Nosocomial pneumonia remains a substantial challenge for hospital infection prevention and critical care medicine. Hospital-acquired pneumonia (HAP) develops during hospitalization, while ventilator-associated pneumonia (VAP) follows invasive mechanical ventilation. Although these syndromes share microbial and therapeutic characteristics, their epidemiology, host factors, prognosis, and resource burdens are not interchangeable. The World Health Organization’s 2024 priority-pathogens framework highlights carbapenem-resistant Acinetobacter baumannii, resistant Enterobacterales, and other difficult-to-treat organisms as major antimicrobial-resistance threats [1]. Antibiotic selection in hospital pneumonia is complicated by heterogeneous susceptibility, prior antimicrobial exposure, biofilm formation, prolonged hospitalization, and microbiological colonization. Initial empirical antimicrobial therapy is ordinarily prescribed before definitive microbiological susceptibility results become available. Clinical guidelines emphasize local antibiograms, previous microbiology, severity of infection, and recognized MDR risk factors when selecting initial regimens for HAP and VAP [2]. Incorrect prediction of resistance can lead to an initially inactive regimen, while unnecessary broad-spectrum coverage increases ecological and patient-level harms. Crucially, MDR status and empirical treatment inappropriateness are related but not identical: a patient with MDR pneumonia may receive active empirical coverage, whereas a patient with susceptible infection may receive an inactive or otherwise inadequate regimen. The clinical consequences of resistance are also confounded by age, comorbidities, immunosuppression, organ dysfunction, and infection severity. The burden of hospital pneumonia extends beyond mechanically ventilated patients. Jones et al. identified 32,797 nonventilator-associated HAP events among 6,022,185 hospitalizations in 284 US hospitals, with 22.4% in-hospital mortality among affected patients [3]. A Chinese multicenter HAP cohort of 2,827 patients reported MDR prevalence of 64.9% and in-hospital mortality of 13.7% [4]. These estimates highlight important population and epidemiological differences, rather than a common global risk. Much previous quantitative research has emphasized mortality associated with initially inappropriate therapy. A separate assessment of MDR involvement at the VAP case level can provide complementary information about microbial ecology and empirical coverage needs. This synthesis therefore examines resistant infection burden, its implications for initial treatment, and associated clinical outcomes, while conducting an exploratory MDR-proportion meta-analysis. 1.1. Objectives • Quantify the proportion of VAP cases involving MDR microorganisms in independent cohorts with extractable numerators and denominators. • Describe organism-specific resistance patterns and differences across healthcare settings. • Examine clinical outcomes, including mortality, length of stay, and treatment complexity. • Assess how MDR status may influence the appropriateness of initial empirical antimicrobial coverage. • Identify methodological constraints and priorities for future research.

2.1. Review design and reporting This systematic review and meta-analysis followed PRISMA 2020 reporting principles [5]. The qualitative synthesis evaluated nosocomial pneumonia epidemiology, antimicrobial resistance, empirical-treatment appropriateness, and clinical outcomes. The primary quantitative endpoint was the proportion of ventilator-associated pneumonia (VAP) cases involving multidrug-resistant (MDR) pathogens. 2.2. Review question Among hospitalized patients with nosocomial pneumonia, what is the burden of MDR bacteria, and how does resistance interact with initial empirical antimicrobial treatment and clinical outcomes? 2.3. Literature identification Literature searches covered PubMed/MEDLINE, Embase, Scopus, Web of Science, the Cochrane Library, and reference lists through 31 January 2026. Search concepts comprised hospital-acquired pneumonia, ventilator-associated pneumonia, nosocomial pneumonia, multidrug resistance, carbapenem resistance, initial antimicrobial therapy, mortality, and clinical outcomes. The principal combined search expression was: ("hospital acquired pneumonia" OR "nosocomial pneumonia" OR "ventilator associated pneumonia") AND ("multidrug resistant" OR "antimicrobial resistance" OR "carbapenem resistant") AND ("empirical antimicrobial therapy" OR "initial antibiotic treatment" OR "mortality" OR "clinical outcomes") Database and citation-search records were screened against the prespecified study-eligibility criteria. Duplicate database records were removed, titles and abstracts were screened, full texts were evaluated, and full-text exclusions were categorized by reason. The study-selection totals and final inclusion decisions are reported in Table 2 and Figure 1. 2.4. Eligibility criteria Original human studies were eligible for qualitative synthesis when they reported MDR pathogens, treatment adequacy, or clinical outcomes in HAP or VAP. For the proportional meta-analysis, reports also had to identify a VAP population, a case-level numerator for MDR-associated disease, and an evaluable VAP denominator without overlap between pooled cohorts. Isolate-level resistance frequencies were not pooled as patient-level MDR involvement. Studies that combined VAP and ventilator-associated tracheobronchitis were discussed narratively if VAP-specific quantitative data were unavailable. 2.5. Data extraction and outcome definitions Study design, geographic location, population, HAP/VAP classification, total evaluable cases, MDR-associated cases, pathogen distribution, resistance definitions, initial empirical-treatment findings, and clinical outcomes were extracted from the cited reports. Definitions of MDR were retained as published. A VAP episode or patient was classified as MDR-associated according to each source study, recognizing that surveillance intensity and culture-positive selection could alter observed proportions. 2.6. Statistical analysis For each study, MDR-associated VAP prevalence was calculated as p = x/n, where x denotes MDR-associated cases and n denotes evaluable VAP cases. Study proportions were transformed using logit(p) = ln[p/(1 − p)]. Inverse-variance weighting with DerSimonian–Laird estimation of between-study variance was used to pool transformed proportions [6]. Cochran’s Q and I² described heterogeneity [7], and study-specific 95% confidence intervals were estimated using the Wilson method. Results were interpreted in light of recognized challenges in pooling prevalence estimates [8]. A sensitivity analysis excluded the Syrian cohort because its eligibility included patients from age 12 years onward. The resulting three-study estimate was compared with the main four-study estimate. A mortality meta-analysis was not undertaken because the evidence lacked sufficiently comparable MDR-versus-non-MDR mortality comparisons with aligned definitions and confounding adjustment. 2.7. Methodological appraisal Methodological appraisal examined sampling strategy, clinical and microbiological diagnosis of pneumonia, culture-positive selection, definitions of MDR, confounding, outcome measurement, and applicability. Study-specific methodological considerations are reported in Table 9 and were considered when interpreting the pooled findings.

Table 1. Review question and analytical framework Component Description Population Patients with HAP and/or VAP in hospital and ICU settings Exposure MDR organisms and inappropriate initial empirical antibiotic therapy Comparator Non-MDR organisms, appropriate antimicrobial therapy, or relevant clinical subgroups Primary quantitative endpoint Proportion of VAP cases involving MDR bacteria Secondary outcomes Mortality, treatment mismatch, clinical deterioration, length of stay, and microbiological outcomes Eligible evidence Prospective and retrospective clinical cohorts, surveillance investigations, and relevant cross-sectional studies 3.1. PRISMA 2020 study selection The searches yielded 603 database records and 21 additional citation-search reports. After removing 153 duplicates, 450 database records underwent title/abstract screening, of which 383 were excluded. Of 67 database reports sought, five were not retrieved, leaving 62 full texts. Of 21 citation-search reports, two were not retrieved, leaving 19 additional full texts. In total, 81 full-text reports were assessed and 70 were excluded: 20 were not original clinical investigations, 17 involved an ineligible population or setting, 14 lacked usable MDR or initial-treatment data, 11 had insufficient extractable clinical outcome data, and eight involved overlapping or duplicate populations. Eleven studies were included in qualitative synthesis, including four in the proportional meta-analysis (467 VAP cases) and seven in narrative synthesis only (Table 2; Figure 1). Table 2. PRISMA 2020 study-selection results PRISMA 2020 stage Number (n) IDENTIFICATION: DATABASES PubMed/MEDLINE 193 Embase 163 Scopus 121 Web of Science 85 Cochrane Library 41 Total records identified from databases 603 Duplicate records removed 153 SCREENING Records screened 450 Records excluded (title/abstract) 383 Reports sought for retrieval 67 Reports not retrieved 5 Full-text reports assessed (database pathway) 62 IDENTIFICATION: OTHER METHODS Reports identified through citation searching 21 Reports not retrieved (citation pathway) 2 Full-text reports assessed (citation pathway) 19 ELIGIBILITY Total full-text reports assessed 81 Full-text reports excluded 70 Not original clinical investigations 20 Wrong population or setting 17 No usable MDR or initial-treatment data 14 Insufficient extractable clinical outcomes 11 Overlapping or duplicate populations 8 TOTAL RECORDS FOUND ACROSS BOTH PATHWAYS 624 INCLUSION Studies in qualitative synthesis 11 Studies in proportional meta-analysis 4 Studies in qualitative synthesis only 7 VAP cases in meta-analysis 467 MDR-associated VAP cases 326 Figure 1. PRISMA 2020 study-selection flow diagram showing identification, screening, eligibility, and inclusion through database and citation-search pathways. Table 3. Included-study and meta-analysis counts Parameter Number Studies in qualitative synthesis 11 Studies contributing to meta-analysis 4 Studies in qualitative synthesis only 7 Evaluable VAP cases in meta-analysis 467 Cases with MDR pathogens 326 Cases without MDR involvement reported 141 Countries in the quantitative synthesis 4 3.2. Characteristics of included investigations The 11 included investigations covered patient-level VAP cohorts, nonventilated HAP surveillance, mixed HAP/VAP investigations, and isolate-level antimicrobial-resistance analyses. Four studies provided comparable case-level MDR proportions for meta-analysis. Larger surveillance studies and isolate-based reports were used to characterize clinical burden and resistance patterns without being added to the pooled denominator. Table 4. Characteristics and principal findings of the original investigations Study Setting and population Principal observation Jones et al., 2023 [3] USA; 32,797 NV-HAP events In-hospital mortality 22.4% Yin et al., 2021 [4] China; 2,827 HAP patients MDR prevalence 64.9%; mortality 13.7% Montrucchio et al., 2024 [9] Italy; 203 first VAP cases 150 MDR-associated cases Elsheikh et al., 2024 [10] Egypt; 62 VAP cases 25 MDR-associated cases Mantri et al., 2025 [11] Oman; 133 VAP cases 96 MDR-associated cases Khalil et al., 2025 [12] Syria; 69 culture-positive VAP cases 55 cases involved MDR organisms Colaneri et al., 2025 [13] Italy; 11,978 VAP cases VAP mortality approximately 30% Kajeekul et al., 2024 [14] Thailand; 1,824 ventilated patients 152 MDR Gram-negative VAP cases Torres et al., 2025 [15] Spain; 573 patients, 585 admissions MDR colonization predicted resistant LRTI Sharma and Shah, 2025 [16] India; 321 VAP cases High isolate-level MDR proportions Mallik et al., 2026 [17] Bangladesh; 100 HAP patients 89 isolates; reported MDR prevalence 87.8% 3.3. MDR involvement in VAP Montrucchio et al. identified MDR pathogens in 150 of 203 first VAP cases (73.9%) [9]. Elsheikh et al. observed MDR involvement in 25 of 62 VAP cases (40.3%) in a prospective Egyptian cohort [10]. Mantri et al. reported 96 MDR-associated cases among 133 VAP cases (72.2%) in an Omani ICU cohort [11]. In the Syrian investigation, 55 of 69 culture-positive VAP cases (79.7%) involved MDR bacteria [12]. The lowest MDR proportion was observed in the Egyptian cohort and the highest in the Syrian culture-positive group. Selection based on positive cultures and age eligibility could partly explain between-study differences; these cohorts should not be treated as epidemiologically equivalent. Table 5. Quantitative characteristics of four VAP cohorts Study MDR / total Proportion (%) 95% CI (%) Montrucchio et al., 2024 150 / 203 73.9 67.4–79.5 Elsheikh et al., 2024 25 / 62 40.3 29.0–52.7 Mantri et al., 2025 96 / 133 72.2 64.0–79.1 Khalil et al., 2025 55 / 69 79.7 68.8–87.5 Descriptive total 326 / 467 69.8 Not pooled by simple addition The descriptive combined proportion of 69.8% represents the ratio of all MDR-associated cases to total cases. It is not identical to the random-effects pooled estimate and should not be interpreted as a universal prevalence. 3.4. Pooled random-effects MDR proportion The logit-transformed random-effects meta-analysis estimated an MDR-associated VAP proportion of 67.8% (95% CI 52.4%–80.2%). Cochran’s Q was 27.65 (3 degrees of freedom; P < 0.001), with I² = 89.1% and τ² = 0.387 on the logit scale. High heterogeneity indicates important clinical and sampling differences between studies. Table 6. Random-effects meta-analysis results Measure Result Number of studies 4 Evaluable cases 467 MDR-associated cases 326 Descriptive combined proportion 69.8% Random-effects pooled proportion 67.8% 95% confidence interval 52.4%–80.2% Cochran’s Q / degrees of freedom 27.65 / 3 Q-test P value < 0.001 I² 89.1% τ² (logit scale) 0.387 Figure 2. Forest plot of MDR-associated VAP proportions from four cohorts. Squares indicate individual proportions, horizontal lines show Wilson 95% confidence intervals, and the diamond shows the random-effects estimate. 3.5. Adult-cohort sensitivity analysis Because Khalil et al. included persons from age 12 years, the mixed-age cohort was removed in a prespecified interpretive sensitivity analysis. The remaining three largely adult studies comprised 398 cases, including 271 MDR-associated cases. The pooled proportion decreased to 63.6% (95% CI 44.4%–79.2%), while heterogeneity remained high (I² = 91.6%). Table 7. Main analysis versus adult-cohort sensitivity analysis Parameter Four-study analysis Three-study sensitivity Studies 4 3 VAP cases 467 398 MDR-associated cases 326 271 Pooled proportion 67.8% 63.6% 95% CI 52.4%–80.2% 44.4%–79.2% I² 89.1% 91.6% 3.6. Clinical outcomes and resource burden Large epidemiological studies underscored the clinical burden of hospital pneumonia. In US electronic surveillance, Jones et al. observed 22.4% in-hospital mortality among patients meeting NV-HAP criteria [3]. Yin et al. reported 13.7% in-hospital mortality among 2,827 Chinese HAP patients, alongside a high proportion of resistant organisms [4]. These values differ by case definition and population and were not pooled with VAP-specific mortality. In Italian surveillance, Colaneri et al. reported 11,978 VAP cases among 122,430 patients at risk during mechanical ventilation, with 10.47 episodes per 1,000 ventilator-days and mortality of approximately 30% [13]. Kajeekul et al. identified 152 MDR Gram-negative VAP cases among 1,824 ventilated patients; the reported incidence was 12.44 per 1,000 ventilator-days [14]. Resistance-related burden therefore has implications for prevention and resource use, not just antibiotic choice. 3.7. MDR colonization and organism-specific resistance Torres et al. identified MDR colonization in 201 admissions among a Spanish cohort involving 573 patients and 585 admissions [15]. Prior MDR colonization was associated with subsequent resistant lower respiratory tract infection (adjusted HR 6.59, 95% CI 2.67–16.26). Colonization was not interchangeable with active infection and should be interpreted with respiratory symptoms, imaging, and microbiological findings. Sharma and Shah reported isolate-level MDR frequencies of 97.01% among A. baumannii complex isolates and 93.53% among K. pneumoniae isolates in their VAP investigation [16]. Mallik et al. reported MDR prevalence of 87.8% among bacteria isolated in their HAP investigation [17]. These are isolate-based measures and were therefore excluded from the patient-level VAP meta-analysis. Table 8. Selected resistance and clinical-burden findings Investigation Finding Interpretation Yin et al. [4] MDR prevalence 64.9% Substantial resistance burden in Chinese HAP cohort Montrucchio et al. [9] 150/203 MDR-associated VAP High MDR case burden in prospective ICU study Elsheikh et al. [10] 25/62 MDR-associated VAP Lower proportion within pooled evidence set Mantri et al. [11] 96/133 MDR-associated VAP High MDR burden in Omani ICU Khalil et al. [12] 55/69 culture-positive VAP Culture-positive selection limits extrapolation Colaneri et al. [13] VAP mortality approximately 30% Substantial disease burden across centers Torres et al. [15] Adjusted HR 6.59 after MDR colonization Potential value of prior microbiological status Sharma and Shah [16] 97.01% MDR among A. baumannii isolates Isolate-level resistance, not patient prevalence Mallik et al. [17] Reported MDR prevalence 87.8% Single-center isolate-level finding 3.8. Empirical treatment appropriateness MDR involvement was the primary quantitative endpoint, but initially inadequate antimicrobial therapy is clinically relevant. The chance of empiric mismatch is influenced by local susceptibility and the clinician’s recognition of prior exposure or colonization. A resistant pathogen does not automatically imply inappropriate treatment; conversely, initial coverage may be inappropriate even for a non-MDR pathogen when selected drugs lack activity or are delayed. The included reports differed in definitions of empirical-treatment appropriateness, timing of regimen modification, and clinical outcomes. Mantri et al. described limitations of single-agent coverage against local VAP pathogens [11], while other reports documented high organism-specific resistance [12,16,17]. These observations support treatment selection informed by susceptibility data and antimicrobial stewardship. A mortality effect of initially inappropriate treatment was not pooled because suitable, consistently defined comparative data were unavailable. 3.9. Study-level methodological appraisal Differences in case definitions, sampling procedures, comparison groups, and units of analysis were considered for each original investigation. Table 9 records the principal concern affecting interpretation and avoids conflating isolate-based prevalence with patient-level VAP outcomes. Table 9. Study-level methodological appraisal and interpretive limitations Study Principal methodological concern Implication Jones et al. [3] Electronic surveillance case definition Potential misclassification compared with clinically adjudicated HAP Yin et al. [4] Long observation period across hospitals Temporal and institutional heterogeneity Montrucchio et al. [9] COVID-19 and non-COVID-19 case mix Different host severity and exposure profiles Elsheikh et al. [10] Single center; 62 VAP cases Limited precision and transportability Mantri et al. [11] Retrospective and single center Selection/information bias Khalil et al. [12] Culture-positive and mixed-age sample Differential denominator and applicability Colaneri et al. [13] Multicenter surveillance definitions Differences in site-specific ascertainment Kajeekul et al. [14] Focused on MDR Gram-negative cases Restricted pathogen spectrum Torres et al. [15] VAP/VAT composite and admissions-based units Not equivalent to VAP-only patient proportions Sharma and Shah [16] Isolate-level resistance frequencies Cannot treat isolate percentage as case prevalence Mallik et al. [17] Single-center isolate reporting Limited generalizability and different denominator

.1. Principal findings Across four identifiable cohorts, the estimated pooled proportion of MDR-associated VAP was 67.8%. This result indicates that resistant bacteria are frequently involved in VAP at the studied centers; however, the high I² and broad confidence interval preclude applying the pooled number indiscriminately to other hospitals. The analysis differs from earlier mortality-centered syntheses by treating MDR case involvement as the primary quantitative outcome. 4.2. Relationship between resistance and empirical antibiotic adequacy MDR organisms can evade routinely selected empirical agents, increasing the risk of inactive initial therapy. Resistance may require agents with narrower therapeutic windows, additional toxicity, or complex dosing. Nevertheless, the association between MDR and treatment inadequacy is mediated by local antibiograms, known colonization, prior culture data, severity, and choice of treatment protocol. Resistance is therefore neither synonymous with inadequate coverage nor, by itself, proof of mortality causation. 4.3. Clinical outcomes and confounding Hospital pneumonia mortality is influenced by age, illness severity, organ dysfunction, immune status, mechanical ventilation, and withdrawal-of-care decisions. Resistant pathogens are often identified in populations with prolonged prior healthcare exposure and multiple baseline risks. Comparisons of MDR versus susceptible pneumonia may thus overstate a direct causal effect unless key confounding factors and timing are appropriately addressed. The disparate HAP and VAP mortality figures in the mapped studies were consequently summarized narratively, not quantitatively pooled. 4.4. Hospital ecology and antimicrobial stewardship The four pooled studies reported MDR-associated proportions from 40.3% to 79.7%. Such variation reflects differences in local microbial ecology, previous antimicrobial exposure, infection-prevention practices, and diagnostic thresholds. A pooled international estimate cannot replace institution- or unit-specific susceptibility surveillance. Empirical treatment should reflect local resistance patterns and individual risk, then be reassessed as clinical and microbiological information emerges. Stewardship includes obtaining appropriate respiratory samples without delaying treatment in unstable patients, verifying activity when susceptibility results arrive, and narrowing or discontinuing antibiotics when justified. Excessive empirical broad-spectrum use may select resistance and cause adverse events; conversely, undercoverage in severely ill patients may lead to progression of infection. 4.5. MDR colonization, diagnostics, and prevention Previous MDR colonization can inform risk assessment, although surveillance positivity is not equivalent to active lower respiratory infection. Clinical presentation, quality of sampling, microbiology, and imaging must be interpreted together. Prevention strategies include hand hygiene, appropriate ventilator-care practices, minimizing unnecessary ventilation, antimicrobial stewardship, and infection-control surveillance. 4.6. Comparison with existing evidence Li et al. reviewed MDR P. aeruginosa in VAP and reported a pooled MDR prevalence of 33% among P. aeruginosa isolates [18]. This outcome is not directly comparable with the present estimate because the earlier denominator consisted of isolates from one species, while this analysis pooled MDR involvement in VAP cases across bacterial pathogens. Numerator and denominator definitions are essential for meaningful comparison between resistance meta-analyses. 4.7. Strengths of the review Strengths include transparent study-level MDR numerators and denominators, the separation of isolate-level resistance from case-level MDR involvement, random-effects synthesis, and a sensitivity analysis addressing mixed-age enrollment. The review incorporates clinical surveillance evidence while maintaining a distinct, predefined quantitative endpoint. 4.8. Limitations The pooled analysis comprised four studies and showed substantial clinical and statistical heterogeneity. The Syrian study enrolled adolescents and was restricted to culture-positive cases, while the other cohorts used differing VAP and MDR definitions. Resistance testing practices and differences in clinical populations may have influenced observed proportions. Comparable case-level outcome data were insufficient for a separate mortality meta-analysis of inappropriate initial treatment. The pooled estimate should therefore be interpreted as the summary of the included populations rather than a universal hospital prevalence. 4.9. Future research Prospective multicenter investigations should employ consistent MDR definitions, standardized VAP diagnosis, and prespecified units of analysis. Reports should provide exact patient-level resistance and initial-treatment adequacy denominators; timing of effective antimicrobial treatment; antibiotic changes; and standardized 28-day and hospital mortality endpoints with confounder adjustment. These data would permit stronger estimates of the independent clinical burden associated with resistance and treatment mismatch.

MDR bacteria accounted for a substantial proportion of VAP in the included cohorts. The four-study meta-analysis of 467 VAP cases estimated an MDR-associated proportion of 67.8% (95% CI 52.4%–80.2%), with substantial between-study heterogeneity. Antimicrobial resistance complicates initial treatment selection, while outcomes remain influenced by underlying illness, microbiological assessment, and subsequent care. Hospital-specific susceptibility surveillance, timely active therapy, infection prevention, and antimicrobial stewardship are key components of care. Future studies should report standardized MDR definitions and comparable clinical outcomes to clarify the independent effects of resistance and initially inappropriate treatment.

1. World Health Organization. WHO bacterial priority pathogens list, 2024: bacterial pathogens of public health importance to guide research, development and strategies to prevent and control antimicrobial resistance. Geneva: WHO; 2024. https://www.who.int/publications/i/item/9789240093461 2. Kalil AC, Metersky ML, Klompas M, et al. Management of adults with hospital-acquired and ventilator-associated pneumonia: 2016 clinical practice guidelines by the Infectious Diseases Society of America and the American Thoracic Society. Clin Infect Dis. 2016;63(5):e61–e111. doi:10.1093/cid/ciw353. 3. Jones BE, Sarvet AL, Ying J, et al. Incidence and outcomes of non–ventilator-associated hospital-acquired pneumonia in 284 US hospitals using electronic surveillance criteria. JAMA Netw Open. 2023;6(5):e2314185. doi:10.1001/jamanetworkopen.2023.14185. 4. Yin Y, Zhao C, Li H, et al. Clinical and microbiological characteristics of adults with hospital-acquired pneumonia: a 10-year prospective observational study in China. Eur J Clin Microbiol Infect Dis. 2021;40:683–690. doi:10.1007/s10096-020-04046-9. 5. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. doi:10.1136/bmj.n71. 6. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;7(3):177–188. doi:10.1016/0197-2456(86)90046-2. 7. Higgins JPT, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ. 2003;327:557–560. doi:10.1136/bmj.327.7414.557. 8. Barendregt JJ, Doi SA, Lee YY, Norman RE, Vos T. Meta-analysis of prevalence. J Epidemiol Community Health. 2013;67(11):974–978. doi:10.1136/jech-2013-203104. 9. Montrucchio G, Balzani E, Sales G, et al. Multidrug-resistant pathogens and ventilator-associated pneumonia in critically ill COVID-19 and non-COVID-19 patients: a prospective observational monocentric comparative study. Respir Res. 2024;25:168. doi:10.1186/s12931-024-02779-1. 10. Elsheikh M, Kuriyama A, Goto Y, et al. Incidence and predictors of ventilator-associated pneumonia using a competing risk analysis: a single-center prospective cohort study in Egypt. BMC Infect Dis. 2024;24:1007. doi:10.1186/s12879-024-09909-6. 11. Mantri T, Burad J, Al Hashmi S, et al. A retrospective closed cohort study on distribution of multidrug-resistant bacteria in ventilator-associated pneumonia and its impact on patient outcome. J Intensive Care Med. 2025;40(4). doi:10.1177/08850666241290468. 12. Khalil KA, Alsultan M, Daher NA. Microbial profile and antimicrobial resistance patterns in ventilator-associated pneumonia (VAP): a cross-sectional study from Syria. J Postgrad Med. 2025;71(1):7–14. doi:10.4103/jpgm.jpgm_565_24. 13. Colaneri M, Montrucchio G, Scaglione G, et al. Incidence, microbiology, and mortality of ventilation-associated pneumonia in a large Italian cohort of critically ill patients: results from the PROSAFE project. Clin Microbiol Infect. 2025;31(9):1491–1499. doi:10.1016/j.cmi.2025.05.026. 14. Kajeekul R, Thamlikitkul V, Wonglaksanapimon S, Rattanaumpawan P. Epidemiology of ventilator-associated tracheobronchitis and ventilator-associated pneumonia caused by multidrug-resistant Gram-negative bacteria at a tertiary care hospital in Thailand. JAC Antimicrob Resist. 2024;6(1):dlad140. doi:10.1093/jacamr/dlad140. 15. Torres I, Ferrer J, Colomina J, et al. Ventilator-associated lower respiratory tract infections and bacteremia in medical and surgical critical care patients colonized by multidrug-resistant bacteria. BMC Infect Dis. 2025;25:1583. doi:10.1186/s12879-025-12027-6. 16. Sharma UK, Shah AD. Antimicrobial resistance patterns and clinical outcomes in ventilator-associated pneumonia. Bioinformation. 2025;21(8):2410–2416. doi:10.6026/973206300212410. 17. Mallik MJA, Rahman MM, Rahman MA, et al. Bacterial etiology and multidrug resistance patterns in hospital-acquired pneumonia: predominance of Gram-negative pathogens. Int J Res Med Sci. 2026. doi:10.18203/2320-6012.ijrms20263490. 18. Li Y, Roberts JA, Walker MM, Aslan AT, Harris PNA, Sime FB. The global epidemiology of ventilator-associated pneumonia caused by multi-drug resistant Pseudomonas aeruginosa: a systematic review and meta-analysis. Int J Infect Dis. 2024;139:78–85. doi:10.1016/j.ijid.2023.11.023.