Infectious brain abscesses and granulomas: analysis of 110 episodes in adults Authors: Zahra Hesari, Mahboubeh Haddad, Fereshte Sheybani, Farzaneh Khoroushi, Ehsan Keykhosravi, Negar Morovatdar Published: 2024 ABSTRACT AbstractBackgroundInfectious brain abscesses and granulomas, characterized by localized collections of pus or inflammatory tissue within the brain parenchyma, pose significant clinical challenges due to their potentially life-threatening nature and complex management requirements.MethodsThis cross-sectional study investigated patients diagnosed with infectious brain abscesses and granulomas from March 1, 2012, to October 22, 2021, in Mashhad, Iran. Data were collected from adult patients admitted to the two primary referral centers for community-acquired neuroinfections and neuroinflammations. Demographic information, clinical features, laboratory and neuroimaging characteristics, and clinical outcomes were analyzed.ResultsA total of 110 episodes were identified in 106 patients, with a median age of 45 years (IQR 30-56.3) and 62.7% male. Predisposing conditions included immunocompromised states (27.5%), preceding otitis/mastoiditis (16.2%), sinusitis (13.3%), and pulmonary infections (17.2%). The most common clinical manifestations were headache (57.3%), fever (49.1%), altered consciousness (44.4%), and seizures (31.8%). Neuroimaging revealed that brain lesions were solitary in 51% and multiple in 48% of episodes. Surgical intervention was performed in 46.4% of cases. The in-hospital mortality rate was 24.5%, with significant associations found between mortality and factors such as age, altered consciousness, multiple brain lesions, and cerebellum and brainstem involvement. The median length of hospital stay was 28 days (IQR 16-46.5).ConclusionOur study underscores challenges in diagnosing and treating brain abscesses and granulomas, with high mortality rates (24.5%) despite advanced techniques. Age, altered consciousness, and lesion characteristics predict death. Addressing changing microbial patterns and improving diagnostics are vital for better outcomes, especially in low- and middle-income countries.Keywords: Brain abscess, Brain granuloma, Neuroimaging, Outcome, Gram-negative bacilli, Multiple ring-enhancing brain lesion Introduction Infectious brain abscesses and granulomas, characterized by localized collections of pus or inflammatory tissue within the brain parenchyma, present a formidable challenge in clinical practice due to their potentially life-threatening nature and complex management requirements [1]. These neurological complications arise from various infectious processes, including bacterial, fungal, and parasitic infections [2], leading to a spectrum of clinical manifestations ranging from subtle neurological deficits to fulminant sepsis. The diagnostic and therapeutic complexities associated with brain abscesses stem from their diverse etiologies, predisposing factors, and anatomical locations within the central nervous system. Epidemiologically, the incidence and burden of brain abscesses exhibit variations across different regions and populations, influenced by factors such as socioeconomic status, healthcare accessibility, and the prevalence of predisposing conditions. The annual incidence has been estimated at 0.4–1.3 per 100 000 inhabitants [3]. In high-income countries, advancements in diagnostic imaging, antimicrobial therapy, and neurosurgical interventions have been expected to enhance patient outcomes. Nevertheless, mortality rates persist at 8–13%, with substantial morbidity linked to neurological deficits and long-term complications [4–8]. Challenges still remain in achieving timely diagnosis, addressing underlying comorbidities, and implementing effective measures to prevent recurrent infections [3]. Conversely, in low and middle-income countries (LMICs), limited access to healthcare resources, diagnostic facilities, and specialized neurosurgical expertise pose significant barriers to effective management. Additionally, socioeconomic disparities and a high prevalence of infectious diseases contribute to the disproportionate burden of brain abscesses in these regions. Although limited high-quality studies on the epidemiology and outcome of brain abscesses in LMICs are available, existing research suggests that mortality rates in these countries range from 5 to 15% [9–15]. The substantial mortality in these regions often attributed to delays in diagnosis, restricted access to healthcare, and a lack of specialized neurosurgical services. Limited availability of diagnostic imaging, neurosurgical expertise, and antimicrobial therapy further exacerbates the burden of disease. Compounding these challenges is the absence of comprehensive epidemiological data and standardized treatment protocols, which hinder efforts to address this pressing public health issue in LMICs. Against this backdrop, community-acquired infectious brain abscesses and granulomas emerge as a focal point for clinical research and healthcare interventions aimed at improving patient outcomes and reducing the global burden of neurological infections. In this context, understanding the epidemiology, pathogenesis, and optimal management strategies for brain abscesses becomes paramount for clinicians and public health authorities alike. This study aims to contribute to this ongoing discourse by elucidating the clinical characteristics, predisposing factors, and outcomes associated with community-acquired infectious brain abscesses and granulomas in adults in a cross-sectional study in Mashhad, Iran between 2012 and 2021. Methods This study conducted a cross-sectional investigation of patients diagnosed with infectious brain abscesses and granulomas between March 01, 2012, and October 22, 2021. Data was collected from adult patients with community-acquired infectious brain abscesses and granulomas admitted to the two primary referral centers for individuals with community-acquired neuroinfections and neuroinflammations in Mashhad, Iran. Mashhad, situated in the Northeast of Iran and serving as the capital of Razavi Khorasan province, has a population of approximately 3.318 million people, making it the second most populous city in Iran [16]. In addition to these main referral centers, Mashhad houses various specialized hospitals attending to specific patient groups such as children, psychiatric patients, rehabilitation patients, eye patients, trauma cases, among others. The private hospital sector primarily offers elective care or specialized services for specific patient demographics. Demographic information, clinical features, laboratory and neuroimaging characteristics, and clinical outcomes of adult patients (≥ 15 years) with clinical syndromes suspected of community-acquired CNS infections were collected. A brain abscess was defined as a localized abnormality observed on CT or MRI scans, situated within the cerebrum, cerebellum, or midbrain, and meeting one or more of the following criteria: Positive culture results from intracerebral material. Positive culture results from intracerebral material. Presence of at least two positive blood cultures; Presence of at least two positive blood cultures; Confirmation of a brain abscess through histological examination; Confirmation of a brain abscess through histological examination; Neuroimaging findings suggestive of a brain abscess, accompanied by clinical and radiological improvement following antimicrobial therapy [6]. Neuroimaging findings suggestive of a brain abscess, accompanied by clinical and radiological improvement following antimicrobial therapy [6]. Samples obtained from cerebral lesions underwent standard microbiological and histological analysis. Statistics Continuous data were described with medians and interquartile range and categorical variables with frequency and percentage. The Mann Whitney U test was used for continuous variables and Fisher’s exact test and Chi-square tests were used for categorical variables, as appropriate. Significant predictors among the tested independent variables analyzed by using binary logistic regression test. A p-value < 0.05 was considered as statistically significant. Research ethics Informed consent was obtained from all participating patients or their legally authorized representatives. The ethics committee of Mashhad University of Medical Sciences approved this study with the code of IR.MUMS.REC.1400.189. Results Overall, 110 episodes of community-acquired infectious brain abscesses and/or granulomas were diagnosed in 106 patients. The median age was 45 (interquartile range [IQR], 30-56.3) years, and 69 (62.7%) were male. Thirteen (11.8%) episodes occurred in the elderly. The predisposing conditions were an immunocompromised state in 30 (27.5%) of 109 cases, preceding otitis/mastoiditis in 17 (16.2%) of 105, sinusitis in fourteen (13.3%), odontogenic infections in five (4.5%) of 110, heart disease (congenital or valvular heart disease, or endocarditis) in nine (8.1%) of 110, pulmonary infections in nineteen (17.2%) of 110, and meningitis in 11 (10%) of 110 cases. Twenty-nine (32.6%) were addicted including two IVDUs, 23 (21.1%) of 109 cases had diabetes mellitus, 10 (9.1%) chronic lung disease (COPD, asthma, bronchiectasis), 10 (9.1%) head trauma, seven (6.4%) cancer or rheumatologic disorders, 6 (5.5%) HIV/AIDS, 2 (1.8%) organ transplantation, and one CKD/ESRD. Twenty-nine (32.6%) were addicted including two IVDUs, 23 (21.1%) of 109 cases had diabetes mellitus, 10 (9.1%) chronic lung disease (COPD, asthma, bronchiectasis), 10 (9.1%) head trauma, seven (6.4%) cancer or rheumatologic disorders, six (5.5%) HIV/AIDS, two (1.8%) organ transplantation, and one CKD/ESRD. The most common clinical manifestations of encephalitis were a headache in 63 (57.3%) of 110 cases, fever (49.1%), altered consciousness in 48 (44.4%), arm or leg paresis in 37 (33.6%), seizures in 35 (31.8%), hallucination or disorganized speech in 15 (13.6%), visual problems in 11 (10%), papilledema in five (4.5%). Overall, in 20 (18.2%) episodes patients presented with the triad of fever, headache, and FNDs and in 36 (37.8%) with SIRS (Table 1). Table 1Patients’ characteristicsCharacteristicsaTotalAge, years45 (30-56.2)Elderly (≥ 65 years)13/110 (11.8)Gender, male69/110 (62.7)Predisposing conditions Immunocompromised state30/109 (27.5) Otitis/mastoiditis17/105 (16.2) Sinusitis14/105 (13.3) Odontogenic5/110 (4.5) Heart disease (congenitalb or valvular heart disease, or endocarditis)9/110 (8.1) Pulmonary infection19/62 (30.6) Meningitis11/110 (10) Other82/109 (75.2)Other underlying comorbidities Diabetes mellitus23/109 (21.1) Chronic lung disease (COPD, asthma, bronchiectasis)10/109 (9.1) Cancer7/109 (6.4) Rheumatologic disorders7/109 (6.4) HIV/AIDS6/110 (5.5) CKD/ESRD1/110 (0.9) Organ transplant2/109 (1.8) Head trauma10/109 (9.1) Addiction29/89 (32.6) IVDU2/110 (1.8)Duration of symptoms, days11 (3–30)Clinical manifestations Headache63/110 (57.3) Fever54/110 (49.1) Altered consciousness48/108 (44.4) Seizures35/110 (31.8) Hallucination or disorganized speech15/110 (13.6) Visual problems11/110 (10) Arm or leg paresis37/110 (33.6) FNDs67/110 (60.9) Papilledema5/110 (4.5) Triad of fever, headache, FNDs20/110 (18.2) SIRS36/95 (37.8)Length of hospital stay, days28 (16-46.5)In-hospital mortality27/110 (24.5)aNumbers are represented by the number of patients with specific characteristics/ the number of patients evaluated (percent) for nominal variables and by median (25th quartile, 75th quartile) for quantitative variablesbFour cases of Tetralogy of Fallot, two cases of Transposition of the Great Arteries, and two cases of Mitral Valve ReplacementCOPD: chronic obstructive pulmonary disease, HIV: human immunodeficiency syndrome, AIDS: acquired immunodeficiency syndrome, CKD: chronic kidney disease, ESRD: end-stage renal disease, IVDU: intravenous drug user, FND: focal neurologic deficit, SIRS: systemic inflammatory response syndrome Patients’ characteristics aNumbers are represented by the number of patients with specific characteristics/ the number of patients evaluated (percent) for nominal variables and by median (25th quartile, 75th quartile) for quantitative variables bFour cases of Tetralogy of Fallot, two cases of Transposition of the Great Arteries, and two cases of Mitral Valve Replacement COPD: chronic obstructive pulmonary disease, HIV: human immunodeficiency syndrome, AIDS: acquired immunodeficiency syndrome, CKD: chronic kidney disease, ESRD: end-stage renal disease, IVDU: intravenous drug user, FND: focal neurologic deficit, SIRS: systemic inflammatory response syndrome Thirty-three (30%) of cases underwent lumbar puncture (LP). CSF analysis revealed CSF pleocytosis in 21 (70%) of 31 episodes, elevated CSF protein in 22 (71%), and hypoglycorrhachia in ten (31.3%). Tests of inflammatory markers showed leukocytosis in 42 (42.4%) of 99 episodes, elevated ESR in 55 (63.9%) of eighty-six, and elevated CRP level in 32 (76.2%) of thirty-two. Twenty-three (23%) of 100 cases had hyponatremia (Table 2). Table 2Imaging and laboratory findingsTotalBrain CT scan76/110 (69.1)aBrain MRI97/110 (88.2)Neuroimaging findingsNumber of brain lesions* Single59/118 (50) Multiple59/118 (50)Abscess location Frontal49/103 (47.6) Temporal28/103 (27.2) Parietal44/103 (42.7) Occipital16/103 (15.5) Cerebellum22/103 (21.4) Brainstem12/103 (11.6) Basal ganglia13/103 (12.6) Extra-axial4/103 (3.8) Ring/rim enhancement97/103 (94.2) DWI restriction70/96 (72.9) Hydrocephalus or ventriculitis11/101 (10.9) Meningeal enhancement24/96 (25) Sinus opacification30/102 (29.4) Mastoid opacification24/102 (23.5)CSF investigation LP33/110 (30) Pleocytosis21/30 (70) Elevated CSF protein22/31 (71) Hypoglycorrhachia10/32 (31.3) Culture positive5/28 (17.8)Blood investigation Leukocytosis42/99 (42.4) Elevated CRP32/42 (76.2) Elevated ESR55/86 (63.9) Hyponatremia23/100 (23) Positive blood culture10/67 (14.9)Abscess culture10/38 (26.3)aNumbers are represented by the number of patients with specific characteristics/ the number of patients evaluated (percent) for nominal variables and by median (25th quartile, 75th quartile) for quantitative variables* The total number of episodes showing both single and multiple brain lesions differs from the total reported number due to missing data regarding the number of brain lesions in two casesCT: computed tomography, MRI: magnetic resonance imaging, DWI: Diffusion-Weighted Imaging, CSF: cerebrospinal fluid, LP: lumbar puncture, CRP: C-reactive protein, ESR: Erythrocyte Sedimentation Rate Imaging and laboratory findings aNumbers are represented by the number of patients with specific characteristics/ the number of patients evaluated (percent) for nominal variables and by median (25th quartile, 75th quartile) for quantitative variables * The total number of episodes showing both single and multiple brain lesions differs from the total reported number due to missing data regarding the number of brain lesions in two cases CT: computed tomography, MRI: magnetic resonance imaging, DWI: Diffusion-Weighted Imaging, CSF: cerebrospinal fluid, LP: lumbar puncture, CRP: C-reactive protein, ESR: Erythrocyte Sedimentation Rate The causative pathogens were isolated from abscess material in ten (9%) out of 38 cases, blood specimens in ten (9%), and CSF in five (4.5%). However, abscess culture, blood culture, and CSF culture were only performed in 38, 67, and 31 episodes, respectively. Finally, in 34 (31.8%) episodes, a cause-specific diagnosis was made, but in others, the etiology remained unknown or presumptive. The causative pathogens of brain abscesses and infectious granulomas included Mycobacterium tuberculosis (Fig. 1), Escherichia coli, and Mucorales (Fig. 2), each in four (3.6%), and Staphylococcus aureus (Fig. 3) and Pseudomonas species (Fig. 4), each in three (2.7%) cases. Additionally, Acinetobacter species, Candida species, Aspergillus species (Fig. 5), and unspecified fungal agents, along with Toxoplasma gondii (Fig. 6), each accounted for 2 (1.8%) episodes. Furthermore, Streptococcus viridans, Enterobacter species, coagulase-negative staphylococci, Enterococci, Corynebacterium species, Brucella species, and Nocardia species (Fig. 7) were each identified as the causative pathogen in one case (Table 3). A 19-year-old previously healthy female presented with headache, double vision, and ataxia. Brain MRI showed regions with low signal intensity on T1-weighted images (Image B), high signal intensity on FLAIR images (Image C), no restriction on DWI images (Image A), and ring enhancement on contrast-enhanced images (Image D), leading to the diagnosis of cerebral tuberculoma A 45-year-old female with diabetes mellitus presented with eye pain and swelling in her right eye, headache, and vomiting after contracting COVID-19 and undergoing systemic steroid therapy. This led to the diagnosis of rhino-orbital mucormycosis. MRI imaging of the brain revealed a region of low signal intensity on T1-weighted sequences (Image B), increased signal intensity on FLAIR sequences (Image C), along with restricted diffusion on DWI sequences (Image A), and ring enhancement on contrast-enhanced images (Image D) in the left frontal lobe, confirmed to be mucormycosis brain abscess A 40-year-old woman with a history of Tetralogy of Fallot developed confusion following a phlebotomy procedure. Brain MRI showed a lesion in the right hemisphere characterized by low signal intensity on T1-weighted images (Image B), high signal intensity on FLAIR images (Image C), restricted diffusion on DWI images (Image A), and ring enhancement on contrast-enhanced images (Image D). She was diagnosed with a brain abscess caused by Staphylococcus aureus a 50-year-old female who has a medical history of diabetes mellitus, recent COVID-19, and stroke. She presented with ataxia, and Brain MRI showed lesions in the left parietal lobe, characterized by low signal intensity on T1-weighted sequences (Image C), high signal intensity on FLAIR sequences (Image B), restricted diffusion on DWI sequences (Image A), and ring enhancement on contrast-enhanced images (Image D). Finally, the diagnosis of post-stroke pseudomonas brain abscesses was made A 45-year-old male diagnosed with recently diagnosed acquired immunodeficiency syndrome (AIDS) presented with confusion and left hemiparesis. MRI of the brain showed a lesion with low signal intensity on T1-weighted sequences (Image B), high signal intensity on FLAIR sequences (Image C), restricted diffusion on DWI sequences (Image A), and ring enhancement on contrast-enhanced images (Image D). Histopathological examination of the brain tissue confirmed the diagnosis of a brain abscess caused by Aspergillus species A 62-year-old female presented with headache, behavioral abnormalities, ataxia and left upper limb paresis. Brain MRI revealed numerous scattered lesions characterized by low signal intensity on T1-weighted sequences, high signal intensity on FLAIR sequences, restricted diffusion on DWI sequences, and ring enhancement on contrast-enhanced images (eccentric target sign). Following evaluation, the patient was diagnosed with advanced HIV (CD4: 9/µL) and cerebral toxoplasmosis A 59-year-old woman, who had previously undergone a kidney transplant, presented with symptoms of seizures and headaches. Brain MRI showed periventricular and cerebellar lesions with low signal intensity on T1-weighted images (Images B, F), high signal intensity on FLAIR images (Images C, G), along with restricted diffusion on DWI images (Images A, E), and ring enhancement on contrast-enhanced images (Images D, H). The patient was diagnosed with a brain abscess due to Nocardia species Table 3Causative pathogens of brain abscesses and granulomas and their predisposing factors and associated comorbiditiesCausative pathogenUnderlying conditionsNumber of episodes Escherichia coli COM/mastoiditis, DM4 Mycobacterium tuberculosis 4MucoralsDM, COVID, ESRD4 Staphylococcus aureus Post-phlebotomy in TOF, COM3Pseudomonas speciesCOM/mastoiditis, DM, COVID3Acinetobacter speciesBronchiectasis2Aspergillus speciesHIV2Candida speciesIVDU, DM2 Toxoplasma gondii HIV, ATLL2FungalCOM/mastoiditis2 Streptococcus viridans Hypertension1Enterobacter speciesHead trauma1EnterococciCOM1Nocardia speciesKidney transplant1Brucella species1Corinebacterium speciesAsthma1 Staphylococcus hemoliticus HCV (liver cirrhosis), MVR1DM: diabetes mellitus, ESRD: end-stage renal disease, TOF: Tetralogy of Fallot, HIV: Human immunodeficiency virus, IVDU: intravenous drug user, ATLL: adult T-cell leukemia/lymphoma, HCV: hepatitis C virus, MVR: mitral valve replacement Causative pathogens of brain abscesses and granulomas and their predisposing factors and associated comorbidities DM: diabetes mellitus, ESRD: end-stage renal disease, TOF: Tetralogy of Fallot, HIV: Human immunodeficiency virus, IVDU: intravenous drug user, ATLL: adult T-cell leukemia/lymphoma, HCV: hepatitis C virus, MVR: mitral valve replacement When comparing the two periods (2012–2016 and 2017–2021) based on the categories of gram-positive bacteria, gram-negative rods, fungal infections, and tuberculosis (TB), notable differences in the spectrum of pathogens can be observed, along with the overall percentages for each category. In the period from 2012 to 2016, the overall frequency of gram-positive bacteria was 40% (4 cases of Staphylococcus aureus, 1 case of Streptococcus viridans, 1 case of Enterococcus, and 1 case of Staphylococcus hemolyticus). Gram-negative rods accounted for 26.7% (4 cases of Escherichia coli and 1 case of Acinetobacter). Fungal infections represented 20% (1 case of Candida, 1 case of Aspergillus, and 2 unspecified fungal agents). Tuberculosis was identified in 6.7% of cases (1 case). In contrast, in the later period from 2017 to 2021, the overall frequency of gram-positive bacteria was 30% (1 case of Staphylococcus epidermidis, 1 case of Group B Streptococcus, and 2 cases of Staphylococcus aureus). Gram-negative rods accounted for 35% (3 cases of Pseudomonas, 1 case of Escherichia coli, 1 case of Acinetobacter, and 1 case of Enterobacter). Fungal infections were more prevalent at 30% (4 cases of Mucormycosis, 1 case of Aspergillus, and 1 case of Candida). Tuberculosis was also present, accounting for 10% of cases (2 cases). Brain CT scans were conducted in 76 (69.1%) cases, and brain MRIs were performed in 97 (88.2%) cases. The neuroimaging results showed that brain lesions were solitary in 56 (51%) episodes, while they were multiple in 52 (48%) episodes. Additionally, neuroimaging findings included ring/rim enhancement in 97 (94.2%) out of 103 episodes, DWI restriction in 70 (72.9%) out of 96, hydrocephalus or ventriculitis in 11 (10.9%) out of 101, and meningeal enhancement in 24 (25%) out of 96 episodes. Cases lacking DWI restriction included three cases of tuberculoma/TB abscess, two cases of staphylococcal abscesses, and abscesses caused by Acinetobacter species, Enterococci, Enterobacter species, and one unspecified fungal agent, each in one. In other episodes, the causative pathogen was not determined. In 49 (47.6%) of 103 episodes, abscesses were located in the frontal lobes, in 44 (42.7%) in the parietal lobes, in 28 (27.2%) in the temporal lobes, in 22 (21.4%) in the cerebellum, in 16 (15.5%) in the occipital lobes, in 13 (12.6%) in the basal ganglia, in 12 (11.6%) in the brainstem, and in four (3.8%) cases, they were accompanied by extra-axial empyema. Head imaging revealed paranasal sinus opacification in 30 (29.4%) of 102 cases and mastoid opacification in 24 (23.5%) (Table 2). In our cohort, 23 patients (21%) had diabetes mellitus, and 19 of 22 diabetic patients (86%) presented with multiple abscesses. Among these, 10 patients (45%) had five or more abscesses, two (9%) had two abscesses, and another 10 (45%) had a single abscess. The most common locations of abscesses in diabetic patients were the frontal lobe (45%), parietal lobe (45%), and temporal lobe (32%), with additional abscesses in the cerebellum (18%) and brainstem (5%). Among the patients with mucormycosis abscesses, one case (50%) was located in the temporal lobe and the other (50%) in the frontal lobe. Both abscesses were associated with preceding COVID-19 infection and diabetes mellitus. Among the three patients (3%) with infective endocarditis, the number of abscesses varied significantly: one patient (33%) had a single parietal abscess, one (33%) had three abscesses located in the cerebellum, parietal, and temporal lobes, and one (33%) had five or more abscesses affecting the frontal and temporal lobes. These findings align with the embolic nature of brain abscesses in patients with cardiac or valvular disease. Furthermore, among patients with sinusitis as the preceding infection, 12 of 18 (67%) had frontal abscesses, with a total of 5 (28%) temporal abscesses, four (22%) parietal, and three (17%) cerebellar abscesses; additionally, seven (39%) patients had multiple abscesses. In patients with otitis as the preceding infection, nine of 19 (47%) had abscesses involving the temporal lobe, while the distribution included four (21%) in the frontal lobe, five (26%) in the cerebellum, two (11%) in the parietal lobe, one (5%) in the brainstem, and seven (37%) had multiple abscesses. In 51 (46.4%) of episodes, in addition to antimicrobial therapy, patients underwent surgical intervention including thirty-six (32.7%) cases of aspiration, nine (8.1%) excision, and four (3.6%) stereotactic biopsies. Eight (7.2%) required more than one episode of surgical intervention. The median time between hospital admission and surgical intervention was three (IQR, 1.5-8) days. Fifty-nine (53.6%) only receive antimicrobials without surgical intervention (Table 4). Table 4Treatment modalities and complications during hospital admissionTotalTreatment Operation51/110 (46.4)a Aspiration36/110 (32.7) Excision9/110 (8.1) Stereotactic operation4/110 (3.6) Reoperation8/110 (7.2) Time from hospital admission to surgery3 (1-8.5) Medical treatment59/110 (53.6)Major complications DVT/PTE14/110 (12.7) Acute kidney injury6/110 (5.4) GIB6/110 (5.4) Respiratory distress6/110 (5.4)aNumbers are represented by the number of patients with specific characteristics/ the number of patients evaluated (percent) for nominal variables and by median (25th quartile, 75th quartile) for quantitative variablesDVT = Deep vein thrombosis, PTE = Pulmonary thromboemboli, GIB = Gastrointestinal bleeding Treatment modalities and complications during hospital admission aNumbers are represented by the number of patients with specific characteristics/ the number of patients evaluated (percent) for nominal variables and by median (25th quartile, 75th quartile) for quantitative variables DVT = Deep vein thrombosis, PTE = Pulmonary thromboemboli, GIB = Gastrointestinal bleeding Overall, 27 (24.5%) cases succumbed to death. The analysis revealed significant associations between certain factors and in-hospital mortality among patients with brain abscesses. Elderly patients exhibited notably higher mortality (61.5%) compared to non-elderly counterparts (19.6%), with a P-value of 0.003, an odds ratio (OR) of 6.568, and a 95% confidence interval (CI) of 1.930-22.353. Similarly, altered consciousness (P = 0.010, OR = 3.754, 95% CI: 1.422–9.908), multiple brain lesions (P = 0.023, OR = 3.176, 95% CI: 1.239–8.145), cerebellum involvement (P = 0.044, OR = 3.046, 95% CI: 1.100-8.432), and brainstem involvement (P = 0.030, OR = 4.934, 95% CI: 1.207–20.167) were significantly associated with increased mortality rates. However, the multiple variable analysis revealed that only age maintained a statistically significant association with in-hospital mortality among patients with brain abscesses. The elderly group had an odds ratio (OR) of 6.05 (95% CI: 0.52–8.008) with a p-value of 0.001, indicating a statistically significant association with the outcome. Other factors such as altered consciousness, multiple brain lesions, cerebellum involvement, and brainstem involvement did not independently predict mortality after adjusting for potential confounders (Table 5). Table 5Univariate and multiple variable logistic regression analysis of prognostic factors for in-hospital mortalityUnivariateMultiple variableDiedP-valueOR95% CIP-valueOR95% CIElderlyaYes8/13 (61.5)0.0036.5681.930-22.3530.00112.6052.949–54.145No19/97 (19.6)Altered level of consciousnessbYes13/30 (43.3)0.0103.7541.422–9.9080.1912.1580.716–7.769No11/65 (16.9)Multiple brain lesionsYes18/52 (34.6)0.0233.1761.239–8.1450.1832.3340.710–8.375No8/56 (14.3)Cerebellum involvementYes9/22 (40.9)0.0443.0461.100-8.4320.4511.6770.330–5.884No15/81 (18.5)Brainstem involvementYes5/9 (55.6)0.0304.9341.207–20.1670.1094.2040.910-50.441No19/94 (20.2)aAge ≥ 65 yearsb GCS < 14 on admissionFIGURE Univariate and multiple variable logistic regression analysis of prognostic factors for in-hospital mortality aAge ≥ 65 years b GCS < 14 on admission FIGURE The median length of hospital stay (LOS) was 28 (IQR, 16-46.5) days. Major complications occurred during hospital admission in patients with brain abscess and infectious granulomas including deep vein thrombosis/pulmonary thromboembolism (12.7%), acute kidney injury (5.4%), gastrointestinal bleeding (5.4%), and respiratory distress (5.4%). Discussion Our study underscored the confusing clinical presentation of brain abscesses and infectious granulomas, its predisposing factors, alongside the difficulties in diagnosing and treating these conditions. Additionally, our study revealed the substantial mortality rate associated with these conditions. We observed that brain abscesses and infectious granulomas commonly affect adults in their mid-forties, although elderly individuals are also susceptible. Notably, a significant portion of patients (approximately 95%) had underlying conditions and associated comorbidities that predisposed them to these infections, such as immunocompromised states, otitis/mastoiditis, sinusitis, and pulmonary infections. Other comorbidities like diabetes mellitus, chronic lung disease, and addiction were also prevalent among the patients. Clinical presentations of brain abscesses and infectious granulomas varied, with headache, fever, altered consciousness, and focal neurological deficits being the most common symptoms. It is worth noting, in line with earlier studies [17, 18], that only a minority of cases (18%) displayed the traditional triad of fever, headache, and focal neurological deficits. The scarcity of the classic triad of fever, headache, and focal neurological deficits poses a significant challenge in early diagnosis of brain abscesses and infectious granulomas, potentially delaying treatment initiation. Physicians need to maintain a heightened suspicion for these conditions, especially in patients with predisposing factors like immunocompromised states or underlying infections, even without the classic triad, to ensure early recognition and optimal patient outcomes. Lumbar puncture is relatively contraindicated in patients with brain abscesses; however, one-third of patients in our study underwent lumbar puncture. This procedure revealed pleocytosis and elevated protein levels in cerebrospinal fluid (CSF) samples in 70% and 71% of those who underwent lumbar puncture, respectively, albeit with a low yield of microbiological confirmation (18%). The findings of our study underscore significant challenges associated with lumbar puncture in patients with brain abscesses. While our results revealed pleocytosis and elevated protein levels in the CSF for a substantial proportion of patients, the low yield of microbiological confirmation (18%) reflects a critical limitation in the diagnostic utility of this procedure. This aligns with existing literature, which has consistently demonstrated a low rate of pathogen identification in cases of brain abscesses, as illustrated by Brouwer et al., where only 24% of CSF cultures yielded positive results [4]. Given the inherent risks associated with lumbar puncture in patients with brain abscesses, including the potential for clinical deterioration, a cautious approach should be advocated when considering this diagnostic procedure. Although we did not assess the rate of clinical deterioration after lumbar puncture in our study, the literature reports a 7% incidence of clinical worsening attributed to this procedure [4]. Therefore, it is imperative to prioritize patient safety by utilizing imaging techniques, such as cranial CT or MRI, to assess for brain shift prior to the procedure. The identification of brain shift is crucial, as it may indicate an increased risk of complications related to lumbar puncture. The low proportion of patients in our study who had microbiological confirmation of the causative agent of their brain abscesses or granulomas is also a notable finding that warrants discussion. Only a minority of cases (about 26% for abscess culture and 15% for blood culture) yielded positive results, indicating a considerable diagnostic challenge in identifying the specific pathogens responsible for these infections. The low rate of microbiological confirmation in brain abscesses and infectious granulomas in our study could be influenced by various factors, including culture-negative infections, empirical antibiotic treatment before diagnosis, challenges in obtaining specimens for culture, and limitations in microbiological techniques and laboratory resources [19]. In low- and middle-income countries (LMICs), limited access to healthcare services, diagnostic challenges, antimicrobial therapy prior to diagnosis, challenges in surgical care, and socioeconomic factors exacerbate the issue. Addressing these challenges requires healthcare system strengthening, capacity building, and resource allocation to enhance diagnostic capabilities and surgical services. The findings of our study regarding the causative pathogens of brain abscesses and infectious granulomas reflect a significant deviation from previous reports, suggesting a possible shifting microbial landscape in this clinical entity. In previous studies, oral cavity bacteria, particularly streptococci, were reported as the most frequent causative pathogens for brain abscesses, followed by less frequently reported Staphylococcus aureus and gram-negative bacteria [4]. Additionally, immunocompromised individuals were commonly associated with infections caused by Nocardia, Aspergillus, Listeria, and Toxoplasma [3]. Notably, a single study by Garg et al. in India in 2007 on patients with multiple brain lesions reported Mycobacterium tuberculosis as the most common causative pathogen, followed by neurocysticercosis [20]. However, in our study, we observed a different pattern of causative pathogens. While the yield of microbiological investigations in our study was relatively low, we identified notable pathogens. M. tuberculosis, Escherichia coli, and Mucorales were each isolated in four cases, indicating a diverse range of microbial agents. Moreover, gram-negative bacteria (GNB) were more prevalent than gram-positive bacteria (GPB), contrasting with previous findings. To further illustrate this evolution, we compared the causative pathogens across two time periods: from 2012 to 2016 and from 2017 to 2021. In the earlier period, GPB constituted 40% of cases (including Staphylococcus aureus, Streptococcus viridans, and Enterococcal species), while GNB accounted for 26.7% (with cases of E. coli and Acinetobacter species). In contrast, during the later period, GPB decreased to 30%, and GNB increased to 35%. The prevalence of fungal infections rose from 20 to 30%, emphasizing a notable shift in pathogen distribution over time. On a global scale, the majority of brain abscesses have been linked to GPB, with streptococci (aerobic, anaerobic, and microaerophilic) responsible for 30–60% and staphylococci for 10–20% of cases. In contrast, enteric GNB have accounted for 23–33% of cases [21]. The shift in our study might mirror alterations in patient demographics or underlying conditions. Earlier research indicated that patients with GNB-induced brain abscesses often have a history of head and neck surgery and hardware placement, unlike those with GPB-induced abscesses. Moreover, GNB-induced abscesses tend to be more healthcare-associated and nosocomial in onset compared to GPB-induced abscesses, which are predominantly community-associated [21]. However, our study only included cases of community-acquired brain abscesses and infectious granulomas. Additionally, it’s imperative to consider the impact of the COVID-19 pandemic, which has correlated with an elevated incidence of GNB infections [22] and fungal infections, notably mucormycosis [23]. Part of our study overlapped with COVID-19 pandemic. This evolving microbial profile in our study underscores the necessity for continual surveillance and adaptation of treatment strategies to address emerging pathogens and combat antimicrobial resistance. Neuroimaging plays a crucial role in diagnosis of brain abscesses, with brain MRI being the preferred modality due to its high sensitivity. In our study, abscesses were often located in the frontal and parietal lobes, and they typically exhibited ring/rim enhancement and diffusion-weighted imaging (DWI) restriction. In the European Society of Clinical Microbiology and Infectious Diseases guidelines on the diagnosis and treatment of post-surgical brain abscesses published in January 2024, 28 studies involving 2128 patients were reviewed. The study reported a sensitivity of 92% and specificity of 91% for brain MRI with contrast injection and DWI and ADC sequences in diagnosing brain abscesses, strongly recommending its use for suspected cases. However, MRI may exhibit lower sensitivity in diagnosing brain abscesses arising from post-surgical conditions, those caused by toxoplasmosis, and after antibiotic use [3]. Notably, our study revealed DWI restriction in only 73% of episodes, with cases lacking DWI restriction including three cases of tuberculoma/TB abscess, two cases of staphylococcal abscesses, and abscesses attributed to various pathogens, including Acinetobacter species, Enterococci, and Enterobacter species, as well as one unspecified fungal agent, each in one episode. In other cases, the causative pathogen remained undetermined. We acknowledge that the interpretation of DWI findings in our study is limited due to the low rate of confirmed diagnoses (only 32%). It is possible that some cases without DWI restriction might not have been abscesses but rather other non-infectious conditions. Therefore, while our findings suggest a 73% sensitivity for DWI in identifying brain abscesses, we cannot comment on its specificity, as the number of DWI-positive cases that might represent non-abscess lesions remains unknown. Previous studies highlighted that among patients with ring-enhancing brain lesions, the presence of restriction on DWI sequences may be characteristic for brain abscesses, although it is not pathognomonic and the sensitivity is not 100% [24]. Additional neuroimaging findings observed in our study included hydrocephalus or ventriculitis and meningeal enhancement, which probably indicated complicated brain abscesses [24]. In our study, 21% of patients had diabetes mellitus, with 86% of these diabetic patients presenting with multiple abscesses. Notably, 45% had five or more abscesses, while 45% had a single abscess, primarily located in the frontal and parietal lobes, followed by the temporal lobe. Among patients with mucormycosis abscesses, lesions were found in both the temporal and frontal lobes, with both cases linked to preceding COVID-19 infections and diabetes. In the cohort with infective endocarditis, abscesses were distributed across the parietal, frontal, temporal, and cerebellar lobes, with a significant proportion exhibiting multiple abscesses. Additionally, patients with sinusitis showed abscesses primarily in the frontal lobe, while those with otitis had a higher tendency for temporal lobe brain abscesses. These findings underscore the relationship between specific comorbidities and patterns of brain abscess formation, revealing a complex interplay in their distribution. Treatment strategies included a combination of surgical interventions such as aspiration, excision, or stereotactic operation, along with medical therapy. However, despite treatment, a considerable proportion of patients succumbed to death, with the in-hospital mortality rate of approximately 24.5%. This high mortality underscores the critical nature of these infections and emphasizes the urgent need for effective therapeutic interventions and early diagnosis. In a systemic review which analyzed 9699 episodes of brain abscesses (1935–2012), the mortality rates varied across continents, with reports of 22% in Europe, 31% in America, 25% in Australia, 19% in Africa, and 15% in Asia. Generally, a mortality rate of 20% was reported, showing a decreasing trend over time [4]. The univariate logistic regression analysis revealed several significant predictors of mortality in patients with brain abscesses and infectious granulomas. Elderly patients (age ≥ 65 years) had a significantly elevated risk of death compared to younger individuals, exhibiting a 6.5-fold increase in mortality risk (p = 0.003). Similarly, patients presenting with altered level of consciousness were more likely to die, showing a 3.7-fold increase in mortality risk (p = 0.010). Presence of multiple brain lesions, cerebellum involvement, and brainstem involvement were also associated with heightened mortality risk, with respective 3.2 (p = 0.023), 3.0 (p = 0.044), and 4.9 (p = 0.030) times higher odds of mortality. However, in the multiple variable logistic regression analysis, only age remained a significant contributor to death, demonstrating a 12.6-fold increase in mortality risk (p = 0.001). These findings underscore the critical role of age as a determinant of mortality in patients with brain abscesses and infectious granulomas, highlighting the necessity for targeted interventions and closer monitoring of elderly individuals to improve clinical outcomes. Most previous studies have indicated lower mortality rates among patients who underwent surgical intervention compared to those treated solely with medical therapy [3, 10, 17, 25]. In our study, we observed no discrepancy in mortality rates between patients who underwent surgery and those treated solely with medical intervention, possibly due to delays in surgical procedures. The average time from hospital admission to surgical intervention in our study was three days, contrasting with the one-day duration reported in the Campioli study [26], for instance. Additionally, the lower frequency of excision surgeries in our research, compared to previous studies [4, 10, 17], might impact mortality rates and the need for subsequent surgical procedures. Other contributing factors could include the higher frequency of surgical interventions in critically ill patients or those with underlying conditions. However, our study did not demonstrate such an association. Additionally, prognostic factors reported in previous studies include extremes of age [5, 27], female gender [18], immunocompromised conditions [5, 17], presence of underlying comorbidities [18], Glasgow Coma Scale (GCS) on admission [6, 18, 26, 27], septic shock [10], oral cavity bacteria [5] and Gram-negative bacteria [15, 18] as causative pathogens, fungal etiology [27], abscess diameter > 3 cm [5], multiple lesions, deep-seated lesions [27], intraventricular rupture [5, 27], delayed or missed diagnosis [27]. Understanding and addressing these prognostic factors are essential for risk stratification and optimizing treatment strategies to improve patient outcomes. Our study had several limitations. Firstly, it was conducted in Mashhad, Iran, where unique healthcare infrastructure, patient demographics, and disease epidemiology may exist. Thus, caution should be taken when extrapolating findings to other geographical regions or healthcare systems due to potential contextual differences. Secondly, a key limitation of our study is the low culture positivity rate, which hampers our ability to accurately identify specific pathogens and limits the generalizability of our findings regarding the microbial landscape. Although we observed a trend towards an increase in gram-negative bacteria, this should be interpreted cautiously due to the small number of cases with definitive pathogen identification. The presence of culture-negative infections may introduce bias, indicating that the observed microbial profile could reflect limitations in our diagnostic methods rather than an actual shift in pathogen prevalence. Therefore, while our results suggest a possible evolving microbial landscape, further research utilizing improved diagnostic techniques is needed to validate these trends. Thirdly, although the study recognized the presence of underlying conditions predisposing patients to infectious brain abscesses and granulomas, the extent to which these comorbidities influenced outcomes was not fully explored. Certain comorbidities may independently affect mortality and morbidity rates, thereby complicating the interpretation of prognostic factors. Lastly, the study did not consider variations in antimicrobial regimens, which could impact patient outcomes. Differences in treatment approaches across providers or centers might confound the observed associations between prognostic factors and clinical outcomes. In conclusion, our study highlights the challenges in diagnosing and treating brain abscesses and infectious granulomas, which commonly affect middle-aged and elderly adults with underlying health conditions. These infections often present with varied symptoms, making early diagnosis difficult. Despite advanced neuroimaging and surgical techniques, the mortality rate remains high at 24.5%, with age, altered consciousness, and multiple or brainstem lesions being significant predictors of death. The changing microbial landscape, with a notable prevalence of gram-negative bacteria, underscores the need for continuous surveillance and adaptation of treatment strategies. Limited microbiological confirmation, especially in low- and middle-income countries, further complicates diagnosis and treatment. Improving diagnostic capabilities, timely surgical intervention, and addressing socioeconomic disparities are essential for better outcomes. Future research should focus on optimizing treatment protocols and understanding the impact of comorbidities to reduce the high morbidity and mortality associated with these infections. Acknowledgements This research was conducted with support from the vice-chancellery for research of Mashhad University of Medical Sciences (Research Project Number 4000604 and ethic code IR.MUMS.REC.1400.189) and is part of the thesis of the first author. Author contributions Z.H., M.H., F.K. and F.S. conceived of the presented idea. Z.H., M.H., F.K. and F.S. wrote the manuscript with support from E.K. and N.M. Z.H., M.H., F.K. and F.S. contributed to the design and implementation of the research, F.K., E.K and N.M. contributed to the writing of the manuscript. F.S. and F.K. prepared Figs. 1, 2, 3, 4 and 5 and tables with support from E.K., Z.H and M.H. All authors reviewed the manuscript. Funding None. Data availability The data that support the findings of this study are available from the corresponding author upon reasonable request. Declarations Ethics approval and consent to participate Informed consent was obtained from all participating patients or their legally authorized representatives. The ethics committee of Mashhad University of Medical Sciences approved this study with the code of IR.MUMS.REC.1400.189. Consent for publication Informed consent was obtained from all participating patients or their legally authorized representatives. Clinical trial number N/A. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request.