Prevalence and antimicrobial susceptibility of multi drug-resistant bacteria isolates in adult intensive care units: a six-year retrospective study from a tertiary hospital in Mogadishu, Somalia Authors: Ali kutta Çelik, Mucahit Türkan, Tigad Abdisad Ali, Suad Abdikarim Isse, Fartuun Ahmed Mohmud, Lıban Ade Hussein, Ahmed Mohamed Ali Published: 2025 ABSTRACT AbstractBackgroundThis study aimed to determine the prevalence, distribution, and antimicrobial susceptibility patterns of multidrug-resistant (MDR) bacterial isolates in adult intensive care units (ICUs) at a tertiary care hospital in Mogadishu, Somalia.MethodsA retrospective study was conducted from January 2019 to August 2024. Blood, urine, and deep tracheal aspirate samples were collected from ICU patients aged ≥ 18 years. Microorganisms were identified using conventional methods at our hospital. Isolates were categorized as Gram-negative (GN), Gram-positive (GP), and Candida species. Subgroup Candida was not identified. Antimicrobial susceptibility testing was performed and interpreted according to Clinical and Laboratory Standards Institute (CLSI) guidelines using disk diffusion and E-test methods. Bacterial strains were classified according to their antibiotic resistance as MDR, possible extensively drug-resistant (XDR), or possible pandrug-resistant (PDR). Temporal trends in microbial distribution and resistance patterns were analyzed and compared with regional and global data.ResultsOf 2,185 isolates, 75.1% were GN bacteria, 18.9% GP bacteria, and 6% Candida species. Among bacterial isolates, 53.6% (n = 1,100) were MDR, with MDR rates of 52.3% in GN and 58.5% in GP bacteria. Most GN-MDR isolates belonged to the possible XDR category, with high carbapenem resistance rates, particularly in Klebsiella [2.8% (n = 16) K. oxytoca, 28.8% (n = 163) K. pneumoniae, and 68.4% (n = 387) Klebsiella spp.] and Acinetobacter [56.8% (n = 133) A. baumannii and 43.2% (n = 101) Acinetobacter spp.]. A marked increase in MDR prevalence and resistance rates was observed from 2022 onward. By 2024, Candida species became the second most common cause of bloodstream infections, surpassing GP bacteria. Colistin and tigecycline showed the highest efficacy against GN-MDR strains, while vancomycin, linezolid, daptomycin, and quinupristin-dalfopristin were most effective against GP-MDR isolates.ConclusionThe high prevalence of carbapenem-resistant Klebsiella and Acinetobacter in our ICUs exceeds global averages and underscores the urgent need for antimicrobial stewardship and robust infection control measures in Somalia. Although treatment options for GP-MDR bacteria are relatively more effective, management of GN-MDR infections remains limited and challenging.Keywords: Intensive care unit infections, Multidrug-resistant bacteria, Extensively drug-resistant, Pandrug-resistant, Carbapenem resistance, Antimicrobial susceptibility, Somalia Introduction The increasing global prevalence of antibiotic-resistant bacteria and the challenges in managing infections caused by these organisms have become major public health concerns. Intensive care units (ICUs) are particularly high-risk settings for the emergence and spread of multidrug-resistant (MDR) organisms due to several factors. These include frequent exposure to invasive procedures and interventions, such as intubation, sedation, and ulcer prophylaxis, which can compromise natural defense mechanisms like coughing and gastric acidity. In addition, the frequent—and sometimes inappropriate—use of broad-spectrum antibiotics further increases this risk. Additional factors, including parenteral nutrition and prior hospitalizations, also contribute to the problem [1, 2]. An international study reported an MDR prevalence rate of 47.8% among ICU infections in European hospitals, with an associated mortality rate of 36%. Of these, 20.5% were classified as possible extensively drug-resistant (XDR) and 0.4% as possible pandrug-resistant (PDR) [3]. Alarmingly, community-acquired MDR infections are also rising. Gram-negative MDR (GN-MDR) bacteria are of particular concern due to their association with increased morbidity, mortality, and healthcare costs [4].Compared to high-income countries, the burden of ICU-acquired infections and antimicrobial resistance is substantially higher in low- and middle-income countries (LMICs), where pathogen distribution also differs significantly [5]. In ICUs of LMICs, more than half of all infections are caused by Acinetobacter baumannii, Pseudomonas aeruginosa, and Klebsiella pneumoniae, whereas in ICUs of high-income countries, Gram-positive bacteria such as Staphylococcus and Enterococcus are more commonly observed [6]. Identifying the most frequently isolated pathogens is critical for guiding empirical therapy and informing infection control policies [7]. Data from local and regional studies are essential for developing context-specific clinical guidelines and effective antimicrobial stewardship interventions. This study aimed to evaluate the distribution of Gram-negative and Gram-positive bacterial isolates, determine the prevalence of MDR, possible XDR, and possible PDR strains, and assess their antimicrobial susceptibility patterns among adult ICU patients in a tertiary hospital in Mogadishu, Somalia, over a six-year period. Methods and materials This retrospective study included patients aged 18 years and older who were admitted to the adult intensive care units (ICUs) of Mogadishu Somalia-Turkey Recep Tayyip Erdoğan Training and Research Hospital between January 1, 2019, and August 30, 2024. Due to the limited number of ICU facilities in Mogadishu, ICU admissions were not strictly divided by medical specialty. All critically ill adult patients requiring intensive care were admitted to one of the three available ICUs: the General ICU, Emergency ICU, or Coronary ICU. Demographic data and microbiological culture results—including blood, urine, and deep tracheal aspirate samples—were retrospectively retrieved from the hospital information system. All bacteria isolated from clinical samples inoculated onto blood agar, chocolate agar, and MacConkey agar were identified using conventional microbiological methods. Gram-positive bacteria were characterized using catalase, coagulase, and bacitracin tests, while Gram-negative bacteria were identified with catalase, oxidase, Triple Sugar Iron (TSI) agar, Sulfur Indole Motility (SIM) test, Simmons Citrate agar, Christensen Urea agar, and Methyl Red/Voges-Proskauer tests. Candida species identification and antifungal susceptibility testing were not performed at our hospital. Therefore, all recovered Candida strains were reported as Candida spp. Only Candida isolates from blood cultures were considered clinically significant. Our microbiology laboratory is not accredited. Antimicrobial susceptibility testing was performed using the Kirby-Bauer disc diffusion method and the E-test (bioMérieux, France), in accordance with standard microbiological protocols. Susceptibility results were interpreted following the Clinical and Laboratory Standards Institute (CLSI) guidelines. To prevent duplication, if multiple culture results were obtained simultaneously from the same patient, only one isolate per organism was included in the analysis. Definitions of antimicrobial resistance Antimicrobial resistance was classified according to the criteria proposed by Magiorakos et al. [8]: MDR (Multidrug-Resistant): Non-susceptibility to at least one agent in three or more antimicrobial categories. MDR (Multidrug-Resistant): Non-susceptibility to at least one agent in three or more antimicrobial categories. XDR (Extensively Drug-Resistant): Non-susceptibility to at least one agent in all but two or fewer antimicrobial categories (i.e., susceptible to only one or two categories). XDR (Extensively Drug-Resistant): Non-susceptibility to at least one agent in all but two or fewer antimicrobial categories (i.e., susceptible to only one or two categories). PDR (Pandrug-Resistant): Non-susceptibility to all agents in all antimicrobial categories (i.e., no active agents remain effective). PDR (Pandrug-Resistant): Non-susceptibility to all agents in all antimicrobial categories (i.e., no active agents remain effective). Resistance among Enterococcus species was evaluated using the same classification. All methicillin-resistant Staphylococcus aureus (MRSA) strains were automatically categorized as MDR. Where the full range of antibiotics required for XDR or PDR classification was not tested, isolates were reported as possible XDR or possible PDR. Statistical analysis Descriptive statistics were used to summarize demographic, clinical, and laboratory features. Data normality was assessed using the Shapiro–Wilk test. Normally distributed variables are presented as mean ± standard deviation (SD), while non-normally distributed variables are presented as median (interquartile range [IQR]). Comparisons between groups were made using the independent-samples t-test or Mann–Whitney U test for continuous variables, and the chi-square or Fisher’s exact test for categorical variables. A p-value < 0.05 was considered statistically significant. All analyses were performed using IBM SPSS Statistics version 27.0 (IBM Corp., Armonk, NY, USA). Ethical approval This study was approved by the Clinical Research Ethics Committee of Mogadishu Somalia-Turkey Recep Tayyip Erdoğan Training and Research Hospital (Approval No: 18319). Results A total of 4,463 adult ICU patients were included in the study, of whom 35.4% (n = 1580) were female and 64.6% (n = 2883) were male. The mean age was 48.25 ± 20.60 years, with the largest age group being 21–40 years (38.1%, n = 1,699). Among the clinical samples collected, blood cultures were the most frequent (47.6%), followed by tracheal aspirates (29.7%), and urine cultures (22.7%) (Table 1). Positive culture results were obtained in 49.0% (n = 2,185) of samples. Of these, 75.1% (1,642/2,185) were Gram-negative (GN) bacteria, 18.9% (412/2,185) were Gram-positive (GP) bacteria, and 6.0% (131/2,185) were Candida spp. The most frequently isolated organisms included Klebsiella spp., Escherichia coli, Pseudomonas spp., Acinetobacter spp., Staphylococcus aureus, and coagulase-negative Staphylococci (CoNS) (Table 1). Table 1Demographic profile, specimen types, and culture outcomes of ICU patients (n = 4463)VariableFrequency(%)GenderFemale1580 (35.4)Male2883 (64.6)Age, Mean ± SD = 48.25 ± 20.60 YearsN (%)18 – 20 Years224 (5.1)21 – 40 Years1699 (38.1)41 – 60 Years1126 (25.2)61 –80 Years1090 (24.4)>80 Years324 (7.3)Years2019239 (5.4)2020323 (7.2)2021222 (5.0)2022715 (16.0)20231741 (39.0)20241223 (27.4)SpecimenBlood2124 (47.6)Deep tracheal aspirate1325 (29.7)Urine1014 (22.7)Culture resultsGrowth2185 (49.0)No growth2278 (51.0)Growing microorganismGram Negative1642 (75.1%)Gram Positive412 (18.9%)Fungus131 (6.0)MicroorganismKlebsiella spp.*566 (25.9%)Escherichia coli500 (22.9%)Pseudomonas spp.265 (12.1%)Acinetobacter spp.**234 (10.7%)Proteus spp.35 (1.6%)Coagulase-negative Staphylococcus (CoNS)186 (8.5%)Staphylococcus aureus181 (8.3%)Candida spp.131 (6.0%)Enterococcus spp.***28 (1.3%)Enterobacter spp.28 (1.3%)Streptococcus spp.14 (0.6%)Serratia spp.5 (0.2%)Citrobacter spp.4 (0.18%)Corynebacterium spp.2 (0.09%)Micrococcus spp.2 (0.09%)Pantoea spp.2 (0.09%)Morganella spp.1 (0.05%)Stenotrophomonas maltophilia1 (0.05%)* 2.8% (n = 16) K. oxytoca, 28.8% (n = 163) K. pneumoniae, 68.4% (n = 387) Klebsiella spp** 56.8% (n = 133) A. Baumannii, 43.2% (n = 101) Acinetobacter spp***Enterococcus sppContinuous variables such as age were summarized as mean ± standard deviation, while categorical variables (e.g., gender, specimen type, and culture results) were presented as frequencies and percentages. No inferential statistical tests were applied in this table Demographic profile, specimen types, and culture outcomes of ICU patients (n = 4463) * 2.8% (n = 16) K. oxytoca, 28.8% (n = 163) K. pneumoniae, 68.4% (n = 387) Klebsiella spp ** 56.8% (n = 133) A. Baumannii, 43.2% (n = 101) Acinetobacter spp ***Enterococcus spp Continuous variables such as age were summarized as mean ± standard deviation, while categorical variables (e.g., gender, specimen type, and culture results) were presented as frequencies and percentages. No inferential statistical tests were applied in this table Multidrug resistance Overall, 53.5% (1,100/2,054) of bacterial isolates were MDR. Among MDR isolates, 78.1% (n = 859) were GN and 21.9% (n = 241) were GP bacteria. The MDR rate was 52.3% among GN bacteria and 65.0% among GP bacteria. The annual distribution of MDR cases varied significantly over the study period (p < 0.001) (Table 2). Table 2Comparison of yearly MDR rates between gram-negative and gram-positive isolates in ICU patients (n = 1100)YearMDR-GN/Total GN Isolates (%)MDR-GP/Total GP Isolates (%)p-value201977/170 (45.3%)23/55 (41.8%)<0.001202083/215 (38.6%)53/89 (59.6%)202134/131 (26.0%)49/81 (60.5%)2022275/477 (57.6%)54/100 (54.0%)2023201/355 (56.6%)22(33 (66.0%)2024189/294 (64.3%)40/54 (74.1%)Abbreviations: MDR: Multidrug-resistant, GN: Gram-negative, GP: Gram-positive, ICU: Intensive Care UnitNote Data represent the annual distribution of MDR isolates and their proportions among total Gram-negative and Gram-positive cultures obtained from ICU patients between 2019 and 2024Pearson’s Chi-square test (χ²) for comparison of categorical variables (yearly MDR rate differences between Gram-negative and Gram-positive isolates) Comparison of yearly MDR rates between gram-negative and gram-positive isolates in ICU patients (n = 1100) Abbreviations: MDR: Multidrug-resistant, GN: Gram-negative, GP: Gram-positive, ICU: Intensive Care Unit Note Data represent the annual distribution of MDR isolates and their proportions among total Gram-negative and Gram-positive cultures obtained from ICU patients between 2019 and 2024 Pearson’s Chi-square test (χ²) for comparison of categorical variables (yearly MDR rate differences between Gram-negative and Gram-positive isolates) Among GN-MDR bacteria, colistin exhibited the highest susceptibility, while ciprofloxacin showed the lowest. Colistin susceptibility declined from 100% in 2019 to 94% in subsequent years, with a low resistance rate of 3.3% (p = 0.011). Carbapenem (meropenem) susceptibility peaked in 2023 at 27.5% and was lowest in 2021 at 8.8% (p < 0.001). Piperacillin–tazobactam resistance decreased from 98.2% in 2019 to 81.8% in 2024 (p < 0.001), while high resistance rates were also observed for cefoperazone–sulbactam (p < 0.001). Amikacin resistance averaged 64%, with annual fluctuations (p < 0.001). Tigecycline resistance increased sharply in 2024 (p < 0.001). Fosfomycin resistance rose from 50% in 2019 to 100% in 2024, though this trend was not statistically significant (p = 0.092). Ciprofloxacin resistance remained consistently above 90% throughout the study period. Among GP-MDR bacteria, resistance remained low for most antibiotics, except trimethoprim-sulfamethoxazole (SXT) and fusidic acid. Linezolid demonstrated the highest efficacy, while vancomycin resistance was observed in 2.1% of isolates. The detailed antibiotic susceptibility profiles for GN-MDR and GP-MDR bacteria are presented in Table 3. Table 3Changes in antibiotic susceptibility detected over the years in MDR-GN and MDR-GP bacteria (n = 1100)Antibiotic Susceptibility201920202021202220232024p-valueThe overall resistance rateMDR *(Gram Negative)(n = 859)Polymyxin E (Colistin)Sensitive77(100%)82(98.8%)32 (94.1%)271(99.3%)189(94.0%)178 (94.2%)0.01128/857 (3.3%)Resistant0(0.0%)1(1.2%)2(5.9%)2(0.7%)12(6.0%)11 (5.8%)MeropenemSensitive16(21.3%)10(14.7%)3(8.8%)31(11.5%)55(27.5%)29 (15.8%)<0.001685/829 (82.6%)Resistant59(78.7%)58(85.3%)31 (91.2%)238(88.5%)145(72.5%)154(84.2%)Piperacillin–TazobactamSensitive1(1.8%)5(9.8%)1(3.6%)8(3.5%)8(5.1%)2 (18.2%)<0.001508/533 (95.3%)Resistant55(98.2%)46(90.2%)27 (96.4%)222 (96.5%)149 (94.9%)9 (81.8%)Cefoperazone–SulbactamSensitive0(0.0%)2(33.3%)5 (41.7%)26 (17.3%)11 (10.2%)17 (18.7%)<0.001307/368 (83.4%)Resistant1(100%)4(66.7%)7 (58.3%)124 (82.7%)97 (89.8%)74 (81.3%)AmikacinSensitive41(54.7%)34(44.7%)12 (41.4%)149 (60.6%)32 (17.2%)39 (27.5%)<0.001547/854 (64%)Resistant34(45.3%)42(55.3%)17 (58.6%)97 (39.4%)154 (82.8%)103 (72.5%)TigecyclineSensitive51(100%)54(94.4%)18 (100%)12(92.3%)0(0.0%)3 (6.0%)<0.00154/192 (28.1%)Resistant0(0.0%)3(5.6%)0(0.0%)1(7.7%)3(100%)47 (94%)FosfomycinSensitive2(50.0%)2(28.6%)0(0.0%)1(5.9%)4(17.4%)0 (0.0%)0.09254/63 (85.7%)Resistant2(50%)5(71.4%)0(0.0%)16 (94.1%)19 (82.6%)12 (100%)Trimethoprim–Sulfamethoxazole (SXT)Sensitive0(0.0%)0(0.0%)0(0.0%)7 (12.5%)1(1.5%)2 (33.3%)<0.001117/127 (92.1%)Resistant0(0.0%)0(0.0%)0(0.0%)49 (87.2%)64 (98.5%)4 (66.7%)CiprofloxacinSensitive2(3.3%)4(7.8%)1(6.7%)11 (5.7%)5(2.6%)6 (3.3%)<0.001664/693 (95.8%)Resistant58(96.7%)47(92.2%)14 (93.3%)181 (94.3%)190 (97.4%)174 (96.7%)MDR* (Gram Positive)(n = 241)VancomycinSensitive21(100%)51(100%)49 (100%)49(96.1%)20(90.9%)39 (97.5%)0.0945/234(2.1%)Resistant0(0.0%)0(0.0%)0(0.0%)2(3.9%)2(9.1%)1 (2.5%)LinezolidSensitive23(100%)47(97.9%)46 (100%)29(96.7%)19(100%)33 (100%)<0.0012/199(1%)Resistant0(0.0%)1(2.1%)0(0.0%)1(3.3%)0(0.0%)0 (0.0%)DaptomycinSensitive22(100%)39(100%)46 (100%)46(93.9%)1(33.3%)0 (0.0%)<0.0015/159(3.1%)Resistant0(0.0%)0(0.0%)0(0.0%)3(6.1%)2(66.7%)0 (0.0%)Fusidic AcidSensitive11(50%)20(96%)23 (76.7%)31(66.0%)2(66.7%)0 (0.0%)<0.00148/135(35.5%)Resistant11(50.0%)9(31.0%)7 (23.3%)16(34.0%)1(33.3%)4 (100%)Quinupristin–DalfopristinSensitive11(100%)11(84.6%)37 (100%)28(90.3%)2(66.7%)0 (0.0%)<0.0016/95(6.3%)Resistant0(0.0%)2(15.4%)0(0.0%)3(9.7%)1(33.3%)0 (0.0%)Trimethoprim–Sulfamethoxazole (SXT)Sensitive0(0.0%)0(0.0%)0(0.0%)2(20.0%)2(22.2%)13 (70.2%)<0.00119/36(52.8%)Resistant0(0.0%)0(0.0%)0(0.0%)8(80.0%)7(77.8%)4 (30.8%)(MDR:multi-drugresistant, GN: gram-negative, GP: gram-positive)*Including; possibleXDR(extremely drug resistant) and possible PDR(pan-drug resistant) strainsPearson’s Chi-square test (χ²)–Used to compare antibiotic susceptibility (sensitive vs. resistant) across years and between MDR categories in ICU isolates Changes in antibiotic susceptibility detected over the years in MDR-GN and MDR-GP bacteria (n = 1100) MDR *(Gram Negative) (n = 859) 77 (100%) 82 (98.8%) 271 (99.3%) 189 (94.0%) 0 (0.0%) 1 (1.2%) 2 (5.9%) 2 (0.7%) 12 (6.0%) 16 (21.3%) 10 (14.7%) 3 (8.8%) 31 (11.5%) 55 (27.5%) 59 (78.7%) 58 (85.3%) 238 (88.5%) 145 (72.5%) 154 (84.2%) 1 (1.8%) 5 (9.8%) 1 (3.6%) 8 (3.5%) 8 (5.1%) 55 (98.2%) 46 (90.2%) 0 (0.0%) 2 (33.3%) 1 (100%) 4 (66.7%) 41 (54.7%) 34 (44.7%) 34 (45.3%) 42 (55.3%) 51 (100%) 54 (94.4%) 12 (92.3%) 0 (0.0%) 0 (0.0%) 3 (5.6%) 0 (0.0%) 1 (7.7%) 3 (100%) 2 (50.0%) 2 (28.6%) 0 (0.0%) 1 (5.9%) 4 (17.4%) 2 (50%) 5 (71.4%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (1.5%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 2 (3.3%) 4 (7.8%) 1 (6.7%) 5 (2.6%) 58 (96.7%) 47 (92.2%) MDR* (Gram Positive) (n = 241) 21 (100%) 51 (100%) 49 (96.1%) 20 (90.9%) 5/234 (2.1%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 2 (3.9%) 2 (9.1%) 23 (100%) 47 (97.9%) 29 (96.7%) 19 (100%) 2/199 (1%) 0 (0.0%) 1 (2.1%) 0 (0.0%) 1 (3.3%) 0 (0.0%) 22 (100%) 39 (100%) 46 (93.9%) 1 (33.3%) 5/159 (3.1%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 3 (6.1%) 2 (66.7%) 11 (50%) 20 (96%) 31 (66.0%) 2 (66.7%) 48/135 (35.5%) 11 (50.0%) 9 (31.0%) 16 (34.0%) 1 (33.3%) 11 (100%) 11 (84.6%) 28 (90.3%) 2 (66.7%) 6/95 (6.3%) 0 (0.0%) 2 (15.4%) 0 (0.0%) 3 (9.7%) 1 (33.3%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 2 (20.0%) 2 (22.2%) 19/36 (52.8%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 8 (80.0%) 7 (77.8%) (MDR:multi-drugresistant, GN: gram-negative, GP: gram-positive) *Including; possibleXDR(extremely drug resistant) and possible PDR(pan-drug resistant) strains Pearson’s Chi-square test (χ²) –Used to compare antibiotic susceptibility (sensitive vs. resistant) across years and between MDR categories in ICU isolates Trends in MDR pathogens MDR rates among GN bacteria exceeded 50% after 2022, while GP-MDR isolates showed a gradual upward trend (Fig. 1, Table 3). Among GN bacteria, the highest MDR rates were observed in Acinetobacter spp., followed by Klebsiella spp., Pseudomonas spp., and E. coli. Possible XDR strains were more common than MDR strains across GN bacteria, while possible PDR strains were observed exclusively in Klebsiella spp. (Table 4). Yearly prevalence of multidrug-resistant (MDR) Gram-negative (GN) bacterial isolates among adult ICU patients from 2019 to 2024 Table 4Resistance rates in Acinetobacter, Klebsiella, Pseudomonas and Escherichia coliMDRPossible XDRPossible PDRMDR + XDR + PDRAcinetobacter (n = 234)39 (16.7%)165 (70.5%)-204 (87.2%)Klebsiella (n = 566)57(10.1%)305 (53.9%)3(0.5%)365 (64.5%)Pseudomonas(n = 265)18(6.8%)73 (27.5%)-91 (34.3%)E.coli (n = 500)71(14.2%)92 (18.4%)-163 (32.6%)MDR presence (GP ve GN all bacteriae)TestGender1:Woman, 2:ManNone MDRChi-square1.089There is MDRP value0.78(E.coli: Escherichia coli, MDR: multidrugresistant, XDR: extremely drug resistant, PDR:pan drug resistant, GN:gram negative, GP: gram positive)Descriptive statistics onlyData are presented as frequencies and percentages of isolates classified as MDR, XDR, or PDR for eachbacterialspecies. No inferential test (such as Chi-square or Fisher’s exact) was applied in this table Resistance rates in Acinetobacter, Klebsiella, Pseudomonas and Escherichia coli Gender 1:Woman, 2:Man (E.coli: Escherichia coli, MDR: multidrugresistant, XDR: extremely drug resistant, PDR:pan drug resistant, GN:gram negative, GP: gram positive) Descriptive statistics only Data are presented as frequencies and percentages of isolates classified as MDR, XDR, or PDR for eachbacterialspecies. No inferential test (such as Chi-square or Fisher’s exact) was applied in this table Until 2022, Acinetobacter and Klebsiella were the predominant GN-MDR organisms. From 2022 onward, E. coli and Pseudomonas became increasingly prevalent, surpassing Acinetobacter as the dominant MDR pathogens (p < 0.001). MDR Proteus strains were identified only in 2022 and 2023 (Table 5, Fig. 2). Table 5Overall antibiotic resistance rates (%) in gram-positive and gram-negative bacteria from 2019 to 2024BacteriaAcinetobacterKlebsiellaPseudomonasE.coliGram-negative bacteriGram-positive bacteria*Antibiotics and resistance rates (%)Polymyxin E0,91,81,11,42,5-Meropenem91,756,747,328,950,8-Piperacillin-Tazobactam94,770,939,437,858,9-Sefoperazon-sulbactam75,060,928,826,943,4-Amikacin56,446,518,512,031,9-SXT(Trimethoprim-Sulfamethoxazole)100,088,4100,074,784,547,0Ciprofloxacin94,778,043,470,474,936.4Tigecycline17,923,5NT14,018,66.1Vancomycin-----1,0Linezolid-----0,6Daptomycin-----2,5Quinupristin-Dalfopristin-----5,2Fusidic Acid-----31,5(E.coli: Escherichia Coli, SXT: Trimethoprim sulfamethoxazole)* Staphylococcus, Streptococcus, EnterococcusDescriptive statistics only – Resistance rates (%) were summarized without inferential statistical testingNo hypothesis test (e.g., Chi-square, Fisher’s exact, or t-test) was applied in this table Overall antibiotic resistance rates (%) in gram-positive and gram-negative bacteria from 2019 to 2024 (E.coli: Escherichia Coli, SXT: Trimethoprim sulfamethoxazole) * Staphylococcus, Streptococcus, Enterococcus Descriptive statistics only – Resistance rates (%) were summarized without inferential statistical testing No hypothesis test (e.g., Chi-square, Fisher’s exact, or t-test) was applied in this table Yearly distribution of gram-negative bacterial species isolated from adult ICU patients between 2019 and 2024 Among GP bacteria, S. aureus was the most frequent MDR organism until 2021. From 2021 onward, CoNS became the predominant GP-MDR isolates (p < 0.001), while Corynebacterium was detected only in 2022. The comprehensive antimicrobial susceptibility profiles of GP-MDR and GN-MDR bacteria are summarized in Table 6. Table 6.Yearly distribution of multidrug-resistant gram-negative and gram-positive microorganisms isolated from ICU patients (2019–2024, n = 1100)Microorganism201920202021202220232024p-valueMDR (Gram Negative)(n = 859)Acinetobacter45(22.1%)34(16.7%)12 (5.9%)54(26.5%)20(9.8%)39 (19.1%)<0.001Klebsiella13(3.6%)29(7.9%)10 (2.7%)91(24.9%)144(39.5%)78 (21.4%)Pseudomonas4(4.4%)6(6.6%)7 (7.7%)46(50.5%)14(15.4%)78 (21.4%)Proteus0(0.0%)0(0.0%)0 (0.0%)7(46.7%)8(53.3%)0(0.0%)E. coli12(7.4%)13(8.0%)4(2.5%)66(40.5%)12(7.4%)56 (34.4%)Serratia0(0.0%)0(0.0%)0 (0.0%)0(0.0%)1(100%)0(0.0%)Enterobacter2(11.1%)1(5.6%)0 (0.0%)11 (61.1%)2(11.1%)2 (11.1%)Citrobacter1(100%)0(0.0%)0 (0.0%)0(0.0%)0(0.0%)0 (0.0%)Pantoae0(0.0%)0(0.0%)1 (100%)0(0.0%)0(0.0%)0 (0.0%)MDR (Gram Positive)(n = 241)S.aureus11(9.8%)37(33.0%)22 (19.6%)12(10.7%)21(18.8%)9 (8.0%)<0.001CoNS12(9.4%)16(12.5%)27 (21.1%)41(32.0%)1(0.8%)31 (24.2%)Corynebacterium0(0.0%)0(0.0%)0(0.0%)1(100%)0(0.0%)0(0.0%)E.coli: Escherichia coli, MDR: multidrug resistant, S. aureus: Staphylococcus aureus, CoNS: coagulase-negative StaphylococcusPearson’sChi-squaretest(χ²)– Used to compare yearly distributions of multidrug-resistant (MDR) isolates across different bacterial species (both Gram-negative and Gram-positive)– P-values (<0.001) indicate significant variation in MDR rates over time for several organisms Yearly distribution of multidrug-resistant gram-negative and gram-positive microorganisms isolated from ICU patients (2019–2024, n = 1100) MDR (Gram Negative) (n = 859) 45 (22.1%) 34 (16.7%) 54 (26.5%) 20 (9.8%) 13 (3.6%) 29 (7.9%) 91 (24.9%) 144 (39.5%) 4 (4.4%) 6 (6.6%) 46 (50.5%) 14 (15.4%) 0 (0.0%) 0 (0.0%) 7 (46.7%) 8 (53.3%) 0 (0.0%) 12 (7.4%) 13 (8.0%) 4 (2.5%) 66 (40.5%) 12 (7.4%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (100%) 0 (0.0%) 2 (11.1%) 1 (5.6%) 2 (11.1%) 1 (100%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) MDR (Gram Positive) (n = 241) 11 (9.8%) 37 (33.0%) 12 (10.7%) 21 (18.8%) 12 (9.4%) 16 (12.5%) 41 (32.0%) 1 (0.8%) 0 (0.0%) 0 (0.0%) 0 (0.0%) 1 (100%) 0 (0.0%) 0 (0.0%) E.coli: Escherichia coli, MDR: multidrug resistant, S. aureus: Staphylococcus aureus, CoNS: coagulase-negative Staphylococcus Pearson’sChi-squaretest(χ²) – Used to compare yearly distributions of multidrug-resistant (MDR) isolates across different bacterial species (both Gram-negative and Gram-positive) – P-values (<0.001) indicate significant variation in MDR rates over time for several organisms Discussion Antimicrobial resistance (AMR) has emerged as a critical threat in hospital settings worldwide, particularly in intensive care units (ICUs), where patients are highly susceptible to infections by multidrug-resistant (MDR) pathogens. The most prominent multidrug-resistant (MDR) bacteria encompass extended-spectrum β-lactamase (ESBL) and carbapenemase-producing Enterobacterales (CRE), carbapenem-resistant Acinetobacter baumannii (CRAB), carbapenem-resistant Pseudomonas aeruginosa (CRPA), methicillin-resistant Staphylococcus aureus (MRSA), and vancomycin-resistant Enterococcus spp. (VRE) [9]. Globally, AMR is escalating rapidly, especially in low- and middle-income countries (LMICs), due to unregulated antimicrobial usage in both human and animal health sectors [10]. A particularly alarming trend is the increasing carbapenem resistance among Enterobacteriaceae, especially Klebsiella pneumoniae, where mortality can reach 75% in carbapenem-resistant infections [11, 12]. For instance, a study from China documented a rise in carbapenem-resistant K. pneumoniae (CRKP) from 3% in 2005 to over 44% in recent years [13]. The global emergence of carbapenemase-producing Gram-negative (GN) bacteria has rendered many last-line antibiotics ineffective, forcing clinicians to resort to older, nephrotoxic agents like colistin and fosfomycin. In Sub-Saharan Africa, AMR is estimated to contribute to 23.5 deaths per 100,000 population [14]. Somalia ranks among the top ten countries globally in terms of AMR-attributable mortality, with AMR being the second leading cause of infectious disease-related deaths [15]. Our six-year retrospective analysis provides important insight into AMR trends within ICUs in Somalia. The study population was predominantly male (64.6%) with a mean age of 48.25 years. Pathogenic organisms were identified in 49% of culture-positive cases, with GN bacteria accounting for the majority (75.1%), followed by Gram-positive (GP) organisms (18.9%) and Candida (6.0%). This distribution aligns with findings from other regions [16, 17], although our study notably identified E. coli as the second most common GN pathogen after Klebsiella spp., ahead of Acinetobacter spp., which is typically cited as the leading ICU pathogen [6, 17–19]. We identified 1,100 MDR bacterial isolates between 2019 and 2024, 78.1% of which were GN bacteria. The overall MDR prevalence among GN organisms was 52.3%, with the highest annual MDR rates recorded in 2021. A marked shift occurred in 2022, with MDR GN rates exceeding 50% and GP-MDR rates also showing an upward trend. Alarmingly, our GN-MDR resistance rates exceed those reported in several international studies [3, 20–22], highlighting a significant regional burden. Our findings also reveal concerning levels of carbapenem resistance. Meropenem resistance was detected in 50.8% of all GN isolates and in 82.6% of GN-MDR isolates—substantially higher than the global average (44%) but consistent with reports from Europe and Russia [1]. This is particularly concerning given the limited availability of newer agents in Somalia. Despite updated 2024 IDSA guidelines recommending newer antimicrobials such as ceftazidime-avibactam, ceftolozane-tazobactam, and cefiderocol [23], none are currently accessible in Somalia. In the literature, the dominant carbapenemase types in Somalia are metallo-β-lactamases and OXA type enzymes [24]. For this reason, local and national measures should be taken in the fight against multidrug-resistant strains. Among GP-MDR isolates, resistance to most antibitics remained below 35%, except for trimethoprim-sulfamethoxazole (SXT), which showed higher resistance rates. Linezolid, vancomycin, daptomycin, and quinupristin-dalfopristin maintained excellent activity against methicillin-resistant S. aureus and CoNS. Our vancomycin resistance rate among GP bacteria overall was 1.0%, and no resistance was observed for daptomycin or linezolid—findings that compare favorably with data from Egypt and other regional studies [25]. Colistin (2.5%) and tigecycline (17.9%) were the most effective agents against GN isolates, likely due to their restricted use in Somalia owing to toxicity concerns. Our results were consistent with a recent Kenyan study [26], although colistin and amikacin resistance rates in our study were notably lower. Additionally, our Pseudomonas isolates demonstrated higher susceptibility than regional averages [27]. A 2023 Somali review categorized resistance levels as low (<1%), moderate (1–5%), or high (>5%) [28]. While carbapenem resistance in E. coli was previously reported as low, our study found a 28.9% meropenem resistance rate—suggesting a rapid and possibly underestimated increase in resistance. Similarly, carbapenem resistance in Klebsiella was 28.7% in our cohort, aligning with global data but still below the 70.6% reported in Greece and 66% in South Asia [29]. Previous Somali studies have reported Enterobacteriaceae resistance between 2.8–13.9% [30], but our findings suggest that this rate has been increasing over the years. We believe that several factors may contribute to this situation. The widespread availability of over-the-counter antibiotics encourages inappropriate antibiotic use, while the lack of an effective surveillance system and the absence of infection control committees or infection control physicians, except at one hospital in Mogadishu, makes early diagnosis of resistant strains and isolation of infected patients more difficult. Furthermore, the limited use of hand antiseptics and personal protective equipment due to economic constraints increases this risk. The fact that female healthcare workers are forced to work in traditional clothing that is prone to cross-contamination also contributes to this problem. Moreover, patients often discontinue treatment early for financial reasons, leading to incomplete recovery from infections and the selection of resistant strains. Furthermore, we believe that the tendency of doctors to initiate antibiotic treatment earlier and more aggressively, observed not only in Somalia but globally following the COVID-19 pandemic, has further exacerbated this problem. Tigecycline resistance remained at 18.6% in our cohort, which is consistent with prior reports [31]. Despite the high prevalence of GP-MDR organisms (notably Staphylococcus spp., with a 65.4% MDR rate), resistance to first-line antibiotics remained low. This suggests that treatment options for GP pathogens are still effective in Somalia, offering a window of opportunity for stewardship interventions. Finally, our findings underscore the urgent need for comprehensive national antimicrobial stewardship programs and robust infection control practices to combat the escalating threat of AMR, particularly in ICU settings. The absence of vancomycin-resistant Enterococcus and the preserved susceptibility to linezolid and daptomycin provide a rare but valuable clinical advantage that should be safeguarded. Limitations This study has several limitations. First, it was conducted in a single tertiary care hospital in Mogadishu, which may limit the generalizability of the findings to other healthcare settings in Somalia or the broader region. Second, the absence of advanced molecular techniques, such as polymerase chain reaction (PCR), for detecting resistance genes restricted the identification of specific carbapenemase-type mechanisms. Third, Candida species were not identified, and antifungal susceptibility testing was not performed, preventing determination of the dominant Candida species and the prevalence of azole resistance. So that the culture result named Candida spp. Fourth, the retrospective design may have led to incomplete data or classification errors in microbiological records. Fifth, patient-level data on outcomes, comorbidities, prior antibiotic use, and ICU length of stay were not available, limiting the assessment of the clinical impact of MDR infections. Finally, antimicrobial susceptibility testing was limited to antibiotics available locally, and internationally recommended newer agents could not be evaluated due to their unavailability. Conclusion This six-year retrospective analysis provides the first comprehensive overview of antimicrobial resistance trends among ICU patients in Somalia. Our study identified a marked rise in multidrug-resistant (MDR) and carbapenem-resistant Gram-negative bacteria, particularly Klebsiella and Acinetobacter species, with resistance rates exceeding those previously reported in regional and global datasets, especially since 2022. Possible pandrug-resistant (PDR) strains were detected only in Klebsiella. Notably, the incidence of candidemia increased steadily and, for the first time, surpassed Gram-positive bacteria as the second most common cause of bloodstream infections in 2024—an emerging issue not previously documented in Somali healthcare settings. Despite a moderate rise in resistance among Gram-positive organisms, several antimicrobial agents remain effective, offering therapeutic options. These findings underscore the urgent need for national antimicrobial stewardship initiatives, reinforced infection control measures, and continuous microbiological surveillance. Future multicenter, prospective studies integrating molecular resistance profiling and clinical outcomes are essential to guide evidence-based interventions and policymaking in Somalia. Given the high resistance rates to carbapenems, piperacillin-tazobactam, and fosfomycin, their empirical use should be restricted, as should both empirical and prophylactic use of fluoroquinolones. Agents with lower resistance rates, such as colistin and tigecycline, should be reserved exclusively for confirmed cases and used only under the supervision of an infectious diseases specialist. Additionally, the procurement and availability of newer-generation antibiotics—such as ceftazidime-avibactam, cefiderocol, and sulbactam-durlobactam—should be expedited for use in selected, culture-proven cases. Acknowledgements Not applicable Abbreviations Antimicrobial Resistance Multidrug-Resistant Extensively Drug-Resistant Pan-Drug-Resistant Intensive Care Unit Gram-Negative Gram-Positive Methicillin-Resistant Staphylococcus aureus Coagulase-negative Staphylococci Vancomycin-Resistant Enterococcus Extended-Spectrum Beta-Lactamase Carbapenemase-producing Enterobacterales Carbapenem-resistant K. pneumoniae Carbapenem-Resistant Acinetobacter baumannii Carbapenem-Resistant Pseudomonas aeruginosa Trimethoprim-sulfamethoxazole Carbapenem-Resistant Acinetobacter baumannii Low- and middle-income countries Clinical and Laboratory Standards Institute Polymerase chain reaction Triple Sugar Iron agar Sulfur Indole Motility test Author contributions All authors (Mucahit Türkan, Tigad Abdisad Ali, Suad Abdikarim Isse, Fartuun Ahmed Mohmud, Liban Ade Hussein, and Fartun Ahmed Mohmud) made substantial contributions to the work, including study conception, design, data acquisition, analysis, and interpretation. All authors participated in drafting, revising, and critically reviewing the manuscript, approved the final version for publication, and agreed to submit the work to BMC Infectious Diseases. Funding This study did not receive any specific assistance from funding agencies or sectors. Data availability The datasets during and/or analysed during the current study available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate This study was approved by the Clinical Research Ethics Committee of Mogadishu Somalia-Turkey Recep Tayyip Erdoğan Training and Research Hospital (Approval No: 18319). As this was a retrospective study using anonymized clinical and laboratory data collected during routine patient care, no direct contact with participants occurred. Therefore, the requirement for written informed consent was waived by the Ethics Committee. All procedures were performed in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Declaration of Helsinki and its later amendments. 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 during and/or analysed during the current study available from the corresponding author on reasonable request.