Temporal trends, antimicrobial non-susceptibility patterns, and clinical predictors of Stenotrophomonas maltophilia in a tertiary care hospital in Palestine: a 7-year retrospective study Authors: Taqwa Abdeljawad, Wafa ’ Sous, Israa Hasan, Nada Janajra, Adham AbuTaha, Ali Sabateen Published: 2026 ABSTRACT AbstractBackgroundStenotrophomonas maltophilia is an opportunistic pathogen of increasing clinical importance because of its intrinsic resistance to multiple antimicrobial agents and its growing presence in healthcare environments. However, data describing its epidemiology and antimicrobial susceptibility patterns in Palestine remain limited. This study examined temporal trends, distribution across specimen types and care settings, and antimicrobial resistance profiles of S. maltophilia isolates.MethodsA retrospective observational study was conducted at a tertiary care hospital between 2018 and 2024. Unique clinical isolates of S. maltophilia were identified using the VITEK® 2 Compact system. Antimicrobial susceptibility to cotrimoxazole, levofloxacin, and ceftazidime was interpreted according to Clinical and Laboratory Standards Institute (CLSI) criteria applied during the respective testing period. Temporal changes in prevalence, distribution across hospital wards and specimen types, and antimicrobial non-susceptibility patterns were evaluated using stratified analyses, trend tests, and multivariable logistic regression models.ResultsA total of 177 unique S. maltophilia isolates were identified, corresponding to a prevalence of 1.08% among all cultures performed. Annual prevalence fluctuated during the study period without evidence of a sustained linear trend. Respiratory specimens accounted for the largest proportion of isolates, and more than one-third originated from intensive care unit (ICU)-related settings. Susceptibility remained highest for cotrimoxazole (80.2%) and levofloxacin (78.5%), whereas ceftazidime showed lower historical non-susceptibility rates (42.9%). Paired analyses demonstrated significantly higher non-susceptibility to ceftazidime than to cotrimoxazole or levofloxacin (p < 0.001). Combined non-susceptibility (non-susceptibility to all three tested agents; CNP) occurred in 5.6% of isolates and was significantly more frequent among ICU-related isolates (21.7% vs. 3.2%, p = 0.004). In multivariable analyses, prior antibiotic exposure (p = 0.044) and COVID-19 infection (p = 0.027) were independently associated with levofloxacin non-susceptibility, whereas invasive medical device use was associated with ceftazidime non-susceptibility (p = 0.004).ConclusionsS. maltophilia isolates demonstrated heterogeneous antimicrobial non-susceptibility patterns, with preserved susceptibility to cotrimoxazole and levofloxacin but reduced activity of ceftazidime. The higher frequency of the combined non-susceptibility phenotype (CNP) among ICU-related isolates suggests that critical care settings may serve as reservoirs for more resistant phenotypes. Continued microbiological surveillance and antimicrobial stewardship are essential to guide empiric therapy and infection-control strategies.Clinical trial registrationNot applicable.Keywords: Stenotrophomonas maltophilia, Antimicrobial resistance, Epidemiology, Intensive care unit, Antimicrobial stewardship, Surveillance, Palestine Background Stenotrophomonas maltophilia is an environmental, non-fermenting, Gram-negative bacillus that has gained increasing clinical attention over the past two decades as an important opportunistic human pathogen [1, 2]. Although multiple species belong to the genus Stenotrophomonas, S. maltophilia is the only species consistently associated with human infection [3]. The incidence of S. maltophilia has increased worldwide, and the organism is now recognized as an emerging nosocomial pathogen, particularly among immunocompromised patients and those admitted to intensive care units (ICUs) [4]. Infections are frequently associated with invasive medical devices, including central venous catheters, endotracheal tubes, mechanical ventilation, and indwelling urinary catheters, largely because of the organism’s ability to adhere to abiotic surfaces and form biofilms [5]. Reported clinical syndromes range from skin and soft tissue, urinary tract, ocular, and wound infections to severe manifestations such as pneumonia, bacteremia, sepsis, endocarditis, osteomyelitis, and meningitis [6]. Although infection occurs predominantly in immunocompromised hosts, cases have also been reported in immunocompetent individuals [7]. Management of S. maltophilia infection is challenging because of its high-level intrinsic resistance to multiple antimicrobial classes [8]. This resistance is mediated by chromosomally encoded L1 and L2 β-lactamases, multidrug efflux pumps, reduced outer membrane permeability, and biofilm formation, which together limit the activity of β-lactams, aminoglycosides, and several other agents [8, 9]. Resistance to fluoroquinolones and cotrimoxazole (trimethoprim-sulfamethoxazole; SXT), although not uniformly intrinsic, has been increasingly reported and further complicates treatment [10, 11]. In addition, S. maltophilia may acquire highly resistant phenotypes that substantially narrow therapeutic options. The spread of resistant strains has been linked to extensive exposure to broad-spectrum antibiotics, a pattern that intensified during the COVID-19 pandemic, especially in critical care settings [12, 13]. SXT remains the first-line therapy for S. maltophilia infection; however, its use may be limited by intolerance, hypersensitivity reactions, drug-drug interactions, and rising resistance rates [11]. Alternative agents, including levofloxacin (LVX) and ceftazidime (CAZ), are often used in clinical practice, yet prior exposure to these antibiotics has been associated with the selection of resistant strains [14]. Resistance rates vary considerably across regions, reflecting differences in prescribing practices, stewardship policies, laboratory standards, and patient case-mix [15]. Inappropriate empirical use of broad-spectrum antibiotics may further accelerate resistance development and increase the risk of treatment failure [15]. Access to reliable local susceptibility data is therefore essential to guide empiric therapy and optimize patient management. Despite the growing clinical importance of S. maltophilia, few studies from low- and middle-income countries have provided longitudinal surveillance of its epidemiology and antimicrobial resistance using structured multivariable analyses, and data from the Eastern Mediterranean region remain scarce. In Palestine, no longitudinal study has previously characterized the burden of this organism. This study aims to address this gap by describing the epidemiology, temporal trends, antimicrobial resistance patterns, and clinical predictors of S. maltophilia isolates recovered at a tertiary hospital over a seven-year period. The findings provide locally relevant evidence to inform clinical decision-making, support antimicrobial stewardship, and improve infection control strategies in resource-limited healthcare settings. Methods Study design and setting This retrospective observational study was conducted at An-Najah National University Hospital (ANNUH) in Nablus, West Bank, Palestine, and included hospitalized patients with clinical isolates positive for Stenotrophomonas maltophilia between January 2018 and December 2024. ANNUH is a tertiary teaching hospital and one of the major referral healthcare institutions in Palestine, providing advanced medical services across a broad range of specialties. The hospital includes multiple specialized departments, including intensive care, coronary care, nephrology, cardiology, neonatology, hematology, and surgical services, and serves patients from both the West Bank and Gaza Strip. Study population and eligibility criteria The study included all hospitalized patients with at least one clinical isolate positive for S. maltophilia identified during the study period. Data were collected retrospectively from patients’ electronic medical records using the hospital information system. Data extraction was independently verified by two investigators to enhance accuracy and completeness. When multiple isolates were obtained from the same patient, only the first isolate per patient was included in the analysis to avoid duplication. Only clinical specimens obtained as part of routine patient care were included. Patients were excluded if medical records were incomplete or if isolates represented duplicate cultures from the same individual. After applying these criteria, 177 unique patients with confirmed S. maltophilia isolates were included in the final analysis. Classification of hospital wards and care settings Hospital departments were categorized according to the level of care provided. Wards were classified as ICU-related settings if patients were admitted to the Intensive Care Unit (ICU), Surgical Intensive Care Unit (SICU), Pediatric Intensive Care Unit (PICU), or Coronary Care Unit (CCU). These units were grouped together because they share defining features of high-dependency care, including intensive monitoring, greater utilization of invasive devices, and higher cumulative antimicrobial exposure relative to general wards. This grouping was primarily motivated by statistical considerations, given the limited number of isolates within individual ICU subtypes, which precluded reliable unit-specific regression analyses. The individual distribution of isolate counts and non-susceptibility rates by ICU subtype is provided in Table 4. It is acknowledged, however, that grouping these units introduces a degree of clinical heterogeneity: the PICU population differs substantially from the adult ICU, SICU, and CCU populations with regard to age distribution, immune status, and device utilization patterns, and these differences may influence both isolate characteristics and resistance phenotypes. Inter-unit heterogeneity within the combined ICU group is therefore recognized as a limitation of this analytical approach. All other departments were classified as non-ICU wards. This classification was applied consistently across all stratified analyses. Table 4Distribution of antibiotic non-susceptibility by hospital wardWardSXT NS1 n (%)2LVX NS n (%)CAZ NS n (%)Surgical Ward (SW)5 (14.3%)6 (18.8%)16 (57.1%)Intensive Care Unit (ICU)7 (30.4%)8 (36.4%)10 (50.0%)Medical Ward (MW)3 (16.7%)3 (16.7%)6 (35.3%)Coronary Care Unit (CCU)4 (23.5%)2 (12.5%)8 (88.9%)Vascular Surgery Ward4 (25.0%)4 (25.0%)6 (40.0%)Surgical Intensive Care Unit (SICU)5 (33.3%)3 (20.0%)7 (46.7%)Emergency Department (ED)0 (0.0%)1 (7.7%)1 (8.3%)Pediatric Intensive Care Unit (PICU)2 (16.7%)0 (0.0%)7 (58.3%)Kidney Unit (KU)2 (16.7%)0 (0.0%)9 (81.8%)Cardiac Surgery Ward (CSW)1 (20.0%)0 (0.0%)1 (50.0%)Outpatient Department (OP)0 (0.0%)0 (0.0%)1 (33.3%)Oncology Ward0 (0.0%)0 (0.0%)0 (0.0%)Pediatric Ward (PW)0 (0.0%)0 (0.0%)0 (0.0%)Bone Marrow Transplant Unit (BMT)1 (100.0%)1 (100.0%)0 (0.0%)Cardiology Ward0 (0.0%)NT*NT1 Non-susceptibility (NS) was defined as resistant (R) or intermediate (I)2 Percentages were calculated using the total number of tested isolates per ward. Percentages for wards with small numbers of isolates should be interpreted cautiously*NT = Not testedAbbreviations: SXT, cotrimoxazole; LVX, levofloxacin; and CAZ, ceftazidime Distribution of antibiotic non-susceptibility by hospital ward 1 Non-susceptibility (NS) was defined as resistant (R) or intermediate (I) 2 Percentages were calculated using the total number of tested isolates per ward. Percentages for wards with small numbers of isolates should be interpreted cautiously *NT = Not tested Abbreviations: SXT, cotrimoxazole; LVX, levofloxacin; and CAZ, ceftazidime Microbiological identification and antimicrobial susceptibility testing Bacterial identification and antimicrobial susceptibility testing (AST) were performed using the VITEK® 2 Compact automated system (bioMérieux, Marcy-l’Étoile, France) according to the manufacturer’s instructions. The antimicrobial agents included cotrimoxazole (SXT), levofloxacin (LVX), and ceftazidime (CAZ), which represent commonly used therapeutic options for S. maltophilia infections. VITEK® 2 AST-N222 cards were used for susceptibility testing. Categorical susceptibility interpretations (Susceptible, Intermediate, Resistant) were extracted from laboratory records. For ceftazidime against S. maltophilia, categorization was based on the historical Clinical and Laboratory Standards Institute (CLSI) M100 guidelines [16], active during the majority of the study period. This historical breakpoint was applied uniformly across the 2018–2024 dataset to allow for standardized longitudinal epidemiological comparison. For analytical purposes, intermediate and resistant categories were combined and reported as non-susceptible (NS). Quality control procedures were performed in accordance with laboratory standards to ensure the accuracy and reliability of susceptibility testing. Throughout this manuscript, the primary descriptive term used is “non-susceptibility” (or “non-susceptible”) to reflect this combined analytical framework. The term “resistance” is used only where established literature is being cited and the original terminology of the cited study is preserved, or where it refers to the broader biological and mechanistic concept of antimicrobial resistance rather than to the specific susceptibility categories analyzed in this study. Because only three routinely reported antimicrobial agents were available for most isolates, standard multidrug-resistance (MDR) classifications, which require testing across three or more distinct antimicrobial categories, could not be directly applied. Therefore, this study used the study-specific operational term “combined non-susceptibility phenotype” (CNP), defined as simultaneous non-susceptibility to all three tested agents: cotrimoxazole (SXT), levofloxacin (LVX), and ceftazidime (CAZ), among isolates with complete susceptibility data for all three agents. This classification does not correspond to standardized MDR criteria and should be interpreted solely as a study-specific operational definition. Stratified resistance analyses Antimicrobial non-susceptibility patterns were examined using stratified analyses based on hospital ward, ICU versus non-ICU setting, specimen type, and patient-related clinical risk factors. Paired comparisons of non-susceptibility between antibiotics were performed using McNemar’s test, restricted to isolates with available susceptibility results for both agents being compared. Exact McNemar tests were used when appropriate. Temporal trends in annual culture-based prevalence were assessed using the Cochran–Armitage test for trend. Poisson regression adjusted for annual culture volume was used to assess linear temporal trends in isolate counts. Simple linear regression was additionally used as a descriptive analysis to explore the association between calendar year and annual isolate counts. Ward- and year-specific isolate distributions were visualized using heatmaps to illustrate temporal and spatial clustering patterns. Statistical analysis Categorical variables were summarized using frequencies and percentages. Comparisons of antimicrobial non-susceptibility across hospital wards, ICU versus non-ICU settings, specimen types, and patient-related clinical factors were performed using the chi-square test or Fisher’s exact test when expected cell counts were below five. Multivariable logistic regression analyses were performed to identify independent predictors of antimicrobial non-susceptibility. Variables with p < 0.10 in univariable analyses or considered clinically relevant were entered into multivariable logistic regression models. Given the limited number of non-susceptible events for some outcomes, specifically, 34 non-susceptible events for cotrimoxazole, 28 for levofloxacin, and 72 for ceftazidime, with 10 isolates meeting the combined non-susceptibility phenotype (CNP) definition, variable selection was restricted to maintain a reasonable events-per-variable ratio and reduce model instability. The number of outcome events included in each regression model is reported in the corresponding table footnotes. To minimize multicollinearity and overfitting, clinically overlapping variables were not entered simultaneously in the same multivariable model [17]. Model fit was assessed using the Hosmer–Lemeshow goodness-of-fit test, with p > 0.05 considered indicative of acceptable calibration. Adjusted odds ratios (aORs) with 95% confidence intervals (CIs) were reported. The variable “COVID-19 infection” was included in regression models as a binary contextual marker for isolates obtained during a period of intensified antimicrobial pressure and altered healthcare practices (2020–2021), rather than as a direct biological exposure; its independent contribution cannot be disaggregated from concurrent co-occurring exposures using the available data. Variables with missing data were analyzed using available-case analysis without imputation. All statistical tests were two-sided, and p < 0.05 was considered statistically significant. Statistical analyses were performed using IBM SPSS Statistics version 21 (IBM Corp., Armonk, NY, USA). Bias control To reduce selection and information bias, only the first isolate per patient was included, and records with missing or inconsistent data were excluded. Because this was a retrospective study and many isolates were recovered from respiratory specimens, misclassification between colonization and true infection cannot be fully excluded. Ethical considerations The study protocol was approved by the Institutional Review Board (IRB) of An-Najah National University, and permission to access patient records was obtained from the Clinical Research Center of ANNUH. Additionally, we ensured the confidentiality of patient data, as the data were only used for clinical research purposes and were not shared. Data were coded by numbers, not by patient names, and access was limited to study team members. Given the retrospective design of the study and the use of anonymized patient data, the requirement for informed consent was waived by the IRB. Reporting standards This study was reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. Results Study flow and isolate selection Between January 2018 and December 2024, a total of 16,393 positive clinical cultures were processed at An-Najah National University Hospital. Among these, 201 cultures yielded Stenotrophomonas maltophilia. After excluding duplicate isolates from the same patient (n = 22) and isolates with incomplete records (n = 2), 177 unique isolates were included in the final analysis. The isolate selection process is illustrated in Fig. 1. Study flow diagram. Study flow diagram illustrating the selection of Stenotrophomonas maltophilia isolates included in the analysis at An-Najah National University Hospital between 2018 and 2024. Duplicate isolates from the same patient were excluded, and isolates with incomplete medical records were removed prior to the final analysis Annual prevalence and temporal trends Between 2018 and 2024, 177 unique S. maltophilia isolates were identified, corresponding to an overall culture-based prevalence of 1.08% among all cultures performed during the study period. Annual prevalence varied across the study period, with marked inter-annual variability. Trend analysis using the Cochran–Armitage test indicated a statistically significant trend in prevalence proportions over time (χ² = 42.12, p < 0.001). In contrast, Poisson regression adjusted for annual culture volume did not demonstrate a significant linear temporal trend (p = 0.106), suggesting fluctuation rather than a sustained directional increase or decrease (Fig. 2). Annual number of Stenotrophomonas maltophilia isolates identified between 2018 and 2024 The highest isolate counts were observed during 2020–2021. Linear regression analysis showed a weak association between calendar year and isolate counts (R² = 0.01), indicating that temporal progression alone explained only a small proportion of the observed variation. Distribution of isolates by hospital wards and specimen types To characterize the epidemiological distribution of isolates within the hospital, their origin across wards and specimen sources was examined. During the study period, isolates originated from a wide range of hospital wards, although the largest proportions were observed in surgical, emergency, and critical care units. When intensive care–related wards were considered together (ICU, SICU, PICU, and CCU), they accounted for 37.9% of all identified isolates. The annual distribution of isolates across hospital wards during the study period is illustrated in Fig. 3. Annual distribution of Stenotrophomonas maltophilia isolates across hospital wards (2018–2024). Abbreviations: SW, Surgical Ward; ICU, Intensive Care Unit; MW, Medical Ward; CCU, Coronary Care Unit; VSW, Vascular Surgery Ward; SICU, Surgical Intensive Care Unit; ED, Emergency Department; PICU, Pediatric Intensive Care Unit; KU, Kidney Unit; CSW, Cardiac Surgery Ward; OP, Outpatient Department; OW, Oncology Ward; PW, Pediatric Ward; BMT, Bone Marrow Transplant Unit; CW, Cardiology Ward Regarding specimen sources, respiratory samples were the most common, particularly sputum (24.3%) and tracheal aspirates (18.6%). These were followed by wound specimens (18.1%) and blood cultures (15.8%). Other specimen types, including body fluids and urine, represented smaller proportions of the total isolates (Table 1). Table 1Distribution of S. maltophilia isolates by specimen typeSpecimen typeN (%)Sputum43 (24.3)Tracheal aspirate (TRAP)33 (18.6)Wound32 (18.1)Blood28 (15.8)Body fluid21 (11.9)Urine10 (5.6)Tissue5 (2.8)Other specimens*5 (2.8) Total 177 (100) *Other specimens include CSF, bedsore, corneal swab, catheter tip, and central-line specimens Distribution of S. maltophilia isolates by specimen type *Other specimens include CSF, bedsore, corneal swab, catheter tip, and central-line specimens Overall antimicrobial susceptibility profile Antimicrobial susceptibility testing for cotrimoxazole was available for almost all isolates included in the study (176/177, 99.4%). Susceptibility results for levofloxacin and ceftazidime were available for 168 (94.9%) and 149 (84.1%) isolates, respectively (Table 2). Table 2Overall antimicrobial susceptibility profile of S. maltophilia isolatesAntibioticSusceptible n (%)Non-susceptible1 n (%)Total Tested (N)2Cotrimoxazole (SXT)142 (80.2)34 (19.2)176Levofloxacin (LVX)139 (78.5)28 (15.8)168Ceftazidime (CAZ)76 (42.9)72 (40.7)149Combined non-susceptibility phenotype (CNP)NA10 (5.6)149*1 Non-susceptibility was defined as isolates categorized as resistant (R) or intermediate (I)2Susceptibility testing was not available for all isolates, as reflected in the reduced denominators reported* = isolates with complete susceptibility results for all three antibiotics (SXT, LVX, and CAZ) Overall antimicrobial susceptibility profile of S. maltophilia isolates 1 Non-susceptibility was defined as isolates categorized as resistant (R) or intermediate (I) 2Susceptibility testing was not available for all isolates, as reflected in the reduced denominators reported * = isolates with complete susceptibility results for all three antibiotics (SXT, LVX, and CAZ) Overall, cotrimoxazole showed the highest activity, with 142 isolates (80.2%) classified as susceptible. A similar susceptibility pattern was observed for levofloxacin, for which 139 of 168 isolates (78.5%) remained susceptible. In contrast, ceftazidime demonstrated considerably lower historical susceptibility (76 of 149 tested isolates, 42.9%), while 72 isolates (40.7%) were categorized as non-susceptible. Among the 149 isolates with complete susceptibility results for all three antimicrobial agents, combined non-susceptibility phenotype (CNP), defined as non-susceptibility to all three tested agents (cotrimoxazole, levofloxacin, and ceftazidime), was identified in 10 isolates (5.6%). The overall distribution of antimicrobial susceptibility is illustrated in Fig. 4. Overall antimicrobial susceptibility profile of Stenotrophomonas maltophilia isolates. Non-susceptibility was defined as isolates categorized as resistant (R) or intermediate (I). Susceptibility testing was available for 168/177 isolates for levofloxacin and 149/177 isolates for ceftazidime. Abbreviations: SXT, cotrimoxazole; LVX, levofloxacin; CAZ, ceftazidime To further explore differences in susceptibility patterns within the same isolates, paired comparisons were performed using McNemar’s test (Table 3). Non-susceptibility to ceftazidime was significantly more frequent than non-susceptibility to either cotrimoxazole (p < 0.001) or levofloxacin (p < 0.001). In contrast, no statistically significant difference was observed between cotrimoxazole and levofloxacin (p = 0.345), indicating broadly comparable activity of these two agents against the isolates included in this study. Table 3Paired comparison of antimicrobial non-susceptibility using McNemar’s testComparison (Drug A vs. Drug B)Non-susceptible to Drug A only nNon-susceptible to Drug B only nMcNemar P valueCotrimoxazole vs. Levofloxacin (N = 168)11170.345Cotrimoxazole vs. Ceftazidime (N = 149)5815 < 0.001* Levofloxacin vs. Ceftazidime (N = 149)5611 < 0.001* *P < 0.05 indicates statistical significance Paired comparison of antimicrobial non-susceptibility using McNemar’s test *P < 0.05 indicates statistical significance Temporal trends in antimicrobial non-susceptibility To assess whether antimicrobial susceptibility patterns changed during the study period, temporal trends in antimicrobial non-susceptibility were evaluated (Fig. 5). Temporal trends in antimicrobial non-susceptibility among Stenotrophomonas maltophilia isolates from 2018 to 2024 Non-susceptibility to ceftazidime demonstrated the greatest variability over time, increasing during 2020–2021 before declining in 2022 and showing moderate fluctuations thereafter. Cotrimoxazole non-susceptibility increased gradually from 2018 to 2021, decreased in 2022, and remained relatively stable in subsequent years. Non-susceptibility to levofloxacin remained generally lower throughout the study period. After reaching its highest level in 2021, levofloxacin non-susceptibility declined in the following years, with a slight increase observed in 2024. Overall, the most pronounced fluctuations in antimicrobial non-susceptibility occurred during the 2020–2021 interval. Stratified antibiotic resistance patterns To further explore heterogeneity in non-susceptibility patterns, antimicrobial non-susceptibility was stratified according to hospital ward and specimen type (Tables 4 and 5). Table 5Distribution of antibiotic non-susceptibility by specimen typeSpecimen type2SXT NS1 n (%)LVX NS n (%)CAZ NS n (%)Sputum7 (16.3%)5 (12.2%)16 (43.2%)Tracheal aspirate (TRAP)8 (24.2%)6 (19.4%)17 (63.0%)Wound8 (25.0%)8 (27.6%)10 (38.5%)Blood5 (17.9%)2 (7.1%)16 (64.0%)Body fluid4 (19.0%)4 (20.0%)6 (35.3%)Urine1 (10.0%)2 (22.2%)5 (62.5%)Tissue1 (20.0%)1 (20.0%)1 (20.0%)CSF (n = 1)0 (0.0%)0 (0.0%)1 (100.0%)1 Non-susceptibility (NS) was defined as resistant (R) or intermediate (I)2Specimen types with no resistant isolates (bed sore, catheter tip, central line, corneal swab) were excluded from the tableAbbreviations: SXT, cotrimoxazole; LVX, levofloxacin; and CAZ, ceftazidime Distribution of antibiotic non-susceptibility by specimen type 1 Non-susceptibility (NS) was defined as resistant (R) or intermediate (I) 2Specimen types with no resistant isolates (bed sore, catheter tip, central line, corneal swab) were excluded from the table Abbreviations: SXT, cotrimoxazole; LVX, levofloxacin; and CAZ, ceftazidime Across most hospital wards, non-susceptibility to ceftazidime was consistently higher than non-susceptibility to cotrimoxazole and levofloxacin. Higher frequencies of non-susceptibility were observed among isolates obtained from critical care and surgical wards, particularly intensive care units and several specialized hospital units. In contrast, isolates originating from pediatric, oncology, outpatient, and emergency settings demonstrated minimal or no resistance (Table 4). A similar distribution was observed across specimen types. Non-susceptibility to ceftazidime was more frequently detected among isolates obtained from invasive specimens, particularly blood and respiratory samples. In comparison, non-susceptibility to cotrimoxazole and levofloxacin remained relatively low across most specimen categories. Wound and urine specimens demonstrated comparatively higher levels of non-susceptibility to levofloxacin and cotrimoxazole than other specimen sources (Table 5). To further evaluate differences according to care setting, isolates were stratified into ICU-related and non-ICU groups. Comparison of antimicrobial non-susceptibility between ICU and non-ICU settings For stratified analysis, isolates obtained from patients admitted to the ICU, SICU, PICU, or CCU were grouped as ICU-related, whereas isolates from all other hospital wards were classified as non-ICU. As illustrated in Fig. 6, non-susceptibility rates to the individual antibiotics were numerically higher among ICU-related isolates than among non-ICU isolates. However, these differences did not reach statistical significance for cotrimoxazole (p = 0.051), levofloxacin (p = 0.399), or ceftazidime (p = 0.127). Comparison of antimicrobial non-susceptibility rates between ICU-related and non-ICU Stenotrophomonas maltophilia isolates. *Triple non-susceptibility or combined non-susceptibility phenotype (CNP) was defined as non-susceptibility to all three tested agents (SXT, LVX, and CAZ) and was calculated among isolates with complete susceptibility results for all three antibiotics (n = 149). Statistical analysis: Fisher’s exact test was used when expected cell counts were < 5; otherwise, the chi-square test was applied. Statistical significance was defined as P < 0.05. Abbreviations: SXT, cotrimoxazole; LVX, levofloxacin; CAZ, ceftazidime In contrast, the combined non-susceptibility phenotype (CNP) was significantly more frequent among ICU-related isolates compared with non-ICU isolates (21.7% vs. 3.2%, p = 0.004). Risk factors associated with antimicrobial non-susceptibility Associations between patient-related characteristics, healthcare exposures, and antimicrobial non-susceptibility among S. maltophilia isolates were examined (Table 6). Table 6Association between antimicrobial non-susceptibility and patient-related clinical characteristics among Stenotrophomonas maltophilia isolatesRisk factor (n)SXT NS1 n (%)P valueLVX NS n (%)P valueCAZ NS n (%)P valueGender (male) (n = 112)25 (22.3%)0.23522 (20.6%)0.8741 (44.6%)0.312Previous Hospitalization (n = 131)25 (19.1%)0.89324 (19.5%)0.15856 (48.7%)0.867Previous Intensive Care Unit Admission (ICU) (n = 55)11 (20.0%)0.87712 (22.6%)0.18526(50%)0.864Use of Immunosuppressive Drugs (n = 88)29 (21.6%)0.56714 (16.3%)0.86232 (41.0%) 0.05* Organ Transplantation (n = 6)2(33.3%)0.3251(16.7%)15(83.3%)0.080Antibiotic Use in the Previous Year (n = 105)24 (22.9%)0.16222 (22.2%) 0.008* 43 (47.8%)0.636Invasive medical device (n = 102)23 (22.5%)0.18819 (19.6%)0.23548 (57.1%) 0.014* Intubation/ventilation (n = 42)12 (28.6%)0.077(17.5%)0.87119 (55.9%)0.315Coronavirus Disease (COVID-19) (n = 17)3(17.6%)0.8636(35.3%) 0.030* 8(50.0%)0.887History of Past Surgeries (n = 112)21 (18.8%)0.83919 (17.9%)0.56749 (49.0%)0.8131 Non-susceptibility (NS) was defined as isolates categorized as resistant (R) or intermediate (I). Percentages were calculated among tested isolates. P values represent comparisons between patients with and without the specified risk factor. Fisher’s exact test was used when expected cell counts were < 5; otherwise, the chi-square test was applied. Statistical significance was defined as *p < 0.05. Abbreviations: SXT, cotrimoxazole; LVX, levofloxacin; CAZ, ceftazidime Association between antimicrobial non-susceptibility and patient-related clinical characteristics among Stenotrophomonas maltophilia isolates 26 (50%) 2 (33.3%) 1 (16.7%) 5 (83.3%) 7 (17.5%) 3 (17.6%) 6 (35.3%) 8 (50.0%) 1 Non-susceptibility (NS) was defined as isolates categorized as resistant (R) or intermediate (I). Percentages were calculated among tested isolates. P values represent comparisons between patients with and without the specified risk factor. Fisher’s exact test was used when expected cell counts were < 5; otherwise, the chi-square test was applied. Statistical significance was defined as *p < 0.05. Abbreviations: SXT, cotrimoxazole; LVX, levofloxacin; CAZ, ceftazidime Prior antibiotic exposure within the previous year was significantly associated with levofloxacin non-susceptibility (p = 0.008). COVID-19 infection was also associated with increased non-susceptibility to levofloxacin (p = 0.030). Non-susceptibility to ceftazidime was significantly associated with the presence of invasive medical devices (p = 0.014). Use of immunosuppressive therapy showed a borderline association with ceftazidime non-susceptibility (p = 0.050). No statistically significant associations were observed between antimicrobial non-susceptibility and gender, previous hospitalization, prior ICU admission, intubation or mechanical ventilation, organ transplantation, or history of previous surgeries (Table 6). Multivariable predictors of antimicrobial non-susceptibility Multivariable logistic regression analyses were performed to identify independent predictors of antimicrobial non-susceptibility among S. maltophilia isolates (Tables 7, 8 and 9). Table 7Multivariable logistic regression analysis of factors associated with cotrimoxazole non-susceptibility in Stenotrophomonas maltophilia IsolatesVariableaOR (95% CI)P valueMale sex (vs. female)0.95 (0.83–1.08)0.405Prior antibiotic use ≤ 1 year (yes vs. no)1.08 (0.95–1.22)0.248Invasive medical device (yes vs. no)0.96 (0.81–1.13)0.620Intubation / mechanical ventilation (yes vs. no)1.05 (0.86–1.28)0.618ICU admission (yes vs. no)1.12 (0.91–1.37)0.300aOR = adjusted odds ratio; CI = confidence interval. Non-susceptible outcome events included in this model: 34Odds ratios represent the association between each risk factor and antimicrobial non-susceptibility compared with the reference category (absence of the listed risk factor)Model calibration was assessed using the Hosmer–Lemeshow goodness-of-fit test (P > 0.05). Collinearity diagnostics indicated acceptable stability, with a maximum VIF of 2.85 and a condition index of 9.97Significant associations: *p < 0.05 Multivariable logistic regression analysis of factors associated with cotrimoxazole non-susceptibility in Stenotrophomonas maltophilia Isolates aOR = adjusted odds ratio; CI = confidence interval. Non-susceptible outcome events included in this model: 34 Odds ratios represent the association between each risk factor and antimicrobial non-susceptibility compared with the reference category (absence of the listed risk factor) Model calibration was assessed using the Hosmer–Lemeshow goodness-of-fit test (P > 0.05). Collinearity diagnostics indicated acceptable stability, with a maximum VIF of 2.85 and a condition index of 9.97 Significant associations: *p < 0.05 Table 8Multivariable logistic regression analysis of factors associated with levofloxacin non-susceptibility in Stenotrophomonas maltophilia IsolatesVariableaOR (95% CI)P valueMale sex (vs. female)0.88 (0.78–0.99) 0.028* Prior antibiotic use ≤ 1 year (yes vs. no)1.14 (1.00–1.29) 0.044* Previous hospitalization (yes vs. no)1.06 (0.92–1.21)0.419Previous ICU admission (yes vs. no)0.98 (0.86–1.12)0.752COVID-19 infection (yes vs. no)1.24 (1.03–1.50) 0.027* aOR = adjusted odds ratio; CI = confidence interval. Non-susceptible outcome events included in this model: 28Odds ratios represent the association between each risk factor and antimicrobial non-susceptibility compared with the reference category (absence of the listed risk factor)Model calibration was assessed using the Hosmer–Lemeshow goodness-of-fit test (P > 0.05). Collinearity diagnostics confirmed model stability with a maximum VIF of 1.21 and a condition index of 9.24Significant associations: *p < 0.05 Multivariable logistic regression analysis of factors associated with levofloxacin non-susceptibility in Stenotrophomonas maltophilia Isolates aOR = adjusted odds ratio; CI = confidence interval. Non-susceptible outcome events included in this model: 28 Odds ratios represent the association between each risk factor and antimicrobial non-susceptibility compared with the reference category (absence of the listed risk factor) Model calibration was assessed using the Hosmer–Lemeshow goodness-of-fit test (P > 0.05). Collinearity diagnostics confirmed model stability with a maximum VIF of 1.21 and a condition index of 9.24 Significant associations: *p < 0.05 Table 9Multivariable logistic regression analysis of factors associated with ceftazidime non-susceptibility in Stenotrophomonas maltophilia IsolatesVariableaOR (95% CI)P valueMale sex (vs. female)1.19 (1.00–1.41) 0.046* Immunosuppressive drugs (yes vs. no)0.85 (0.73–1.00) 0.046* Organ transplant (yes vs. no)1.40 (0.94–2.09)0.095Invasive medical device (yes vs. no)1.27 (1.08–1.50) 0.004* aOR = adjusted odds ratio; CI = confidence interval. Non-susceptible outcome events included in this model: 72Odds ratios represent the association between each risk factor and antimicrobial non-susceptibility compared with the reference category (absence of the listed risk factor)Model calibration was assessed using the Hosmer–Lemeshow goodness-of-fit test (P > 0.05). Collinearity diagnostics confirmed model stability with a maximum VIF of 1.11 and a condition index of 8.76Significant associations: *p < 0.05 Multivariable logistic regression analysis of factors associated with ceftazidime non-susceptibility in Stenotrophomonas maltophilia Isolates aOR = adjusted odds ratio; CI = confidence interval. Non-susceptible outcome events included in this model: 72 Odds ratios represent the association between each risk factor and antimicrobial non-susceptibility compared with the reference category (absence of the listed risk factor) Model calibration was assessed using the Hosmer–Lemeshow goodness-of-fit test (P > 0.05). Collinearity diagnostics confirmed model stability with a maximum VIF of 1.11 and a condition index of 8.76 Significant associations: *p < 0.05 For cotrimoxazole non-susceptibility, the multivariable model included sex, prior antibiotic exposure within the preceding year, intensive care unit (ICU) admission, invasive medical device use, and intubation or mechanical ventilation. After adjustment, none of these variables demonstrated a statistically significant independent association with cotrimoxazole resistance (p > 0.05 for all variables) (Table 7). For levofloxacin non-susceptibility, prior antibiotic exposure within the previous year and COVID-19 infection were independently associated with higher odds of resistance. Male sex also showed a modest but statistically significant association with levofloxacin resistance. In contrast, previous hospitalization and ICU admission were not significantly associated with resistance after adjustment (Table 8). For ceftazidime non-susceptibility, invasive medical device use emerged as the strongest independent predictor of non-susceptibility. Male sex was also associated with increased odds of ceftazidime non-susceptibility. In contrast, immunosuppressive drug use demonstrated a modest inverse association with non-susceptibility, while organ transplantation showed a trend toward increased non-susceptibility risk but did not reach statistical significance in the adjusted model (Table 9). All models demonstrated acceptable goodness of fit according to the Hosmer–Lemeshow test (p > 0.05). Because several adjusted odds ratios were close to unity, these associations should be interpreted as epidemiological signals rather than strong predictors of individual-level clinical risk. Discussion This study provides the first longitudinal evaluation of the epidemiology and antimicrobial resistance patterns of Stenotrophomonas maltophilia in a tertiary-care hospital in Palestine, offering insight into the local dynamics of this opportunistic pathogen within a Middle Eastern healthcare setting [1, 18, 19]. Over the seven-year study period, the organism accounted for a relatively small but clinically meaningful proportion of all cultures performed. Although the overall culture-based prevalence remained modest, the annual frequency of isolation fluctuated noticeably rather than showing a consistent upward trend. Respiratory specimens represented the predominant source of isolates, and more than one-third of isolates originated from ICU-related settings, reflecting the well-established association of S. maltophilia with critically ill and device-dependent patients [4, 20, 21]. In terms of antimicrobial susceptibility, cotrimoxazole and levofloxacin retained the highest activity, whereas ceftazidime showed substantially reduced effectiveness. Although non-susceptibility to individual antibiotics did not differ significantly between ICU and non-ICU settings, the combined non-susceptibility phenotype (CNP) was significantly more frequent among ICU isolates, suggesting that multidrug-resistant phenotypes may accumulate preferentially in critical care environments [4, 20]. In addition to describing local epidemiological patterns, these findings provide longitudinal surveillance data that help clarify the resistance landscape and clinical contexts associated with S. maltophilia isolation in tertiary-care hospitals. The temporal variability observed in this cohort is consistent with international surveillance reports indicating that the epidemiology of S. maltophilia often fluctuates across healthcare systems rather than expanding in a simple linear manner [18, 19, 22]. Observational studies from different regions have reported similar patterns, suggesting that antimicrobial consumption, infection-control practices, and patient case-mix may shape the ecological dynamics of this organism [1, 18, 19, 22, 23]. For example, long-term observational data from Europe have shown that hospital incidence may vary considerably over time and may correlate with systemic antibiotic use [22, 24]. Likewise, global systematic reviews and meta-analyses have shown that the frequency and resistance patterns of S. maltophilia vary substantially across regions and institutions, supporting the view that local ecological pressures, rather than uniform global expansion alone, influence its emergence in hospital settings [15, 25–28]. The fluctuations observed in the present study therefore align with broader international observations and underscore the value of sustained local surveillance. The predominance of respiratory isolates in this study is also consistent with previous reports identifying the respiratory tract as a major ecological niche of S. maltophilia in hospitalized patients [29]. Studies from tertiary-care hospitals and ICUs have consistently identified respiratory specimens as the most common source of isolation, particularly among patients requiring prolonged mechanical ventilation or extended hospitalization [23, 27, 29, 30]. In critically ill populations, respiratory colonization may precede the development of ventilator-associated pneumonia and other respiratory infections, especially in patients with chronic lung disease or prolonged exposure to invasive respiratory devices [20, 21]. These observations suggest that the respiratory predominance in our cohort reflects the known ecological behavior of the organism in ventilated and device-dependent patients. At the same time, respiratory isolation does not necessarily indicate true infection, which should be considered when interpreting these findings [4, 23]. Beyond isolate frequency, the temporal dynamics reveal significant annual fluctuations in antimicrobial non-susceptibility rates. This non-linear progression aligns with long-term international surveillance data suggesting that S. maltophilia resistance patterns are highly dynamic and responsive to shifting ecological pressures [1, 22, 31]. Notably, the observed peak in resistance during the 2020–2021 interval coincides with the unique clinical challenges of the COVID-19 pandemic, a period characterized by intensified empirical use of broad-spectrum antibiotics, which has been identified as a driver of resistance in several settings [12, 13, 32]. The subsequent stabilization or decline in resistance for some agents after 2021 further supports the hypothesis that local antimicrobial stewardship intensity and changing patient case-mix may be primary determinants of these temporal trends [22, 24]. The antimicrobial susceptibility profile observed in this study is broadly consistent with contemporary international surveillance data. Recent global analyses have shown that trimethoprim–sulfamethoxazole remains the most active antimicrobial agent against S. maltophilia, with susceptibility rates generally ranging from approximately 70% to 90% across different healthcare settings [15, 25, 26]. In contrast, resistance to β-lactam agents such as ceftazidime remains substantially higher because of intrinsic resistance mechanisms carried by the organism [6, 33, 34]. Data from the SENTRY antimicrobial surveillance program similarly showed preserved susceptibility to cotrimoxazole across isolates collected from Europe, Asia, and Latin America, while susceptibility to fluoroquinolones and β-lactam agents showed greater regional variability [35]. The susceptibility rates observed in the present cohort therefore fall within the range reported internationally [8, 15, 25, 26]. Taken together, these findings support the continued clinical relevance of cotrimoxazole as the preferred first-line therapy for most S. maltophilia infections, while also indicating that levofloxacin remains a useful alternative when cotrimoxazole cannot be used [36, 37]. The resistance patterns observed in this study can also be understood in light of the intrinsic multidrug resistance mechanisms characteristic of S. maltophilia. The organism possesses multiple chromosomally encoded resistance determinants, including the inducible β-lactamases L1 and L2 and multidrug efflux pumps that actively expel antimicrobial agents from the bacterial cell [6, 9, 31, 38–42]. These mechanisms reduce the activity of many β-lactam agents and contribute to the organism’s naturally resistant phenotype [33, 34, 43]. In addition, S. maltophilia can form biofilms on respiratory surfaces and medical devices, which promotes persistence within hospital environments and further limits antimicrobial penetration during treatment [5, 44–46]. Together, these intrinsic and adaptive mechanisms provide a plausible biological explanation for the resistance patterns observed among clinical isolates in this study. A notable finding of this study is that ICU-related isolates did not show significantly higher non-susceptibility to individual antibiotics than isolates from other hospital settings, yet multidrug-resistant phenotypes were significantly concentrated among ICU isolates. This pattern may be more clinically informative than isolated drug-specific differences. Intensive care units are characterized by extensive antimicrobial exposure, frequent invasive procedures, prolonged hospitalization, and high device utilization, all of which promote the emergence and persistence of resistant organisms [4, 20, 24, 47]. Previous observational studies have similarly reported that resistant S. maltophilia isolates occur more often among critically ill patients with prolonged hospital stays or prior exposure to broad-spectrum antibiotics [22, 24]. The clustering of the combined non-susceptibility phenotype (CNP) in ICU environments observed in the present study therefore mirrors patterns described in multiple healthcare systems worldwide [1, 21, 23]. The association between levofloxacin non-susceptibility and prior antibiotic exposure further supports the role of antimicrobial selection pressure in shaping resistance patterns. Several epidemiological investigations have reported similar associations between antimicrobial use and increased incidence or resistance of S. maltophilia, particularly in settings where fluoroquinolones are widely used [10, 11, 22, 24, 41]. The association with COVID-19 infection observed in this cohort should be interpreted carefully: the COVID-19 infection variable in the regression model functions as a contextual marker for a period of intensified antimicrobial pressure and profoundly altered healthcare practices, including increased empirical broad-spectrum antibiotic use, higher ICU occupancy, modified infection-control protocols, and a shifted patient case-mix, rather than as evidence of a direct biological effect of COVID-19 on S. maltophilia resistance. Multiple co-occurring exposures during this interval may have independently or synergistically contributed to the observed resistance patterns, and the independent contribution of COVID-19 infection per se cannot be isolated from the available data [12, 13]. The association observed should therefore be understood as reflecting the complex ecological and clinical context of the pandemic period rather than a causal mechanism. During the COVID-19 pandemic, widespread empirical use of fluoroquinolones, including levofloxacin, for suspected bacterial co-infections has been well documented and has been associated with increased selection pressure for levofloxacin-resistant nosocomial pathogens in multiple healthcare settings [12, 13, 32]. Several studies have reported a parallel increase in S. maltophilia incidence and resistance during pandemic-era periods, consistent with the antimicrobial stewardship challenges observed globally during this interval [1, 32]. From a clinical and antimicrobial stewardship perspective, the susceptibility data generated by this study provide locally relevant evidence to guide empirical treatment decisions. Based on the high in vitro activity observed for cotrimoxazole (80.2%) and levofloxacin (78.5%), cotrimoxazole should remain the preferred empirical first-line agent for suspected S. maltophilia infection in this setting, consistent with current IDSA guidance [36]. Levofloxacin represents a suitable alternative, particularly for patients who cannot tolerate cotrimoxazole or who have a history of prior exposure to SXT [36, 37]. In contrast, ceftazidime showed substantially lower in vitro activity (42.9% susceptibility), and empirical use of ceftazidime against S. maltophilia should be approached with considerable caution and guided by individual susceptibility data were available [16, 33]. Although ceftazidime has historically been used as part of combination regimens for severe S. maltophilia infections, the evolved CLSI and FDA positions on ceftazidime breakpoints for this organism further limit the interpretability of susceptibility results and underscore the need for careful therapeutic decision-making in affected patients [16, 37]. Similarly, the observed association between ceftazidime non-susceptibility and invasive medical device use is biologically plausible. S. maltophilia has a well-documented capacity to colonize abiotic surfaces such as endotracheal tubes and intravascular catheters, where biofilm formation facilitates persistent colonization and reduced antimicrobial susceptibility [5, 44–46]. Device-associated biofilms therefore likely contribute to the persistence of resistant isolates in patients requiring prolonged mechanical ventilation or vascular access [20, 21]. Furthermore, patients with invasive devices typically sustain higher cumulative antimicrobial exposure, which further selects for resistant phenotypes through ongoing selection pressure on device-associated organisms [10, 24]. Within biofilms, horizontal gene transfer may additionally accelerate the spread of resistance determinants among colonizing bacteria [31, 42]. Regarding the rationale for ceftazidime use in clinical practice, it is worth noting that although current CLSI guidelines no longer provide breakpoints for ceftazidime against S. maltophilia, this agent has historically been employed in combination therapy, particularly with SXT or LVX, for severe or refractory infections [37]. The intrinsic resistance of S. maltophilia to ceftazidime is largely attributable to the constitutively expressed chromosomal L1 zinc-dependent metallo-β-lactamase and the L2 serine-β-lactamase, which together confer broad hydrolytic activity against most β-lactam agents [33, 38]. These enzymes provide a compelling biological explanation for the substantially lower susceptibility rates observed for ceftazidime in our cohort compared with cotrimoxazole and levofloxacin. An additional clinically important observation is that cotrimoxazole non-susceptibility was not independently associated with any of the evaluated patient-related or healthcare-exposure variables in the multivariable analysis. This suggests that, within our cohort, reduced susceptibility to cotrimoxazole was less strongly influenced by the clinical and hospital factors that appeared to affect other agents, particularly levofloxacin and ceftazidime [11, 40]. From a practical standpoint, this relative stability strengthens the role of cotrimoxazole as the most reliable therapeutic option in this setting and supports its continued use as the preferred first-line agent for S. maltophilia infection. This interpretation is in line with current treatment guidance and recent global surveillance findings, both of which indicate that trimethoprim–sulfamethoxazole remains the agent with the most consistently preserved in vitro activity against S. maltophilia [35, 36]. Male sex was associated with resistance in some adjusted models; however, this finding should be interpreted cautiously, as sex may act as a proxy for unmeasured differences in disease severity, healthcare exposure, or care setting rather than representing a direct biological determinant [7]. More broadly, several statistically significant associations in the multivariable analyses demonstrated modest effect sizes, with certain adjusted odds ratios approaching unity (e.g., aOR values in the range of 0.88–1.27). While these associations achieved conventional statistical significance thresholds, their clinical relevance, defined as a meaningful difference in the probability of non-susceptibility at the individual patient level, warrants careful interpretation [15, 25]. Statistical significance in a relatively small cohort does not automatically translate into clinically actionable information. The absolute differences in non-susceptibility risk associated with factors such as prior antibiotic exposure, COVID-19 infection, and invasive device use are modest in magnitude, and should be considered in the context of the overall study limitations, including the sparse-outcome constraints of some models. These associations are best understood as hypothesis-generating observations that warrant confirmation in larger, prospective studies, rather than as definitive predictors to guide individual clinical decision-making or antimicrobial stewardship policy in isolation [10, 11]. To our knowledge, this study is the first longitudinal investigation of S. maltophilia epidemiology and antimicrobial resistance patterns conducted in a Palestinian tertiary-care hospital. Although studies from neighboring regions have examined susceptibility profiles of this organism, published data from Palestinian healthcare institutions remain extremely limited [8, 23, 30, 47]. Because antimicrobial resistance profiles may vary substantially across healthcare systems due to differences in antibiotic prescribing practices, infection-control measures, and patient populations, locally generated surveillance data are essential for informing empirical treatment strategies and antimicrobial stewardship policies [1, 2, 15, 28]. Overall, the findings suggest that antimicrobial resistance in S. maltophilia within this tertiary hospital setting is shaped by a complex interaction of antimicrobial exposure, patient-related clinical factors, and healthcare-environment pressures [6, 22, 24]. The preserved activity of cotrimoxazole and levofloxacin supports their continued relevance in the local context, whereas the clustering of multidrug-resistant phenotypes in ICU environments highlights the importance of cautious empirical antibiotic use and continuous microbiological monitoring [20, 29, 36]. Study limitations Several limitations should be considered. First, the retrospective single-center design may limit generalizability, as differences in patient populations, prescribing practices, and infection-control policies across institutions can influence S. maltophilia epidemiology and resistance patterns. Second, misclassification between colonization and true infection cannot be excluded. A considerable proportion of isolates came from respiratory specimens, sputum (24.3%) and tracheal aspirates (18.6%), for which no standardized clinical adjudication criteria (e.g., CDC/NHSN VAP definitions) were applied. In ICU settings, respiratory colonization by S. maltophilia is common and does not invariably indicate active infection. Third, susceptibility testing was limited to three agents, precluding broader characterization of resistance phenotypes or assessment of additional therapeutic options. Fourth, no molecular analyses were performed, including assessment of sul genes, efflux pump expression (SmeABC, SmeDEF, SmeVWX), quinolone resistance determinants, or whole-genome sequencing for clonal relatedness. The biological mechanisms underlying the observed non-susceptibility patterns and any clonal transmission within the hospital therefore cannot be determined. Future studies incorporating molecular and genomic approaches are strongly recommended. Fifth, reliance on categorical susceptibility data (S/I/R) rather than continuous MIC values precludes analysis of MIC creep. Ceftazidime data were derived from historical CLSI breakpoints applied uniformly for longitudinal comparison; these should be interpreted strictly as epidemiological surveillance data, not as clinical guidance, given that CLSI withdrew these breakpoints in 2023 due to intrinsic L1/L2 β-lactamase activity and poor clinical correlation. Sixth, susceptibility testing was performed exclusively with VITEK® 2 (AST-N222) without confirmatory broth microdilution. Automated platforms may show variable performance and systematic category errors for non-fermenters such as S. maltophilia, and some misclassification between susceptibility categories cannot be excluded. Future studies should incorporate reference MIC confirmation, particularly for isolates near breakpoint thresholds. Finally, detailed treatment data, coinfection profiles, and clinical outcomes were not systematically available, limiting assessment of the relationship between antimicrobial exposure, resistance development, and patient outcomes. The small number of CNP isolates also precluded multivariable regression for predictors of the combined non-susceptibility phenotype. Despite these constraints, the overall sample size is comparable to published single-center S. maltophilia surveillance studies, supporting the epidemiological relevance of the findings. Conclusion This study provides the first longitudinal characterization of the epidemiology, temporal trends, antimicrobial resistance patterns, and clinical predictors of Stenotrophomonas maltophilia in a Palestinian tertiary-care hospital, filling a meaningful gap in the regional literature on this increasingly important nosocomial pathogen. Over the seven-year study period, the organism maintained a modest but clinically relevant presence, with isolation predominantly from respiratory specimens and a consistent predilection for ICU-related settings, reflecting its well-established ecological affinity for critically ill and device-dependent patients. Cotrimoxazole and levofloxacin retained the highest in vitro activity and remain the most defensible empirical options in this setting. Ceftazidime susceptibility, reported here solely as a historical epidemiological observation, was substantially lower, a finding best understood through the lens of the organism’s intrinsic β-lactamase-mediated resistance. Notably, while non-susceptibility to individual agents did not differ significantly between care settings, the combined non-susceptibility phenotype was significantly concentrated among ICU-associated isolates, suggesting that critical care environments may function as selective reservoirs for more complex resistance profiles. Prior antibiotic exposure, COVID-19 infection, as a marker of intensified antimicrobial pressure, and invasive medical device use, emerged as the principal clinical predictors of non-susceptibility, findings that, carry direct implications for stewardship and infection-control prioritization. Taken together, these findings underscore a central message, resistance in S. maltophilia is not uniform, and its clinical significance is shaped as much by the healthcare environment and patient exposures as by the organism’s intrinsic biology. Despite being conducted at a single tertiary-care center; hospital-based surveillance studies remain essential for identifying local resistance dynamics that are often not captured by national or global antimicrobial resistance monitoring programs. Continued microbiological surveillance, combined with stewardship-guided antimicrobial use, will be important for monitoring evolving resistance patterns, guiding empiric antimicrobial therapy, informing infection-control, and optimizing the management of S. maltophilia infections in comparable tertiary-care environments. Acknowledgements The authors acknowledge the laboratory staff of ANNUH for their assistance and support. Abbreviations Antimicrobial susceptibility testing Ceftazidime Coronary Care Unit Confidence interval Clinical and Laboratory Standards Institute Coronavirus disease 2019 Intensive Care Unit Levofloxacin Non-susceptible Minimum inhibitory concentration Odds ratio Adjusted odds ratio Variance Inflation Factor Pediatric Intensive Care Unit Surgical Intensive Care Unit Cotrimoxazole (trimethoprim–sulfamethoxazole) Combined non-susceptibility phenotype (study-specific operational definition; non-susceptibility to all three tested agents) An-Najah National University Hospital Author contributions TA, WS, IH, and NJ collected the clinical data. AS and AA supervised the laboratory procedures and microbiological analyses. TA performed the statistical analyses and drafted the manuscript. TA and AA reviewed and revised the manuscript. All authors read and approved the final manuscript. Funding This research received no external funding. Data availability The datasets used and analyzed during the current study are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate The Institutional Review Boards (IRB) of An-Najah National University approved the study protocol, and the collected data were only used for clinical research purposes. Information was kept confidential and was not used for any other purpose. Patient information was coded to protect their identities. Because retrospective data were used, the IRB of An-Najah National University waived the requirement for informed consent. The authors confirmed that all methods were performed in accordance with the relevant guidelines and regulations. All procedures performed in this study were in accordance with the ethical standards of the institutional research committee and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. Consent for publication Not applicable. 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 datasets used and analyzed during the current study are available from the corresponding author on reasonable request.