Bacteriological profile and multidrug resistance of Gram-negative bacteria in wound infections at a secondary healthcare facility in Ghana Authors: Anthony Agborgbor, Lukeman Abubakar, Hilda Agyei, George Lassey, Anabel Acheampong, Prince-Charles Kudzordzi, Prince Nyarko, Kenneth Bentum Otabil Published: 2026 ABSTRACT BackgroundGlobally, antimicrobial resistance is a major clinical and therapeutic challenge, causing significant morbidity and mortality, with wounds being a major reservoir of multi-drug resistant bacteria. Data on the prevalence of bacterial resistance in secondary health care settings, particularly in the middle belt of Ghana, remain limited, which hampers effective treatment and infection control measures. The aim of the study was to determine the bacteriological profile and the prevalence of multidrug-resistant Gram-negative bacteria in wounds at the Holy Family Hospital in Berekum, a secondary health care facility in the Central Belt of Ghana.MethodsThe cross-sectional study was performed between February and June 2021 and 210 patients (50.5% female) with wound infections were systematically recruited. Wound swabs were taken from participants and transported immediately to the laboratory. Bacteria were isolated and identified using standard culture and biochemical tests. Antimicrobial susceptibility was performed on Mueller Hinton agar using Kirby-Bauer disk diffusion method according to standards by the Clinical and Laboratory Standards Institute. Isolates resistant to one or more antibiotics in three or more antimicrobial classes were classified as multi-drug resistant.ResultsA total of 210 swabs were taken from study participants and 173 (82.4%) of these showed bacterial growth with 143/210 (68.1%) of being Gram-negative bacteria and 30/210 (14.3%) being Gram-positive bacteria. The major Gram-negative bacteria isolated were 24.8% Pseudomonas spp., 17.3% Citrobacter spp., 13.3% Enterobacter spp., and 10.4% Proteus mirabilis whilst Staphylococcus aureus (17.3%) was the only Gram-positive bacteria isolated. The prevalence of antibiotic resistance was high among the isolates tested being 90.6% for cefuroxime, 86.9% for ceftriaxone, 84.2% for tetracycline, 81.1% for ampicillin and cefotaxime, and 74.8% for meropenem. The prevalence of multidrug resistant Gram-negative bacteria was 136/143 (95.1%) with amikacin and ciprofloxacin being the most effective antibiotics with sensitivity rates at 83.6% and 57.5%, respectively.ConclusionsOur study showed that the most common Gram-negative bacteria found in infected wounds was Pseudomonas spp. We also observed a high prevalence of multidrug resistance in the Gram-negative bacteria tested against selected common antibiotics used in Ghana. The high resistance to carbapenems and third generation cephalosporins (cefotaxime and ceftriaxone) considered to be the antibiotics of last resort was a particularly worrying finding. This indicates the need to review first-line and last-line antibiotic regimens and to improve and maintain antimicrobial surveillance in secondary health care settings. In addition, empirical wound healing should be monitored on a regular basis to assess the development of resistance and to ensure that wound healing is still effective against wound infections.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12866-026-04813-2. Background Wound infections pose a major threat to global health, especially in countries with limited resources and inadequate health care infrastructure [1]. The diversity of bacterial pathogens associated with wounds is broad and constantly changing and depends on the immune system of the patient, the type of wound and environmental conditions [2]. These infections can delay healing of wounds, increase morbidity and, in severe cases, lead to sepsis and mortality [3]. For example, infections caused by wounds are among the most common infections acquired in hospitals and account for 70–80% of all fatalities [4]. The increasing prevalence of antimicrobial resistance (AMR) is making wound infection management even more difficult [5]. In 2017, the World Health Organization published its first list of global priority pathogens, identifying Gram-negative bacteria as the most critical pathogens, because they are resistant to even last-resort antibiotics [6]. Studies have subsequently confirmed that Gram-negative bacteria are the most common and resistant group of bacteria in infected wounds [7]. The impact of antimicrobial resistance is particularly severe in sub-Saharan Africa, where access to advanced diagnosis and effective treatment is often limited [8]. In 2019, an estimated 1.27 million deaths worldwide were attributed to bacterial AMR, with West Sub-Saharan Africa accounting for the highest number of deaths attributed to AMR [9]. Data on the bacteria in wounds and their antibiotic resistance patterns are critical for effective wound treatment, development of treatment guidelines and public health strategies [4, 10]. The National Action Plan on Antimicrobial Resistance in Ghana thus highlights the need for antimicrobial resistance surveillance at different levels of healthcare [11]. Unfortunately, most surveillance efforts and research studies are concentrated in tertiary healthcare institutions [10, 12–15]. However, primary and secondary health care institutions remain key in the management of antimicrobial resistance, playing a key role in infection prevention and serving as key entry points for patients who may later require hospitalization in tertiary and quaternary care. The aim of the study was to determine the bacteriological profile and multidrug resistance of Gram-negative bacteria in wounds in a secondary health care facility in the middle belt of Ghana. The findings of this study can inform infection control protocols and help reduce the transmission of AMR from secondary to tertiary care, ultimately reinforcing national infection prevention strategies. Methods Study area The study was performed in the Berekum Holy Family Hospital in the Berekum Municipality in the Bono Region of Ghana. Berekum is the capital of the Berekum Municipality in the Bono Region of Ghana with an estimated population of 62,364 and a total land size of\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\:{955km}^{2}$$\end{document} [16]. It is located between latitudes 7°15’ and 8°00’ North and Longitudes 2°25’ and 2°50’ West. It is bordered to the North-East and North-West by Tain Municipal and Berekum West District, respectively. It is bordered to the South-West by Dormaa East Municipal and to the South-East by the Sunyani West District [16]. The Berekum Holy Family Hospital is a secondary health facility which serves as the main health facility for the Berekum municipality. The hospital provides both outpatient and inpatients services including general medicine, surgery, internal medicine, obstetrics and gynaecology, pediatrics, dental, ophthalmology, dermatology, orthopedic and urology. The Hospital has 162 inpatient beds and 100 bed capacity maternity unit and averages an annual outpatient attendance of 100,000 and 2,700 deliveries. Its patient volume and community reach make it an appropriate setting for assessing antimicrobial resistance in a primary care context. The microbiology department of the Hospital is well equipped and has the human resource to perform microbial and biochemical analysis. Ethical approval and consent Ethical approval (certificate No. CHRE/CA/12/012) for this research was obtained from the Committee for Human Research and Ethics at the University of Energy and Natural Resources, Sunyani, Ghana. Written informed consent was obtained from all participants or their parents/legal guardians (in the case of minors <18 years), and the participants were free to withdraw from the study at any time without consequence. In addition, assent was sought for minors aged 10–17 years to affirm their willingness to participate in the study. All study procedures conformed to the principles of protection of human subjects of research embodied in the Declaration of Helsinki [17]. Study design and participant recruitment The study adopted a cross-sectional design where participants were recruited from patients who presented with clinical signs of wound infections (both inpatients and outpatients) in the Berekum Holy Family Hospital. To account for seasonal variations in bacterial infections and to allow for sufficient sample collection to analyze resistance patterns effectively the sampling was performed from 1st February to 30th June 2021. The purpose and procedures of the study was discussed with the participants and written informed consent was taken prior to recruitment and wound sampling. Demographic information as well as history of previous infections and antibiotic usage were obtained using a structured questionnaire developed for this study (see Additional file 1). This questionnaire was pretested in a small pilot prior to the start of this study at the same health facility and only minor edits on clarity of questionnaires were necessitated and effected. The questionnaire was administered by members of the research team who had received a one-week training on questionnaire administration at the Consortium for Neglected Tropical Diseases and One Health at the University of Energy and Natural Resources, Sunyani, Ghana. The study population comprised of inpatients and outpatients who had clinical signs of wound infections. Inpatients were taken from the surgical ward (SW), maternity ward (MAT), emergency ward (EW), male ward (MW), female ward (FW) and neonatal intensive care unit (NICU) and outpatients department (OPD) of study hospital. Inclusion and exclusion criteria Patients were considered eligible for inclusion in the study if they presented with clinically diagnosed wounds as confirmed by clinical evaluation (e.g., presence of pus discharge, erythema, warmth, swelling, or pain), had not taken any antibiotics in the 48 h prior, and gave their informed consent to be included in the study. Classification of wounds and collection of samples Classification of wounds were made based on the etiologies of injury sustained. Thus, wounds were categorized as post-surgical, diabetic, burns and others (animal bites, cuts, laceration wounds and abrasion wounds) [18]. Wound swabs were taken from patients with clinical signs of wound infections following established protocols [19]. Briefly, the wounds were cleaned with sterile, normal saline-soaked gauze to remove debris and any possible contaminants. The tip of the cotton swab was moistened with 0.9% sodium chloride and rotated through a one-centimeter square area of visible tissue for a few seconds away from the edges of the cut. Two samples were taken at random from each patient as a control to determine whether contamination occurred during the collection procedure. The samples collected were then sent immediately to the Holy Family Hospital Microbiology Laboratory for bacterial culture, Gram staining, microscopy, biochemical and sensitivity testing. Bacteria culture, biochemical analysis and identification Using the streak-plating technique, wound swabs were inoculated onto blood agar and MacConkey agar [20]. All cultures were incubated aerobically at 37 °C overnight and observations were made for the identification of the species of bacteria [20]. The preliminary identification of the bacteria was based on the growth characteristics of the organisms observed on each plate [20]. These included haemolysis on blood agar, changes in the appearance of the different media and enzymatic activity [20]. Gram staining, microscopy and biochemical tests were then carried out to further confirm the findings according to established procedures [20]. The choice of biochemical tests was determined by the results of the Gram-reaction [20]. Gram-positive bacteria were identified with coagulase and catalase tests as well as their growth characteristics on blood agar. Gram-negatives were identified with series of biochemical tests namely indole, citrate utilization, urease, Triple Sugar Iron (TSI), oxidase and motility tests. Given that wound swabs were inoculated on blood agar and MacConkey agar using the streak plate technique and cultures were incubated aerobically at 37 °C overnight, our epidemiologic estimates reflect aerobically recoverable flora under swab-based sampling. Antimicrobial susceptibility testing Antimicrobial susceptibility testing was performed on Mueller Hinton agar using Kirby-Bauer disk diffusion method [21]. Morphologically identical pure bacterial colonies from overnight cultures were suspended in 5 ml nutrient broth and incubated for 4–6 h at 37 °C [20, 22]. This 4–6 h of incubation was necessary because the initial overnight culture likely contained bacteria in the stationary phase, where cell division has slowed due to nutrient depletion and waste accumulation [23]. Suspending these colonies in fresh nutrient broth and incubating for a specific duration of 4–6 h at the optimal temperature of 37 °C allows the culture to enter the log phase improving the accuracy and reproducibility of the AST [23]. The turbidity of the suspension was equilibrated to match with 0.5 McFarland standards [20]. The bacterial suspensions were then inoculated on the surface of the Mueller Hinton agar using a sterile cotton swab to test for antimicrobial susceptibility of the isolates [20]. The antibiotics tested against the Gram-negative bacteria isolates were ampicillin (10 µg), tetracycline (30 µg), cotrimoxazole (25 µg), gentamicin (10 µg), cefuroxime (30 µg), chloramphenicol (10 µg), ceftriaxone (30 µg), cefotaxime (30 µg), ciprofloxacin (5 µg), amikacin (30 µg), meropenem (10 µg). Antibiotics disks were aseptically placed on the surface of inoculated agar and incubated at 37 °C for 18–24 h [21]. After incubation, the diameters of the zone of inhibition were measured and interpreted based on the criteria set forth by the Clinical and Laboratory Standard Institute [21]. MDR was defined as resistance to three or more antimicrobial categories [21]. Quality control For bacteria cultures, control strains of Pseudomonas aeruginosa (NCTC 10662), Escherichia coli (NCTC 10418) and Staphylococcus aureus (NCTC 6571) obtained from the Microbiology Laboratory of the Holy Family Hospital were used as control bacteria [21]. These strains are used for routine quality control in the facility. In this study, the organisms were used to control for the quality of our freshly prepared culture media, reagents and culture and AST procedures. Two swabs were taken from each patient to ascertain whether there was contamination during swab collection or culture procedures. Sample size calculation The sample size (n) for the study was calculated using the following formula []:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:n\:=\:\frac{{Z}^{2\:\:}P\:(1-P)}{{d}^{2}\:}$$\end{document} Where Z (1.96) is the statistics corresponding to the level of confidence (95%), P (0.16) is the assumed prevalence, and d (0.05) is the margin of error. The calculated sample size was 206.5 which was adjusted to 210 individuals. Statistical analysis Categorical variables were summarized using frequency and percentages. The prevalence of bacterial isolates was calculated by dividing the number of laboratory confirmed cases by the total number of subjects screened and multiplying this figure by a factor of 100. The 95% confidence limits (CI) of prevalence values have been calculated using the Wilson-range interval [24]. For MDR, we report isolate-level estimates reflecting routine laboratory practice and provide confidence intervals to quantify uncertainty. All statistical analyses were performed with Jamovi Desktop 2.5.3 (Open-Source Software, Sydney, Australia) and GraphPad Prism 8 for Windows (GraphPad Software, San Diego, CA, USA). Results Demographic information and distribution of wounds Out of the 210 samples received from both inpatient and outpatient departments, the total number of wound swabs taken from females were 106 (50.5%) and 104 (49.5%) males. The median age of the study participants was 40.0 (interquartile range = 55.0–30.0) years, with an age range between 1 and 90 years. The 31–40 year old group had the highest number [39/173(18.6%)] of bacterial growth (see Table 1). Table 1Distribution of bacterial isolates by sex and age of study participants Demographic factors Number of wound cases Sex Male104 (49.5%) Female106 (50.5%) Age (years) Number of bacterial isolates 0–106 (3.5%) 11–209(5.2%) 21–3033(19.1%) 31–4039(22.5%) 41–5032(18.5%) 51–6023(13.3%) 61 and above31 (17.9) Table 1Distribution of bacterial isolates by sex and age of study participants Demographic factors Number of wound cases Sex Male104 (49.5%) Female106 (50.5%) Age (years) Number of bacterial isolates 0–106 (3.5%) 11–209(5.2%) 21–3033(19.1%) 31–4039(22.5%) 41–5032(18.5%) 51–6023(13.3%) 61 and above31 (17.9) Distribution of bacterial isolates by sex and age of study participants The frequencies of the different categories of wounds were 9.0% for surgical site infections, 7.6% for diabetic wounds, 4.8% for cellulitis and 78.6% for other wounds. The category ‘other’ consisted of burns, animal bites, cuts, laceration and abrasion wounds. Prevalence of bacterial infections in wounds among participants Overall, the prevalence of bacterial infections was 173/210 (82.4%), with 143/210 (68.1%) Gram-negative bacteria (GNB) under aerobic, swab-based culture conditions (Table 2). Pseudomonas spp. were the most prevalent aerobically recoverable isolate in this facility (24.8% overall; 30% of GNB), followed by Citrobacter spp. (17.3%), Enterobacter spp. (13.3%) and Proteus mirabilis (10.4%). The prevalence of GNB was higher in males (55%) than in females (46%). The only Gram-positive organism isolated was Staphylococcus aureus with a prevalence of 17.3%. The distribution of GNB isolates in the various hospital departments is described in Table 2. The OPD recorded the highest prevalence 117/143 (82%) of GNB isolates, followed by the FW 8.0%, 11/143(8.0%), MW 8/143((6%), SW 4/143(3%), with the least prevalence recorded in the NICU 1/143(0.7%). Table 2Distribution of bacterial isolates among inpatient and outpatient departmentsGNBFrequencyOPDN (%)MWn (%)EWn (%)SWn (%)FWn (%)MATn (%)NICUn (%)Prevalence(%) Pseudomonas spp. 43(100)34 (79.0)3 (7.0)0 (0.0)1 (2.3)4 (9.3)0 (0.0)1(2.3) 24.8 S. aureus 30(100)22 (73.3)0 (0.0)0 (0.0)0 (0.0)6 (20.0)2 (6.7)0 (0.0) 17.3 Citrobacter spp. 30(100)24 (80.0)1 (3.3)0 (0.0)0 (0.0)4 (13.4)1 (3.3)0 (0.0) 17.3 Enterobacter spp. 23(100)20 (87.0)0 (0.0)0 (0.0)3 (13.0)0 (0.0)0 (0.0)0 (0.0) 13.3 P. mirabilis 18(100)15 (83.3)2 (11.1)0 (0.0)0 (0.0)0 (0.0)1 (5.6)0 (0.0) 10.4 P. vulgaris 14(100)13 (9.9)1 (7.1)0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0) 8.1 E. coli 7(100)6 (85.7)1 (14.3)0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0) 4.0 Klebsiella spp. 4(100)2 (50.0)0 (0.0)0 (0.0)0 (0.0)2 (50.0)0 (0.0)0 (0.0) 2.3 Providencia spp. 3(100)2 (66.7)0 (0.0)0 (0.0)0 (0.0)1 (33.3)0 (0.0)0 (0.0) 1.7 S. marcescens 1(100)1(100)0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0) 0.6 Total 173 (100) 139 (80.3) 8 (4.6) 0 (0) 4 (2.3) 17(9.8) 4 (2.3) 1(0.6) - GNB Gram negative bacteria, OP Outpatient department, MW male ward, EW Emergency ward, SW Surgical ward, FW Female ward, MAT maternity ward, NICU neonatal intensive care unit Table 2Distribution of bacterial isolates among inpatient and outpatient departmentsGNBFrequencyOPDN (%)MWn (%)EWn (%)SWn (%)FWn (%)MATn (%)NICUn (%)Prevalence(%) Pseudomonas spp. 43(100)34 (79.0)3 (7.0)0 (0.0)1 (2.3)4 (9.3)0 (0.0)1(2.3) 24.8 S. aureus 30(100)22 (73.3)0 (0.0)0 (0.0)0 (0.0)6 (20.0)2 (6.7)0 (0.0) 17.3 Citrobacter spp. 30(100)24 (80.0)1 (3.3)0 (0.0)0 (0.0)4 (13.4)1 (3.3)0 (0.0) 17.3 Enterobacter spp. 23(100)20 (87.0)0 (0.0)0 (0.0)3 (13.0)0 (0.0)0 (0.0)0 (0.0) 13.3 P. mirabilis 18(100)15 (83.3)2 (11.1)0 (0.0)0 (0.0)0 (0.0)1 (5.6)0 (0.0) 10.4 P. vulgaris 14(100)13 (9.9)1 (7.1)0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0) 8.1 E. coli 7(100)6 (85.7)1 (14.3)0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0) 4.0 Klebsiella spp. 4(100)2 (50.0)0 (0.0)0 (0.0)0 (0.0)2 (50.0)0 (0.0)0 (0.0) 2.3 Providencia spp. 3(100)2 (66.7)0 (0.0)0 (0.0)0 (0.0)1 (33.3)0 (0.0)0 (0.0) 1.7 S. marcescens 1(100)1(100)0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0) 0.6 Total 173 (100) 139 (80.3) 8 (4.6) 0 (0) 4 (2.3) 17(9.8) 4 (2.3) 1(0.6) - GNB Gram negative bacteria, OP Outpatient department, MW male ward, EW Emergency ward, SW Surgical ward, FW Female ward, MAT maternity ward, NICU neonatal intensive care unit Distribution of bacterial isolates among inpatient and outpatient departments GNB Gram negative bacteria, OP Outpatient department, MW male ward, EW Emergency ward, SW Surgical ward, FW Female ward, MAT maternity ward, NICU neonatal intensive care unit Antibiotic resistance profile of the GNB isolates The antibiogram of the GNB isolates is presented in Table 3. For Pseudomonas spp., most isolates demonstrated resistance to the antibiotics tested: 41/43(95.3%), 40/43(93.0%), and 37/43(86.0%) were resistant to ampicillin/tetracycline, meropenem, and ceftriaxone, respectively. For Citrobacter spp., 25/30(83.3%) of the isolates demonstrated resistance to meropenem and cefuroxime whereas 24/30(80.0%) were resistant to cefotaxime, and 23/30(76.7%) were resistant to ceftriaxone, ampicillin, tetracycline, and cotrimoxazole. Among the 23 isolates of Enterobacter spp., 21(91.3%) were resistant to ceftriaxone, meropenem, and tetracycline whilst 20(87.0%) were resistant to cefuroxime, cefotaxime, ampicillin, and cotrimoxazole. For the E. coli isolates, 7/7(100.0%) were resistant to ampicillin and co-trimoxazole whereas 6/7(85.7%) were resistant to meropenem, cefuroxime, and cefotaxime. For Klebsiella spp. isolates, 4/4(100.0%) were resistant to ampicillin, tetracycline, cefuroxime and ceftriaxone, whilst 3/4(75.0%) were resistant to meropenem, gentamicin, chloramphenicol, and cotrimoxazole. Other bacterial isolates like Proteus mirabilis and P. vulgaris, Providencia spp. and Serratia marcescens also demonstrated varying resistance patterns to different antibiotics as seen in Table 2. Table 3Antibiotic resistance profiles of the various Gram-negative bacteria isolated from wound infections among study participantsAMPTETCOTGENCRXCHLCTRCTXCIPAMKMEM Pseudomonas 95.395.383.762.8-69.886.090.734.94.693.0 Citrobacter 76.776.776.740.083.363.376.780.030.030.083.3 Enterobacter 87.091.387.065.287.065.291.387.060.913.091.3 P. mirabilis 77.894.472.244.489.966.777.872.233.327.877.8 P. vulgaris 92.985.785.778.678.664.378.664.350.014.385.7 E. coli 100.014.3100.057.185.742.971.485.757.10.085.7 Klebsiella 100.0100.075.075.0100.075.0100.050.050.025.075.0 Providencia 100.0100.066.766.7100.0100.0100.0100.066.733.366.7 S. marcescens 0.0100.00.00.0100.0100.0100.0100.00.00.0100.0Total81.184.271.954.490.671.986.981.142.516.474.8AMP Ampicillin, TET Tetracycline, COT Cotrimoxazole, GEN Gentamicin, CHL Chloramphenicol, CRX Cefuroxime, CTR Ceftriaxone, CTX Cefotaxime, CIP Ciprofloxacin, AMK Amikacin, MEM Meropenem Table 3Antibiotic resistance profiles of the various Gram-negative bacteria isolated from wound infections among study participantsAMPTETCOTGENCRXCHLCTRCTXCIPAMKMEM Pseudomonas 95.395.383.762.8-69.886.090.734.94.693.0 Citrobacter 76.776.776.740.083.363.376.780.030.030.083.3 Enterobacter 87.091.387.065.287.065.291.387.060.913.091.3 P. mirabilis 77.894.472.244.489.966.777.872.233.327.877.8 P. vulgaris 92.985.785.778.678.664.378.664.350.014.385.7 E. coli 100.014.3100.057.185.742.971.485.757.10.085.7 Klebsiella 100.0100.075.075.0100.075.0100.050.050.025.075.0 Providencia 100.0100.066.766.7100.0100.0100.0100.066.733.366.7 S. marcescens 0.0100.00.00.0100.0100.0100.0100.00.00.0100.0Total81.184.271.954.490.671.986.981.142.516.474.8AMP Ampicillin, TET Tetracycline, COT Cotrimoxazole, GEN Gentamicin, CHL Chloramphenicol, CRX Cefuroxime, CTR Ceftriaxone, CTX Cefotaxime, CIP Ciprofloxacin, AMK Amikacin, MEM Meropenem Antibiotic resistance profiles of the various Gram-negative bacteria isolated from wound infections among study participants AMP Ampicillin, TET Tetracycline, COT Cotrimoxazole, GEN Gentamicin, CHL Chloramphenicol, CRX Cefuroxime, CTR Ceftriaxone, CTX Cefotaxime, CIP Ciprofloxacin, AMK Amikacin, MEM Meropenem This study demonstrated a high prevalence of antibiotic resistance against the antibiotics tested including tetracycline (90.2%), cefuroxime (86.5%), ampicillin (86.7%), meropenem (86%), Vancomycin (85.3%), ceftriaxone (83.9%), cefotaxime (83.2%), cotrimoxazole (81.1%), chloramphenicol (67.8%) and gentamicin (57.3%). The most sensitive antibiotics among the GNB isolates were amikacin with 83% being sensitive, and ciprofloxacin with 59% being sensitive. Multi-drug resistance among bacterial isolates The prevalence of multidrug resistance in Gram negative bacteria was 95.1% [95% CI: 90.2–97.6] with 100.0% of isolates of Enterobacter spp., P. vulgaris, E. coli, Klebsiella spp., Providencia spp. and S. marcescens being multidrug resistant (Table 4). Table 4Frequency of multi-drug resistance among bacterial isolatesBacterial isolatesNumber of isolatesMDR, n (%, 95%CI) Pseudomonas spp. 4341[95.3, 84.5–98.7] Citrobacter spp. 3026[86.7, 70.3–94.7] Enterobacter spp. 2323[100.0, 85.7–100] P. mirabilis 1818[100.0, 82.4–100] P. vulgaris 1413[92.9, 68.5–98.7] Escherichia coli 77[100.0, 64.6–100] Klebsiella spp. 44[100.0, 51.0-100] Providencia spp. 33[100.0, 43.9–100] S. marcescens 11[100.0, 20.7–100]Total143136[95.1, 90.2–97.6]MDR Multidrug resistance, n number of isolates Table 4Frequency of multi-drug resistance among bacterial isolatesBacterial isolatesNumber of isolatesMDR, n (%, 95%CI) Pseudomonas spp. 4341[95.3, 84.5–98.7] Citrobacter spp. 3026[86.7, 70.3–94.7] Enterobacter spp. 2323[100.0, 85.7–100] P. mirabilis 1818[100.0, 82.4–100] P. vulgaris 1413[92.9, 68.5–98.7] Escherichia coli 77[100.0, 64.6–100] Klebsiella spp. 44[100.0, 51.0-100] Providencia spp. 33[100.0, 43.9–100] S. marcescens 11[100.0, 20.7–100]Total143136[95.1, 90.2–97.6]MDR Multidrug resistance, n number of isolates Frequency of multi-drug resistance among bacterial isolates MDR Multidrug resistance, n number of isolates Discussion This study evaluated the bacteriological profile and multidrug resistance of Gram-negative bacteria in wound infections in the middle belt of Ghana. The overall prevalence of bacteria in wounds from this study was 82.4%. This prevalence is higher than the one reported in the study at a secondary health care facility in the Eastern Region of Ghana in a three-year retrospective study [25] and the 61.6% reported in the study of northern Ethiopia by Haliu et al. [26]. However, our findings are consistent with a study from Ethiopia where Mohammed et al. [27] reported 83.9% prevalence of bacterial infections. In general, our findings are not surprising as wounds are a major health problem, with many people dying each year from preventable and curable wounds [28, 29]. This is a particular problem in hospitals and especially in surgical procedures, where clean operations may be contaminated by virulent organisms and subsequently infected [30]. The most prevalent bacteria isolated from wounds in this study was 24.8% Pseudomonas spp. Our findings agree with that of other studies who reported a Pseudomonas spp. prevalence of 30.2% in Ghana [31], 30.6% in Nigeria [32], 26.9% in Tanzania [33], and 46% in Sierra Leone [34]. However, our finding contrast with that of other studies who found that S. aureus and Klebsiella were the most prevalent species associated with wounds with prevalence of 28.7% [35] and 33.9% [36], respectively. Our findings on the predominance of Pseudomonas were not surprising as Pseudomonas spp. infection is known to be a serious problem in patients hospitalized with various medical conditions including cancer, cystic fibrosis, wound infections, urinary tract infections and burns with a case fatality of 50% [37, 38]. The high prevalence of Pseudomonas observed in our study may be because the organism is well adapted to a moist environment such as that caused by wounds [31]. The prevalence of carbapenems resistance among all GNB isolates ranged from 66.7% to 100.0%. This finding is especially worrying as carbapenem antibiotics (e.g., meropenem) are considered last-resort drugs for the treatment of severe bacterial infections as these infections are caused by bacteria that are resistant to other antibiotics [39, 40]. Some studies have demonstrated that the presence of genes such as blaOXA-48 and blaNDM-1 in bacteria such as Proteus spp., Citrobacter spp., and E. coli confers resistance to carbapenem antibiotics, though we did not confirm their presence in this study. Carbapenem-resistant strains are associated with higher mortality rates and treatment failure thus our finding suggests a higher likelihood of treatment failure when carbapenems are used as a treatment option for wounds [40, 41]. Additionally, the finding that all the bacteria from our study had some isolates being resistant to carbapenem increases the risk of distribution of resistance genes in the study population and there is the need to investigate these resistance genes. The need for better pharmacovigilance of antimicrobials, but especially of last resort antibiotics, cannot be overemphasized. The study also found that the prevalence of resistance to third generation cephalosporins (cefotaxime and ceftriaxone) among the wound isolates was substantially high, ranging from 50.0 to 100.0%. This is especially worrying as resistance to third generation cephalosporins is a consistent marker for the presence of extended-spectrum β-lactamases (ESBLs) in bacteria, particularly the Enterobacteriaceae including Citrobacter, Enterobacter, Proteus mirabilis, E. coli, Klebsiella, Providencia, and Serratia marcescens [42]. ESBL-producing bacteria are resistant to these antibiotics because the enzymes they produce break down the antibiotics, and this resistance can often be used as a proxy for identifying ESBL producers. However, we did not investigate for the presence of ESBLs in bacteria in this study. The national action plan on antimicrobial resistance in Ghana highlights the need for surveillance at different levels of health [11], but most surveillance efforts in the country are concentrated at tertiary institutions [39] whilst primary and secondary health facilities remain underrepresented in national reports. This study contributes to an enhanced understanding of the development of resistance by providing data from a setting which is not usually represented in the national reports. The high proportion of bacteria resistant to carbapenems, and third generation cephalosporins (cefotaxime and ceftriaxone) which are regarded as last resort drugs in some serious infection highlights the need to consider modifying national treatment guidelines. Additionally, given the high level of observed resistance, empirical prescribing policies in Ghanaian health care settings may need to be reviewed to ensure that first-line treatments remain effective [39]. A relatively high prevalence of MDR Gram-negative bacteria of 95.1% was observed in our study. The MDR prevalence we report in the present study is in tandem with the findings of 97.9% MDR reported by Aboagye et al. [10] in a study at a primary health care facility in Accra, Ghana. The MDR prevalence in our study ranged from 86.7% in Citrobacter spp., to 100% in Pseudomonas spp., Enterobacter spp., P. mirabilis, E. coli, Klebsiella spp., Providencia spp., and S. marcescens. The widespread MDR resistance amongst all the isolates is indicative of the growing global challenge of treating bacterial infections effectively, particularly in settings where antibiotic stewardship may be suboptimal [10]. The resistance observed suggests overuse or misuse of antibiotics, both within and outside healthcare facilities, which can promote the development of resistant strains. These findings underscore the need for stricter antibiotic prescription guidelines, improved infection control measures, and enhanced efforts in monitoring antibiotic resistance patterns [10]. Amikacin and ciprofloxacin were the most effective antibiotics against the bacterial isolates from wounds in this study. The low resistance of common pathogens to these antibacterials has been highlighted by studies in Ghana [43, 44] and comparable to studies in the African sub-region [45, 46]. This finding is important for guiding treatment and highlights a positive trend in an area where antibiotic resistance is a growing concern for many other drugs. There are some limitations in this study. We utilized phenotypic techniques for bacterial species identification and did not confirm resistance with molecular methods (e.g., PCR and sequencing) due to resource constraints. Anaerobic culture and enriched media were not performed, and tissue aspirate/curettage sampling was not used; therefore, Gram-negative predominance reflects organisms recoverable under aerobic, swab-based conditions. Another limitation is that the MDR isolates were not also sequenced to characterize the specific determinants of MDR. Lastly, findings reflect one secondary hospital and may not be generalizable nationally. Despite these limitations, the phenotypic methods used are widely accepted and used in routine surveillance activities. Our study therefore provides valuable information on the AMR and MDR situation in a secondary health care facility in the middle belt of Ghana. Conclusion Our study showed that the most common Gram-negative bacteria found in infected wounds was Pseudomonas spp. We also observed a high prevalence of multidrug resistance in the Gram-negative bacteria tested against selected common antibiotics used in Ghana. The high resistance to carbapenems and third generation cephalosporins (cefotaxime and ceftriaxone) considered to be the antibiotics of last resort was a particularly worrying finding. This indicates the need to review first-line and last-line antibiotic regimens and to improve and maintain antimicrobial surveillance in secondary health care settings. In addition, empirical wound healing should be monitored on a regular basis to assess the development of resistance and to ensure that wound healing is still effective against wound infections. Supplementary Information Additional file 1: PDF Questionnaire on patient data Additional file 1: PDF Questionnaire on patient data