the context of emerging resistance to ceftazidime/avibactam. The emergence of KPC variants, especially the D179Y mutation in the ST258 clone, associated with resistance to ceftazidime/avibactam and an apparent recovery of sensitivity to carbapenems, has been described [532–534]. These mutations can also generate false negatives in carbapenemase detection tests and promote clonal spread in hospital settings, particularly in ICUs [534,535]. In this scenario, meropenem/vaborbactam remains active against most of these strains, although isolates with combined resistance mechanisms have been described [535]. The IDSA 2024 guidelines place meropenem/vaborbactam as one of the preferred options for the treatment of KPC-producing CRE infections outside the urinary tract, along with ceftazidime-avibactam and imipenem-relebactam, reserving cefiderocol as an alternative [39]. Overall, the available evidence positions meropenem/vaborbactam as a key, effective, and safe tool for the management of severe KPC-CRE infections in current clinical practice. Cefiderocol. Cefiderocol is a cephalosporin that acts as a siderophore, utilizing bacterial iron uptake systems to cross the outer membrane of Gram-negative bacilli. This unique mechanism confers stability against multiple resistance determinants, including metallo-β-lactamase production, porin alterations, and efflux pump overexpression. Recent narrative reviews and position papers have emphasised its potential in treating infections caused by metallo-β-lactamase producers, for which therapeutic alternatives are limited [536,537]. Severe pneumonia, especially that associated with mechanical ventilation in the ICU, continues to pose a significant clinical challenge due to multidrug-resistant Gram-negative bacilli. ENVIN-HELICS reports confirm that P. aeruginosa remains the predominant pathogen and that there has been a progressive increase in Stenotrophomonas maltophilia. Furthermore, a significant proportion of patients receive inadequate initial antibiotic treatment, which is associated with a worse clinical prognosis [93,538]. The high in vitro activity of cefiderocol against P. aeruginosa and other non-fermenting Gram-negative bacilli has been consistently demonstrated in European studies. Recent studies have shown that cefiderocol exhibits higher sensitivity rates than other β-lactams against carbapenem-resistant P. aeruginosa isolates, including those not susceptible to modern β-lactam/β-lactamase inhibitor combinations, according to the breakpoints established by CLSI and EUCAST [539–542]. Data from the SENTRY surveillance programme (2020–2023) in Europe and the United States confirm an almost 99% susceptibility rate against P. aeruginosa, including difficult-to-treat resistance phenotypes and isolates that produce carbapenemases, such as VIM or GES-5. Of particular clinical relevance is the fact that a lower level of cross-resistance is observed with cefiderocol than with ceftolozane-tazobactam, ceftazidime-avibactam, or imipenem-relebactam [542–546]. The potential for resistance selection during treatment with new β-lactams is a growing concern. Although experimental studies have shown that cefiderocol presents a low risk of selecting resistant mutants, under certain conditions, mechanisms of emerging resistance related to alterations in TonB-dependent transport or modifications of AmpC and PBP3 have been described. This reinforces the need for microbiological re-evaluation in the event of relapse or new episodes of sepsis [541,546,547]. From a pharmacokinetic standpoint, cefiderocol has demonstrated adequate pulmonary penetration. Studies in healthy volunteers and population pharmacokinetic analyses in patients with nosocomial pneumonia requiring mechanical ventilation have shown concentrations in the epithelial lining fluid that are proportional to plasma concentrations. The epithelial lining fluid (ELF)/plasma ratios are higher than those observed in healthy subjects, and the minimum concentrations are generally above the sensitivity cut-off points. This supports the achievement of pharmacodynamic targets with the recommended doses, which are adjusted for renal function [548]. The use of cefiderocol in the treatment of pneumonia is further supported by results obtained in animal models, where it has demonstrated efficacy in treating Stenotrophomonas maltophilia pneumonia under conditions of persistent neutropenia, a particularly challenging therapeutic scenario [549]. Evidence from real-world clinical practice is increasingly consistent. In prospective cohorts and multicentre observational studies such as CEFI-SITA and CEFI-ID, cefiderocol has frequently been used to treat lower respiratory tract infections and bacteraemia caused by multidrug-resistant Gram-negative bacilli. The clinical cure and mortality rates are consistent with the severity of the included patients [545,550]. In the PERSEUS study, which was conducted as part of a compassionate use programme in Spain, cefiderocol showed particularly favourable results in the treatment of P. aeruginosa infections. Clinical cure rates were above 80%, and the drug was more effective when administered early or as a first-line treatment. Consistent with this, the European results of the PROVE observational study confirm the effectiveness of cefiderocol in critical and high-risk patient groups [546,547]. The 2024 Infectious Diseases Society of America guidelines recommend cefiderocol as an alternative treatment for challenging P. aeruginosa infections outside the urinary tract. However, it is considered the preferred option for metallo-β-lactamase-producing isolates. Similarly, for metallo-β-lactamase-producing Enterobacterales, cefiderocol is considered one of the preferred strategies, alongside aztreonam-avibactam [39,537]. Overall, cefiderocol emerges as a particularly interesting therapeutic option for treating severe pneumonia caused by multidrug-resistant Gram-negative bacilli in the ICU. Its high microbiological activity, adequate intrapulmonary exposure, and favorable results in clinical trials and real-world clinical practice (especially when used early and guided by the resistance profile) justify its inclusion in the current therapeutic arsenal against respiratory infections caused by P. aeruginosa and metallo-β-lactamase-producing pathogens [39,93,536–538,542,545,547,550]. Isavuconazole. Isavuconazole is a broad-spectrum triazole that has established its role in the treatment of invasive fungal infections, especially in hospitalised patients with relevant comorbidities and severe respiratory disease [376,551,552]. Its efficacy against Aspergillus spp. and Mucor spp., together with a predictable pharmacokinetic profile and better tolerability compared to other azoles, has positioned it as a first-line alternative in complex scenarios, as reflected in pivotal trials and regulatory documents [553–556]. Clinical trials show that invasive fungal infections are still associated with a high mortality rate, particularly among patients with respiratory failure, ICU admission, or immunosuppression. Aspergillus fumigatus is the most common pathogen [376,557]. In this context, isavuconazole has produced consistent clinical outcomes consistent with those observed in clinical trials, registry studies, and European clinical guideline recommendations [553–556,558]. From a pharmacological point of view, isavuconazole has high oral bioavailability, linear pharmacokinetics, and lower interindividual variability [551,554,559]. Unlike voriconazole, it shortens the QT interval and has a more favourable drug interaction profile, although its metabolism by CYP3A4 remains relevant [560–562]. Studies in renal and moderate hepatic impairment in critically ill patients, including those on extracorporeal support, suggest that systematic dose adjustments are not necessary, although individualised clinical and pharmacological monitoring is recommended [563–566]. The adequate tissue distribution of isavuconazole in the lung has been confirmed in experimental models and clinical trials, reaching effective concentrations at the site of pulmonary infection, even in situations of complex immunodeficiency [567,568]. These data support its use in invasive fungal pneumonia, one of the most frequent and serious scenarios in the hospital setting. The emergence of azole resistance, especially in Aspergillus fumigatus, represents a growing challenge in Europe [557]. Although the prevalence of isavuconazole resistance is lower than for other triazoles, its use should be integrated into antifungal optimisation programmes and aligned with international guidelines and consensus recommendations [555,567]. Therapeutic monitoring, although not systematic, may be useful in selected situations such as clinical failure, relevant drug interactions, or critically ill patients [554,555]. In conclusion, the available evidence supports the use of isavuconazole as an effective and safe option for the treatment of invasive fungal infections, offering clear pharmacological advantages in complex patients with severe pneumonia. Conclusions Patient-centred strategies are key to improving outcomes in patients with CAP, HAP and VAP. Personalised therapy in critically ill patients is key to achieve better outcomes. Great inter and intraindividual variability poses a challenge to tailor therapies; therefore, resources such as TMD integrated with point-of-care tools, information and communication technologies (ICT) and artificial intelligence AI represent an opportunity to improve the management of these patients. The diagnostic process for pneumonia can be challenging. Chest radiography has important limitations for diagnosis of HAP/VAP. Chest ultrasound and computed tomography are good alternatives that improve the diagnostic accuracy. Nevertheless, in the absence of a new lung infiltrate, treatment of VAT is associated with better outcomes. Although biomarkers do not replace clinical assessment, they provide useful information for diagnosis and risk stratification and guide therapeutic decisions. MR-proADM and PCT are currently the most useful tools for implementing protocols in the ED and ICU. Regarding microbiological diagnosis, the combination of classical techniques, rapid diagnostic methods, and careful clinical correlation remains the most effective strategy to optimize it in critically ill patients with pneumonia. Syndromic molecular panels are very useful in guiding the empirical treatment of severe CAP, HAP, and VAP because they enable microorganisms and resistance mechanisms to be detected earlier than classical techniques. They are cost-effective and increase the likelihood of appropriate initial antimicrobial therapy. New technologies and therapeutic approaches can improve outcomes for CAP. Current evidence supports the view that severe CAP is the result of a complex interaction between incomplete aetiological diagnosis, host characteristics, respiratory microbiome, and bacterial virulence mechanisms. The integration of clinically driven diagnosis, rational use of antibiotics, and adjuvant therapies targeting pathogenesis is a promising avenue towards a more precise and personalised approach to this entity. Contemporary management of CAP requires the integration of scientific evidence, molecular diagnosis, antimicrobial stewardship and individual risk stratification, shaping a new paradigm aimed at improving clinical outcomes and optimising healthcare resources utilisation. Patients with severe CAP may benefit from corticosteroids. Those with higher inflammation, respiratory failure, and elevated CRP are more likely to obtain benefit from this treatment. For patients with severe CAP, an antibiotic regimen containing a macrolide and a beta-lactam is more effective than other regimens. Moreover, HAP and VAP are usually difficult to diagnose and treat. However, other factors relating to healthcare, such as the presence of risk or MDR microorganisms, an absence of treatment response, and concurrent surgery, make them even more challenging. In multi-resistant settings, the therapeutic approach to VAP should be based on a multidimensional strategy that combines early diagnosis, local epidemiological knowledge, appropriate empirical treatment, and continuous clinical reassessment. The ultimate goals are to improve prognosis and optimise the use of antimicrobials. A key aspect in the diagnosing of refractory pneumonia is close follow-up of patients with severe disease. The differential diagnosis includes non-infectious causes and infectious processes outside the lungs. PK/PD factors are also crucial for achieving good outcomes, so clinicians facing this complex problem must consider ARC, suboptimal focus control and proper drug delivery to the lung parenchyma. Postoperative pneumonia is a common and serious complication that can be prevented. Its development reflects the interaction between patient frailty, surgical complexity, and the quality of perioperative management. Prevention requires a multidisciplinary approach based on the early identification of risks and the implementation of integrated strategies aimed at improving surgical outcomes. Viruses, other than influenza and SARS-CoV-2, also pose challenges in the critically ill patient, and this is something that healthcare professionals must be aware of. RSV infection is common and causes morbidity and mortality comparable to that of Influenza. The clinical course differs between immunosuppressed patients and older adults. There are limited treatment options, and corticosteroids should be avoided. Fortunately, vaccines and monoclonal antibodies with high efficacy are now available to prevent adverse outcomes. Reactivation of HSV-1 and CMV is common in critically ill patients and is associated with poorer outcomes. Latent or active CMV infection has a negative immunomodulatory effect. Prophylaxis and treatment of viral replication have not been shown to benefit patients, but specific treatment is associated with better outcomes if clinical signs of viral infection are present. Respiratory viruses constitute a frequent and clinically relevant cause of pneumonia in adults. Their impact on disease severity and outcomes is comparable to that of influenza, particularly in elderly and comorbid patients. Diagnostic uncertainty, frequent coinfection, and limited therapeutic options continue to pose major challenges, emphasising the need for improved risk stratification and evidence-based management strategies. Finally, fungi are also an emerging problem in patients with severe pneumonia. Fungal pneumonia in critically ill patients is a growing clinical challenge. A better prognosis depends on adequate risk stratification, the rational use of biomarkers, the integration of specific diagnostic algorithms, and the early initiation of targeted antifungal treatment. Collaboration between specialists in intensive care, microbiology, and infectious diseases is essential to optimise the management of these complex infections. The management of invasive pulmonary aspergillosis in critically ill patients requires an integrated approach that combines early recognition with an understanding of the evolving epidemiology and potential emerging resistance mechanisms of the fungus. It also requires careful evaluation of drug interactions and individualised pharmacokinetic-guided dosing. Achieving optimal outcomes depends not just on antifungal susceptibility, but also on delivering the right drug, at the right dose, at the right time. P. jirovecii is a significant cause of severe respiratory failure in immunocompromised patients who are not infected with HIV. TMP-SMX is the recommended treatment for PjP, but its use in critically ill patients is limited due to potential serious side effects. Combining caspofungin with clindamycin is associated with a faster clinical response and better tolerance. Echinocandins also modulate inflammation by inhibiting 1,3-β-D-glucan synthesis, which is associated with reduced lung damage and improved oxygenation. Funding Statement CRL has received funding from 312D program “Fundaciones de Investigación Biomédica” from Madrid Community Expenditure Budget for 2025. Footnotes AI statement No artificial intelligence tools were used in the writing, analysis, interpretation, or review of this article. Funding CRL has received funding from 312D program “Fundaciones de Investigación Biomédica” from Madrid Community Expenditure Budget for 2025. Collaborators Salavert M (Hospital Universitari i Politecnic. La Fe. Valencia. GEIPC-SEIMC, SEQ), Barrasa H (Hospital Universitario de Araba. Vitoria. GTEIS-SEMICYUC, GEIPC-SEIMC), Rodríguez A (Hospital Universitari Joan XXIII. Tarragona. GTEIS-SEMICYUC, GEIPC-SEIMC), Andaluz D (Complejo Asistencial Universitario. Palencia. GTEIS-SEMICYUC, GEIPC-SEIMC), González del Castillo JM (Hospital Clínico Universitario San Carlos. Madrid. SEMES-Infurgsemes), Navarro D (Hospital Clínic Universitari de València. SEIMC), Sancho S (Hospital Universitari i Politecnic. La Fe. Valencia. GTEIS-SEMICYUC, GEIPC-SEIMC), Rascado P (Complejo Hospitalario Universitario de Santiago. A Coruña. GTEIS-SEMICYUC, GEIPC-SEIMC), Carbonell N (Hospital Clínic Universitari de València. GTEIS-SEMICYUC), Maseda E (Hospital Quirón Sur. Madrid. GTIPO-SEDAR, GEIPC-SEIMC), Abril V (Consorci Hospital General Universitari València, SEIMC), Del Pozo JL (Clinica Universitaria. Campus Pamplona. GEIPC-SEIMC), Merino E (Hospital General Universitario Dr. Balmis. Alicante. SEIMC), Fortún J (Hospital Universitario Ramón y Cajal. Madrid. SEIMC), De la Fuente C (Hospital Universitario Reina Sofía. Córdoba. GTEIS-SEMICYUC), Garnacho J (Hospital Universitario Virgen del Rocío. Sevilla. GTEIS-SEMICYUC), Ramirez-Galleymore P (Hospital Universitari i Politecnic. La Fe. Valencia. GTEIS-SEMICYUC), Nuvials X (Hospital Universitari Vall d’Hebron. Barcelona. GTEIS-SEMICYUC). Zaragoza R (Hospital Universitai Doctor Peset. València. GTEIS-SEMICYUC, GEIPC-SEIMC). Conflicts of interest No commercial, financial, or personal conflicts of interest were identified in relation to the research, authorship, and/or publication of this article. References