Focused Update to International Consensus Recommendations for the Use of Prolonged‐Infusion Beta‐Lactam Antibiotics Among Severely Ill Adult Patients: Endorsed by the American College of Clinical Pharmacy, British Society for Antimicrobial Chemotherapy, Cystic Fibrosis Foundation, European Society of Clinical Microbiology and Infectious Diseases, Infectious Diseases Society of America, Society of Critical Care Medicine, and Society of Infectious Diseases Pharmacists Authors: Lisa T Hong, Alireza FakhriRavari, Marta Ulldemolins, Jason A Roberts, Robert A Bonomo, Marc H Scheetz, the Prolonged‐Infusion Beta‐Lactam Consensus Panel Published: 2026 ABSTRACT ABSTRACTIn 2023, international consensus recommendations suggested the use of prolonged‐infusion beta‐lactam antibiotics over short‐infusion dosing in severely ill adult patients to improve mortality or clinical cure, although the certainty of evidence was very low. With the availability of several new randomized controlled trials (RCTs), this focused update re‐evaluates the efficacy of prolonged versus short infusion among severely ill adult patients. The original Population, Intervention, Comparator, and Outcome (PICO) question VII was separated into two questions addressing mortality (VIIa) and clinical cure (VIIb). Methods used for study identification, screening, and evidence grading were consistent with the original guideline, with the addition of Bayesian meta‐analyses. A total of 28 RCTs evaluating mortality in severely ill adults were analyzed, with pooled estimates favoring prolonged infusion (RR 0.91; 95% confidence interval 0.85 to 0.97), supported by Bayesian analyses demonstrating a 98.76% posterior probability of benefit. Across 21 RCTs evaluating clinical cure, prolonged infusion improved outcomes compared with short infusion (RR 1.11; 95% confidence interval 1.06 to 1.17), supported by Bayesian analyses demonstrating a 99.97% posterior probability of benefit. Certainty of evidence was strengthened to moderate for mortality and low for clinical cure. Consistent with the original recommendation (PICO X), subgroup analyses evaluating loading doses suggested improved outcomes when a loading dose is used before continuous infusion. Overall, the cumulative evidence reinforces and strengthens the recommendation for use of prolonged‐infusion beta‐lactam therapy in severely ill adult patients to improve survival and clinical cure. Continued use of loading doses when initiating continuous‐infusion therapy is suggested. Future research should identify subpopulations most likely to benefit, clarify optimal prolonged‐infusion strategies, and define the role of therapeutic drug monitoring.Keywords: beta‐lactams, consensus, infectious diseases, prolonged infusion Introduction In 2023, we published the international consensus recommendations for the use of prolonged‐infusion beta‐lactam (β‐lactam) antibiotics [1]. In those consensus recommendations, Population, Intervention, Comparator, and Outcome (PICO) question VII asked, “Should prolonged‐infusion (PI) β‐lactam antibiotics be preferred over short infusion (SI) dosing in severely ill adult patients to improve mortality or clinical cure?” Based on the reviewed evidence at the time, we suggested PI β‐lactam antibiotics over SI to reduce mortality or increase clinical cure among severely ill adult patients, particularly those with Gram‐negative infections (including Enterobacterales and non‐fermenting Gram‐negative bacilli). This was a conditional recommendation, based on very low certainty of evidence with the panel vote 17–0 in favor of the recommendation. As this was an important recommendation from the consensus document and a number of recent prospective studies are now available to inform the question, we have opted to repeat our original analysis [1] and summarize the updated results. Methods Consensus Recommendation Panel The author panel aimed to update recommendations associated with clinically important efficacy outcomes (mortality and clinical cure) in severely ill adult patients. Microbiological cure was summarized for completeness, consistent with the methodology of the 2023 consensus. Definitions In line with our original document [1], PI were defined as continuous infusion (CI) or extended infusion (EI) over at least 3 h and short infusions (SI) as infusions up to 60 min. PICO Question Revision For this update, PICO VII has been separated to address mortality (VIIa) and clinical cure (VIIb). Literature Search, Screening Process, and Grading of Evidence The literature search strategy was conducted as previously described [1] with dates between October 18, 2020, and January 26, 2025, which generated 517 articles for screening within Rayyan [2]. The screening process required agreement by two authors and adjudication by a third author for any disagreements. The quality of evidence and strength of recommendation were evaluated using the Grading of Recommendations Assessment Development and Evaluation (GRADE) system [3]. Meta‐Analyses All randomized controlled trials (RCTs) were meta‐analyzed; observational studies were summarized narratively and were not meta‐analyzed as previously described [1]. Frequentist meta‐analyses were conducted in R (version 4.4.0) using the “meta” package. We used a random‐effects model for all primary analyses; however, for transparency, the figures present results from both random‐effects and fixed‐effect (common‐effect) models. Bayesian meta‐analyses were conducted using the “bayesmeta” package under the normal–normal hierarchical random‐effects model. Bayesian meta‐analyses were included as probabilistic summaries of treatment effect to complement frequentist estimates and were not intended to replace GRADE assessments of certainty of evidence. The prior for the overall treatment effect was centered at no effect (log[RR] = 0) with a standard deviation of 2, reflecting equipoise while allowing a broad range of plausible treatment effects and minimizing directional influence toward benefit or harm. Accordingly, posterior probabilities are conditional on the specified prior and model assumptions and should be interpreted alongside effect estimates and GRADE certainty ratings, rather than as direct measures of certainty of clinical benefit. Updated Recommendations PICO VIIa Should PI β‐lactam antibiotics be preferred over SI dosing in severely ill adult patients to improve mortality? Recommendation 7a We suggest PI β‐lactam antibiotics over SI to reduce mortality among severely ill adult patients, particularly those with Gram‐negative infections. (Conditional recommendation; moderate certainty of evidence). Evidence Summary Eight additional RCTs [4, 5, 6, 7, 8, 9, 10, 11] were included with the updated literature search and screening process, for studies including patients within intensive care units (ICUs), totaling 28 RCTs that evaluated mortality in these severely ill patients. None of these RCTs individually demonstrated a mortality benefit with prolonged compared with SI β‐lactam treatment. However, collectively, the mortality rate across all RCTs was lower with prolonged administration methods (RR 0.91 [95% confidence interval 0.85 to 0.97; I 2 = 0]) [4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31] (Figure 1). In Bayesian random‐effects meta‐analysis, the pooled RR was 0.89 (95% credible interval 0.77 to 0.99), with low heterogeneity (I 2 = 8%) and a posterior probability of benefit of 98.76%. CI was used in 19 studies and EI was used in the other 9 studies. Agents studied included piperacillin‐tazobactam [5, 8, 9, 10, 15, 22, 24, 25, 26, 27, 28, 31], meropenem [4, 6, 8, 9, 10, 22, 24, 26, 29, 30], cefepime [8, 11, 14, 26], ceftazidime [12], ticarcillin‐clavulanate [22, 24], imipenem‐cilastatin [9, 23], doripenem [21], ceftriaxone [16], amoxicillin‐clavulanate [9], and ampicillin‐sulbactam [7]. Meta‐analysis and funnel plot of RCTs evaluating mortality in severely ill adult patients. The RCT that contributed the most weight (i.e., patients) on the overall effect size in the meta‐analysis was the BLING III study, comparing PI versus SI β‐lactam antibiotics [10]. It was conducted in 104 adult ICUs in Australia, Belgium, France, Malaysia, New Zealand, Sweden, and the United Kingdom, and it randomized 7202 sepsis and septic shock patients with 3498 patients receiving CI and 3533 patients receiving SI. Patients receiving kidney replacement therapy were excluded. In the evaluable population, 79.6% of patients received piperacillin‐tazobactam and 20.4% received meropenem. The primary site of infection was pulmonary in 59.5% of patients and intraabdominal in 13% of patients. Only 40% had infective organisms identified, and 30% of those were Gram‐positive organisms. The study did not find a statistically significant difference in 90‐day mortality (odds ratio [OR] 0.91 [95% confidence interval 0.81 to 1.01], p = 0.08) in the primary analysis. The absolute difference in mortality was −1.9% [95% confidence interval −4.9% to 1.1%], in favor of CI. The study was designed to detect an absolute difference of 3.5%, but it was underpowered to detect smaller yet meaningful differences (such as the 1.9% observed). The MERCY trial was the second largest RCT, which randomized 607 patients with presumed sepsis in 31 ICUs across Croatia, Italy, Kazakhstan, and Russia to receive meropenem as a CI (n = 303) or SI (n = 304) [6]. The most common sites of infection were respiratory tract (33%), gastrointestinal tract (about 9%), and catheter‐related bloodstream (about 7%). About 30% of patients did not have a pathogen identified, about 36% had Gram‐positive organisms identified (50% of those being coagulase‐negative staphylococci), and about 32% had carbapenem‐resistant organisms. No difference in all‐cause mortality at 28 days was seen between groups (relative risk [RR] 0.92 [95% confidence interval 0.73 to 1.17], p = 0.50). With an absolute difference of −2.5% and the study being designed to detect a 12% absolute risk reduction, this study was also underpowered and enrolled a large number of patients that would not be expected to benefit from a dose optimization strategy. A third randomized trial, BEATLE [32], met our predefined criteria for severely ill patients based on mean APACHE II scores. However, the study enrolled predominantly ward hematological patients with febrile neutropenia, and the quantitative value of APACHE II scores in this setting is unclear as underlying hematologic malignancy and chemotherapy‐related effects may drive differences rather than severity of illness. Therefore, BEATLE was excluded from the primary analysis and included in a sensitivity analysis to maintain consistency with our severity definition. Mortality at 30 days occurred in 2/77 (2.5%) patients receiving EI and 1/73 (1.7%) patients receiving SI, contributing an overall weight of only 0.1% to the meta‐analysis. Inclusion of the BEATLE trial did not alter the pooled estimate (RR 0.91, 95% confidence interval 0.85 to 0.97 [Figure S1]). In Bayesian sensitivity analysis including BEATLE, the pooled RR remained 0.89 (95% credible interval 0.77 to 0.99), with I 2 = 7% and a posterior probability of benefit of 98.68%. With regard to observational data, the updated literature search identified 10 new studies evaluating the effects of prolonged antibiotic infusions on mortality in patient populations largely composed of severely ill individuals [33, 34, 35, 36, 37, 38, 39, 40, 41]. Overall, these studies were retrospective and included a variety of broad‐spectrum anti‐pseudomonal β‐lactam antibiotics, which were more frequently administered as extended infusions than as continuous infusions. Many of these studies used a pre–post quasi‐experimental design [33, 34, 38, 41]. Although not all patients were in the ICU (ranging from 10% to 100% across studies), most had high illness severity reflected by the high proportion of patients diagnosed with sepsis (40%–100%) [33, 34, 36, 40, 42] or septic shock (33%–100%) [33, 37, 40, 42]. Across the 16 identified observational studies, mortality was generally reduced in patients receiving PI [35, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. Among the studies that reported a significant protective effect with PI [35, 38, 40, 41, 42, 43, 44, 45, 46, 47, 48], a greatest benefit was observed for patients with confirmed Gram‐negative infections, including P. aeruginosa [35, 42, 43, 44, 45, 46, 48]. We also note a recently published systematic review and meta‐analysis [49], but that analysis was focused on adult patients with sepsis and septic shock. Limitations and Future Research Needs Mild funnel plot asymmetry was observed (Figure 1); however, no statistically significant publication bias was found, and pooled estimates were driven by the large, multicenter RCTs. Importantly, none of the included individual RCTs have demonstrated a mortality benefit and this raises concerns about the effect size of use of PI. Furthermore, most of the literature heralds from CI rather than EI dosing interventions which complicates the ability to infer the same level of effect across both infusion durations. Powering a study to detect the difference between CI and EI for all patients will likely be impossible as the study numbers to discern partial differences (e.g., 1% absolute difference in 90‐day mortality) will require upwards of 100,000 patients. Because of this, there are various important future research needs to be explored and these include: Post hoc analyses of existing RCT data to identify sub‐populations of severely ill adult patients who are most likely to have greater benefit from PI of β‐lactams, including infection syndromes, patient severity of illnesses, and less susceptible pathogens (e.g., with an increased minimum inhibitory concentration). These data can then be used to develop targeted future studies that have a greater ability to detect meaningful differences in outcomes. Post hoc analyses of existing RCT data to identify sub‐populations of severely ill adult patients who are most likely to have greater benefit from PI of β‐lactams, including infection syndromes, patient severity of illnesses, and less susceptible pathogens (e.g., with an increased minimum inhibitory concentration). These data can then be used to develop targeted future studies that have a greater ability to detect meaningful differences in outcomes. A powered RCT for CI vs. EI (and preferably as a 3‐arm study, also including SI), in adult and pediatric sub‐populations that focuses on surrogates for mortality (e.g., clinical cure). A powered RCT for CI vs. EI (and preferably as a 3‐arm study, also including SI), in adult and pediatric sub‐populations that focuses on surrogates for mortality (e.g., clinical cure). RCTs comparing PI combined with a therapeutic drug monitoring intervention with a standard PI protocol or versus SI. RCTs comparing PI combined with a therapeutic drug monitoring intervention with a standard PI protocol or versus SI. PK/PD studies comparing exposure of β‐lactams in both CI and EI scenarios with patient outcomes, in order to define specific exposures which could be subsequently tested in RCTs of maximal β‐lactam effectiveness. PK/PD studies comparing exposure of β‐lactams in both CI and EI scenarios with patient outcomes, in order to define specific exposures which could be subsequently tested in RCTs of maximal β‐lactam effectiveness. Studies focusing on individual molecules or specific β‐lactam subclasses, rather than treating the β‐lactam class as homogenous, to better elucidate whether important differences exist in terms of optimal exposure for each agent. Studies focusing on individual molecules or specific β‐lactam subclasses, rather than treating the β‐lactam class as homogenous, to better elucidate whether important differences exist in terms of optimal exposure for each agent. PICO VIIb Should PI β‐lactam antibiotics be preferred over SI dosing in severely ill adult patients to improve clinical cure? Recommendation 7b We suggest PI β‐lactam antibiotics over SI to increase clinical cure among severely ill adult patients, particularly those with Gram‐negative infections. (Conditional recommendation; low certainty of evidence). Evidence Summary The updated literature search identified eight additional RCTs that evaluated clinical cure in severely ill adult patients [4, 5, 7, 8, 9, 10, 11, 32]. As explained above, the BEATLE trial [32] was not included in our primary analysis. Thus, seven RCTs were added to the primary analysis, totaling 21 RCTs [4, 5, 7, 8, 9, 10, 11, 14, 16, 18, 19, 20, 21, 22, 24, 26, 27, 29, 30, 31, 50]. CI was used in 14 studies and EI was used in the other 7 studies. Agents studied included piperacillin‐tazobactam [5, 8, 9, 10, 22, 24, 26, 27, 31], meropenem [4, 8, 9, 10, 22, 24, 26, 29, 30], cefepime [8, 11, 14, 26], ceftazidime [50], ticarcillin‐clavulanate [22, 24], imipenem‐cilastatin [9], doripenem [21], ceftriaxone [16], amoxicillin‐clavulanate [9], and ampicillin‐sulbactam [7]. Again, the BLING III study contributed the most weight on the overall effect size and found significantly greater clinical cure with CI compared with SI (OR 1.26 [95% confidence interval 1.15 to 1.38], p < 0.001) [10]. Collectively, these 21 studies demonstrated significant benefit with the use of PI β‐lactam antibiotics compared to SI (RR 1.11 [95% confidence interval 1.06 to 1.17; I 2 = 12.6%]) (Figure 2). Sensitivity analysis including the BEATLE trial yielded similar results (Figure S2). In Bayesian random‐effects meta‐analysis, the pooled RR was 1.11 (95% credible interval 1.05 to 1.19; I 2 = 12.1%) with a posterior probability of benefit of 99.97%. Notably, the heterogeneity in clinical cure definitions across studies led to a downgrade for indirectness, and in the GRADE approach, any domain requiring downgrade contributes to a lower overall certainty of evidence, regardless of how compelling the effect size may appear. Consistent with the findings from intervention trials, most observational studies evaluating clinical cure have reported improved cure rates with PI [33, 40, 44, 50, 51, 52, 53]. Meta‐analysis and funnel plot of RCTs evaluating clinical cure in severely ill adult patients. Two additional RCTs that evaluated microbiological cure were identified in our updated literature search and both failed to find significant differences in microbiological cure between PI and SI β‐lactam therapy [8, 11]. However, overall, meta‐analysis of 8 RCTs found improved microbiological success among severely ill patients who received PI compared with SI of β‐lactams (RR 1.17 [95% confidence interval 1.06 to 1.30], I 2 = 3.3%) (Figure 3). Meta‐analysis and funnel plot of RCTs evaluating microbiological cure in severely ill adult patients. Limitations and Future Research Needs Key gaps exist in our knowledge of the outcome benefits associated with using PI of β‐lactam antibiotics over SI dosing in severely ill adult patients. Although there are a number of studies that individually demonstrate significantly improved clinical cure in severely ill adults receiving PI, some studies reported only a combined outcome of clinical cure and clinical improvement, limiting our ability to distinguish complete from partial resolution. Additionally, consistent data relating to microbiological cure or hospital length of stay are lacking. More consistent data collection of these endpoints is needed in future evaluations. Studies in pediatric patients are also sparse and should be a priority for future research. PICO X Should a loading dose be administered over no loading dose when using PI β‐lactam antibiotics in severely ill adults to improve mortality or clinical cure? Recommendation 10 When initiating CI β‐lactams in severely ill patients, we suggest administering a loading dose to improve clinical success; for EI, evidence is insufficient to recommend for or against a loading dose. (Conditional recommendation; very low certainty of evidence). Evidence Summary Loading doses are used at the start of β‐lactam therapy to rapidly achieve therapeutic levels and prevent delays in target attainment. Of the 28 RCTs evaluating mortality in severely ill patients, 18 studies utilized a loading dose [4, 6, 7, 9, 10, 12, 13, 15, 16, 17, 19, 20, 22, 23, 24, 25, 26]. Subgroup analysis showed a statistically significant survival benefit with PI compared to SI when pooling studies that utilized a loading dose (RR 0.91 [95% confidence interval 0.84 to 0.99], I 2 = 2%), whereas pooling studies without a loading dose showed no significant difference (RR 0.82 [95% confidence interval 0.63–1.08], I 2 = 0%) (Figure 4). While there was no significant heterogeneity, the difference between a loading dose and no loading dose was not statistically significant (p = 0.48). These findings remained the same in sensitivity analysis, which included the BEATLE trial in the subgroup without a loading dose (Figure S3). Meta‐analysis of RCTs evaluating mortality in severely ill adult patients by loading dose. Of the 21 RCTs evaluating clinical cure in severely ill patients, 13 studies utilized a loading dose [4, 7, 9, 10, 16, 19, 20, 22, 24, 26, 29, 30, 50]. Subgroup analysis showed a statistically significant benefit with PI compared to SI when pooling studies that utilized a loading dose (RR 1.13 [95% confidence interval 1.08–1.19], I 2 = 6%), whereas pooling studies without a loading dose showed no significant difference (RR 1.04 [95% confidence interval 0.93–1.16], I 2 = 11%) (Figure 5). There was no significant heterogeneity (I 2 < 25%), similar to mortality results. As with the mortality analysis, the difference between using a loading dose versus no loading dose was not statistically significant with the random‐effects model (p = 0.18). In contrast, the fixed‐effect model showed a statistically significant difference (p = 0.04). These findings remained the same in the sensitivity analysis including the BEATLE trial in the subgroup without a loading dose (Figure S4). Meta‐analysis of RCTs evaluating clinical cure in severely ill adult patients by loading dose. Because most CI RCTs (and a minority of EI trials) used a loading dose, and because ~88% of the clinical‐cure dataset and ~92% of the mortality dataset came from studies that administered a load, it is prudent to continue using a loading dose until further studies clarify its impact. For CI, timing has not been uniform: several earlier RCTs began the maintenance infusion immediately after the load, whereas BLING III prespecified an initial 30‐min bolus followed by starting the continuous infusion at t50% of the intended intermittent interval (e.g., 4 h if q8h), not immediately after the bolus. In EI trials, two of four began the first maintenance dose on the usual q6–8 h schedule, one used q8h dosing initiated 6 h after the load, and one did not specify the post‐load start time [7]. Limitations and Future Research Needs Evidence syntheses and large CI trials generally prespecified a bolus‐then‐maintenance sequence. In contrast, EI trials have been heterogeneous and sometimes under‐specified in their post‐load timing, limiting clean comparisons. Future studies should standardize reporting of whether a load was given, its dose/infusion duration, and the exact interval from load completion to CI/EI start. They should also include adequately powered comparisons of PI with versus without a loading dose (stratified by illness severity and by CI versus EI), as well as dose‐finding to match loading doses to planned start times. Embedding β‐lactam TDM to link early exposure to outcomes is also needed. Conclusion This update to the international consensus recommendations for the use of PI β‐lactam antibiotics adds recommendations for severely ill patients based on the outcomes of mortality (7a) and clinical cure (7b). The updated analysis led us to recommend PI β‐lactams in severely ill patients with moderate (7a) and low (7b) certainty of evidence (Table 1 and Table S1) versus the previously combined recommendation which was determined to be very low certainty of evidence. With the new literature (from 2020 to 2025), eight new RCTs were considered and included both BLING III and MERCY, which added over 7900 patients for compiled analysis. While neither BLING III nor MERCY individually demonstrated improvements for mortality or clinical cure, both contribute meaningfully to the cumulative evidence indicating a modest but consistent benefit for both mortality and clinical cure. As such, we suggest PI for severely ill patients for both mortality and clinical cure reasons. An absolute mortality reduction of 2% corresponds to a number‐needed‐to‐treat of 50 to prevent one death. For loading doses, recommendations are similar to those made previously, as moderate benefits were seen for clinical cure and mortality; yet, it is difficult to attribute benefits to that intervention alone as studies did not randomize on that basis. Future studies are needed to understand the quantitative benefits for both EI interventions and loading doses for CI. TABLE 1Evidence summary table for efficacy outcomes among severely ill adult patients.Question: PI compared to SI in severely ill adult patientsCertainty assessmentNo of patientsEffectCertaintyImportanceNo of studiesStudy designRisk of biasInconsistencyIndirectnessImprecisionOther considerationsPISIRelative (95% CI)Absolute (95% CI) Mortality 28Randomized trialsSerious a Not seriousNot seriousNot seriousNone1199/5172 (23.2%)1326/5201 (25.5%)RR 0.91 (0.85 to 0.97)23 fewer per 1000 (from 38 fewer to 8 fewer) ⨁⨁⨁◯ Moderate a Critical Clinical cure 21Randomized trialsSerious a Not seriousSerious b Not seriousNone2664/4617 (57.7%)2405/4651 (51.7%)RR 1.11 (1.06 to 1.17)57 more per 1000 (from 31 more to 88 more) ⨁⨁◯◯ Low a , b Important Microbiological cure 8Randomized trialsSerious a Not seriousSerious c , d Not seriousNone178/243 (73.3%)150/240 (62.5%)RR 1.17 (1.06 to 1.30)106 more per 1000 (from 38 more to 188 more) ⨁⨁◯◯ Low a , c , d ImportantAbbreviations: CI, confidence interval; RR, risk ratio. a Lack of blinding. b Six trials combined clinical cure and clinical improvement (Roberts 2009, Bao 2017, Zhao 2017, Naiim 2022, Saad 2023, Alvarez‐Moreno 2024) and one did not define clinical efficacy (Lyu 2018). c Surrogate outcome. d Microbiological efficacy not defined by Lyu 2018. Evidence summary table for efficacy outcomes among severely ill adult patients. ⨁⨁⨁◯ Moderate a ⨁⨁◯◯ Low a , b ⨁⨁◯◯ Low a , c , d Abbreviations: CI, confidence interval; RR, risk ratio. Lack of blinding. Six trials combined clinical cure and clinical improvement (Roberts 2009, Bao 2017, Zhao 2017, Naiim 2022, Saad 2023, Alvarez‐Moreno 2024) and one did not define clinical efficacy (Lyu 2018). Surrogate outcome. Microbiological efficacy not defined by Lyu 2018. Author Contributions Lisa T. Hong: conceptualization, investigation, writing – original draft, methodology, formal analysis, validation, funding acquisition, writing – review and editing, project administration, data curation, supervision. Alireza FakhriRavari: validation, software, formal analysis, investigation, visualization, writing – review and editing, data curation, writing – original draft, methodology. Marta Ulldemolins: investigation, writing – original draft, validation, writing – review and editing. Jason A. Roberts: investigation, writing – original draft, conceptualization, validation, writing – review and editing. Robert A. Bonomo: writing – review and editing, validation, resources, methodology. Marc H. Scheetz: conceptualization, methodology, investigation, project administration, writing – original draft, validation, writing – review and editing, supervision. Funding The authors did not receive any specific funding for this study. Disclosure AI Use: No AI tools were used in the drafting of this manuscript, tables, or figures. Conflicts of interest Personal and financial relationships with industry and other entities are disclosed for all members of the panel in Appendix S1. Supporting information Appendix S1: Conflict of Interest Disclosures. Figure S1: Sensitivity analysis and funnel plot of RCTs evaluating mortality in severely ill adult patients, including BEATLE trial. Figure S2: Sensitivity analysis and funnel plot of RCTs evaluating clinical cure in severely ill adult patients, including BEATLE trial. Figure S3: Sensitivity analysis and funnel plot of RCTs evaluating mortality in severely ill adult patients by loading dose, including BEATLE trial. Figure S4: Sensitivity analysis and funnel plot of RCTs evaluating clinical cure in severely ill adult patients by loading dose, including BEATLE trial. Table S1: Evidence summary table for efficacy outcomes among severely ill adult patients, including BEATLE trial. Hong L. T., FakhriRavari A., Ulldemolins M., et al., “Focused Update to International Consensus Recommendations for the Use of Prolonged‐Infusion Beta‐Lactam Antibiotics Among Severely Ill Adult Patients: Endorsed by the American College of Clinical Pharmacy, British Society for Antimicrobial Chemotherapy, Cystic Fibrosis Foundation, European Society of Clinical Microbiology and Infectious Diseases, Infectious Diseases Society of America, Society of Critical Care Medicine, and Society of Infectious Diseases Pharmacists,” Pharmacotherapy: The Journal of Human Pharmacology and Drug Therapy 46, no. 9 (2026): e70188, 10.1002/phar.70188. The Prolonged‐Infusion Beta‐Lactam Consensus Panel includes the original 2023 consensus document authors who reviewed and approved this update and are listed at the end of the article. Scientific Editor: Lindsay DeVane The following individuals served as members of the Prolonged‐Infusion Consensus Panel for this focused update: Jacinda C. Abdul‐Mutakabbir, Mohammad H. Alshaer, Matteo Bassetti, Robert A. Bonomo, Kevin J. Downes, Alireza FakhriRavari, Mark Gilchrist, Lisa T. Hong, Jason A. Roberts, Marc H. Scheetz, Thomas Tängdén, Marta Ulldemolins, David C. Young. Contributor Information Lisa T. Hong, Email: lhong@llu.edu. the Prolonged‐Infusion Beta‐Lactam Consensus Panel: Jacinda C. Abdul‐Mutakabbir, Mohammad H. Alshaer, Matteo Bassetti, Robert A. Bonomo, Kevin J. Downes, Alireza FakhriRavari, Mark Gilchrist, Lisa T. Hong, Jason A. Roberts, Marc H. Scheetz, Thomas Tängdén, Marta Ulldemolins, and David C. Young Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request. References Associated Data Supplementary Materials Appendix S1: Conflict of Interest Disclosures. Figure S1: Sensitivity analysis and funnel plot of RCTs evaluating mortality in severely ill adult patients, including BEATLE trial. Figure S2: Sensitivity analysis and funnel plot of RCTs evaluating clinical cure in severely ill adult patients, including BEATLE trial. Figure S3: Sensitivity analysis and funnel plot of RCTs evaluating mortality in severely ill adult patients by loading dose, including BEATLE trial. Figure S4: Sensitivity analysis and funnel plot of RCTs evaluating clinical cure in severely ill adult patients by loading dose, including BEATLE trial. Table S1: Evidence summary table for efficacy outcomes among severely ill adult patients, including BEATLE trial. Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request.