Get 20M+ Full-Text Papers For Less Than $1.50/day. Subscribe now for You or Your Team.

Learn More →

Antimicrobial resistance in hospitalized surgical patients: a silently emerging public health concern in Uganda

Antimicrobial resistance in hospitalized surgical patients: a silently emerging public health... Background: Surgical site infections (SSIs) are difficult to treat and are associated with substantially longer hospital stay, higher treatment cost, morbidity and mortality, particularly when the etiological agent is multidrug-resistant (MDR). To address the limited data in Uganda on SSIs, we present the spectrum of bacteria isolated from hospitalized patients, the magnitude and impact of MDR bacterial isolates among patients with SSIs. Methods: A descriptive cross sectional study was conducted from September 2011 through April 2012 involving 314 patients with SSIs in the obstetrics & gynecology, general surgery and orthopedic wards at Mulago National Hospital in Kampala, Uganda. Wound swabs were taken and processed using standard microbiological methods. Clinico-demographic characteristics of patients were obtained using structured questionnaires and patients’ files. Results: Of the 314 enrolled patients with SSIs (mean age 29.7 ±13.14 years), 239 (76.1%) were female. More than half of the patients were from obstetrics and gynecology (62.1%, 195/314). Of 314 wound swabs taken, 68.8% (216/314) were culture positive aerobically, yielding 304 bacterial isolates; of which 23.7% (72/304) were Escherichia coli and 21.1% (64/304) were Staphylococcus aureus. More than three quarters of Enterobacteriaceae were found to be extended spectrum beta lactamase (ESBL) producers and 37.5% of S. aureus were Methicillin resistant S. aureus (MRSA). MDR occurred in 78.3% (238/304) of the isolates; these were more among Gram-negative bacteria (78.6%, 187/238) compared to Gram-positive bacteria (21.4%, 51/238), (p-value < 0.0001, χ2 = 49.219). Amikacin and imepenem for ESBL-producing Enterobacteriacea and vancomycin for MRSA showed excellent performance except that they remain expensive drugs in Uganda. Conclusion: Most SSIs at Mulago National Hospital are due to MDR bacteria. Isolation of MRSA and ESBL-producing Enterobacteriaceae in higher proportions than previously reported calls for laboratory guided SSIs- therapy and strengthening of infection control surveillance in this setting. Keywords: Antimicrobial resistance, Surgical patients, Uganda Background and rank third among all types of nosocomial infections in Surgical site infections (SSIs) are the infections involving the USA [2,3]. Infection rates among operated patients var- skin, subcutaneous tissue and organs/spaces opened or ies with hospital settings reflecting infection control prac- manipulated during an operation, occurring within 30 days tices as well as factors related to the agent, environment after the procedure or within one year if orthopedic im- and the host [4]; for example, infection rates reported were plant is in situ [1,2]. SSIs account for approximately 15.0% less than 3.0% in German and France, 16.9% in Rio de Janeiro, Brazil and 26.0% in Mwanza, Tanzania [5-8]. SSIs * Correspondence: senijj80@gmail.com can involve any surgical procedure ranging from obstetrics Department of Medical Microbiology, Makerere University College of Health & gynecology, general surgery and orthopedic [6,9,10]. Sciences, P.O. Box 7072, Kampala, Uganda The most predominant bacteria in hospital-acquired SSIs Department of Microbiology and Immunology, Catholic University of Health and Allied Sciences-Bugando, P.O. Box 1464, Mwanza, Tanzania are Staphylococcus aureus, Enterococcus spp, Pseudomonas Full list of author information is available at the end of the article © 2013 Seni et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Seni et al. BMC Research Notes 2013, 6:298 Page 2 of 7 http://www.biomedcentral.com/1756-0500/6/298 aeruginosa, Escherichia coli, and other Enterobacteriaceae; occurring within 30 days after the operative procedure of these, single bacterial isolates are common whereas 9.0% or within one year if orthopedic implant was in situ were to 27.0% of bacterial isolates from different surgical sites are included, whereas surgical patients with community- attributed to polymicrobial [7-9,11]. These infections pose acquired pyogenic infections such as abscess, furuncle and therapeutic challenges and are associated with substantially carbuncles; patients with infection of an episiotomy; and longer duration of hospital stay, increased hospital cost, patients with open fractures were excluded from the study. higher morbidity and mortality [5,12], particularly when the agents are Methicillin resistant S. aureus (MRSA), Extended Study clearance and ethical considerations spectrum beta lactamase (ESBL) producing Enterobacte- The study got ethical clearance from the Institutional riaceae and/or other agents collectively referred to as Review Board (IRB) of Makerere University College of multidrug-resistant (MDR) [11,13,14]. Studies from devel- Health Sciences (# REC REF 2011–183), Mulago Hos- oping countries have shown high level of resistance (ran- pital Research Committee (MREC #125) and the Uganda ging from 50 to 100%) to the commonly used antibiotics National Council for Science and Technology (UNCST) like ampicillin, trimethoprim – sulphamethoxazole, genta- (REF # HS 1080). A written informed consent from each micin, chloramphenicol and third generation cephalospo- patient/caretaker and assent for minors (11 to 17years) rins among S. aureus, E. coli,and P. aeruginosa [8,15] as were obtained whereas for each minor (<11 years), con- opposed to low rates of resistance ranging from 0-50% sent was obtained from his/her parent or caretaker. All in developed countries [16]. In both settings however, sub- patient information was kept confidential and anonym- stantial rates of resistance to oxacillin, erythromycin and ous using codes. clindamycin reported for S. aureus, ranged from 10-60% [8,9,15,16] whereas vancomycin (for S. aureus and other Data collection and laboratory procedures Gram-positive bacteria), amikacin, piperacillin-tazobactam Demographic and clinical characteristics from patients were and imepenem (for E. coli, P. aeruginosa and other Gram- collected using structured questionnaire and from patients’ negative bacteria) showed resistant rates of less than 25% files (see Additional file 1). The infected site was cleaned [17,18]. using normal saline and sterile gauze then, from each pa- It is well known that specific therapeutic options to tient, two wound swabs were collected using sterile cotton patients with SSIs are largely dependent on data from swabs in Amies transport media (Biolab, HUNGARY®). antimicrobial sensitivity tests generated by clinical la- boratories or sound epidemiological data from ongoing Isolate identification nosocomial infection surveillance [6,11,19]. Wound swabs were processed in the bacteriology labora- In Uganda, about 10% of the surgical procedures be- tory of the Department of Medical Microbiology, Makerere come septic accounting for an increasing morbidity and University College of Health Sciences, within 2 hours of mortality, with the commonest organism isolated being collection. The first wound swab was used to make Gram S. aureus [20-22]; however data on the spectrum of bac- stain smears while the second one was inoculated into teria isolated from hospitalized patients and their anti- blood agar, MacConkey agar, and mannitol-salt agar and microbial susceptibility patterns to guide SSI-therapy in incubated at 35-37°C for 24–48 hours. Identification of Mulago National Hospital remains scanty. Furthermore, bacteria was based on conventional physiological and bio- the magnitude and impact of MDR bacteria from SSIs chemical methods such as Gram stain, catalase reaction, are unknown. Thus, this study aimed at addressing these coagulase test, DNase test, hemolytic activity on sheep areas. Data herein will be crucial in guiding SSIs-therapy blood agar plates, bacitracin, optochin and trimethoprim- and will form a baseline for nosocomial SSIs surveillance. sulphamethoxazole (SXT) antimicrobial identification disks and bile esculin test for Gram-positive bacteria. Gram- Methods negative bacteria were identified based on colony morph- Study design and sampling process ology on blood agar and MacConkey agar, followed by This descriptive cross-sectional study was conducted at biochemical reactions namely oxidase, triple sugar iron Mulago National Hospital in Kampala, Uganda. The hos- (TSI), sulphur indole and motility (SIM), citrate, and ure- pital is located on Mulago Hill in the northern part of the ase tests [23]. city of Kampala and is the largest hospital in Uganda with an estimated 1,500 beds. Drug susceptibility tests The study was conducted for a period of 8 months Following identification of the bacterial isolates, a from September 2011 to April 2012 and involved 314 standard disc diffusion technique for drug susceptibility patients with clinical SSIs who consented to participate. test (DST) was performed as recommended by Clinical The patients were from obstetrics & gynecology, general and Laboratory Standard Institute (CLSI) [24]. For Gram- surgery and orthopedic wards. All patients with SSIs positive bacteria, discs (Biolab®, HUNGARY) tested were Seni et al. BMC Research Notes 2013, 6:298 Page 3 of 7 http://www.biomedcentral.com/1756-0500/6/298 ampicillin (10 μg), oxacillin (1 μg), trimethoprim-sul- where appropriate. To determine factors associated with phamethoxazole (1.25/23.75 μg), tetracycline (30 μg), bacteria isolation from SSI, we used univariate followed ciprofloxacin (5 μg), chloramphenicol (30 μg), gentami- by multivariate logistic regression analysis. At univariate cin (10 μg) [high level gentamicin (120 μg) for Entero- level all factors which had a p-value of less than 0.05 coccus spp], erythromycin (15 μg), clindamycin (2 μg), were subjected to multivariate analysis. The strength of and vancomycin (30 μg). For Gram-negative bacteria association between factors and outcome was measured discs (Biolab®, HUNGARY) tested included ampicillin using odds ratio with respective 95% confidence interval. (10 μg), piperacillin(100 μg), piperacillin-tazobactam Factors with p-value of less than 0.05 on multivariate lo- (100/10 μg), amoxicillin-clavulanic acid (20/10 μg), gistic regression analysis were considered as independent trimethoprim-sulphamethoxazole (1.25/23.75 μg), tetra- association of bacteria isolation from SSI. cycline (30 μg), ciprofloxacin (5 μg), chloramphenicol (30 μg), gentamicin (10 μg), amikacin (30 μg), ceftriax- one (30 μg), ceftazidime (30 μg), cefepime (30 μg), and Results imipenem (10 μg). These were incubated at 35-37°C for This study enrolled 314 patients with clinical SSIs. Among 24 hours. these, 239 (76.1%) were female. The overall mean age was Isolates which were not identifiable by the standard 29.7 ± 13.14 years (minimum 12 and maximum 83 years). conventional methods, colistin DST for all Acinetobacter More than half of the patients were from obstetrics and spp and P. aeruginosa as well as vancomycin DST for all gynecology wards, 62.1% (195/314), whereas 33.1% (104/ S. aureus were confirmed using the Phoenix Automated 314) and 4.8% (15/314) were from general surgery and instrument® (Becton-Dickson, Sparks Maryland) as per orthopedic wards respectively. The most common surgical manufacturer’s instruction. procedures were caesarean section 46.2% (145/314) and For determining inducible clindamycin resistance, clin- laparotomy 42.7% (134/314); open reduction and internal damycin disk (2 μg) and erythromycin disk (15 μg) were fixation (ORIF) accounted for 3.5% (11/314) while other placed side by side approximately 15-26 mm apart. Flat- surgical procedures contributed 7.6% (24/314). tening of the zone of inhibition adjacent to the eryth- Of the 314 non-repeat wound swabs collected, 216 romycin disk was regarded as a positive D-test. As (68.8%) were culture positive aerobically. The most pre- recommended by CLSI, isolates were screened for ESBL dominant bacterial isolates were E. coli, 23.7% (72/304) production using the double disc method and MRSA was and S. aureus, 21.1% (64/304) (see Figure 1). Single bac- identified by the use of cefoxitin disc (30 μg) [24,25]. terial isolates were recovered from 137 (63.4%) patients MDR was defined as an isolate with resistance to three or whereas 79 (36.6%) had polymicrobial infections. more antimicrobial classes [26]. All P. aeruginosa isolates were sensitive to colistin, Results on isolate identity and antimicrobial suscepti- whereas 3.9% (2/52) of Acinetobacter spp were resistant bility patterns were promptly reported to the attending to colistin. All Gram-negative bacteria were susceptible doctor for patient care. to imipemen, except 3.9% (2/52) of Acinetobacter spp. All S. aureus isolates were sensitive to vancomycin, Quality control whereas one Enterococcus spp [4.4% (1/23)] was resistant Reference strains S. aureus ATCC 25923 and Staphylo- to vancomycin i.e. VRE (see Table 1). coccus epidermidis ATCC 12228 for Gram-positive bac- More than three quarters of Enterobacteriaceae were teria and E. coli ATCC 25922 and P. aeruginosa ATCC phenotypically found to be ESBL producers [E. coli,79.2% 27853 for Gram-negative bacteria were used to quality- (57/72); Klebsiella spp, 92.3% (36/39) and other Entero- control microbiological procedures such as Gram staining , bacteriaceae, 73.5% (25/34)]. The proportion of MRSA growth of bacteria on respective media, microscopy, bio- among S. aureus isolates was 37.5% (24/64). chemical identification tests and drug susceptibility testing. Of the 304 isolates recovered from SSIs, 78.3% (238/304) were MDR, these were found significantly more among Data analysis Gram-negative bacteria (78.6%, 187/238) compared to Variables from the clinical and demographic data in the Gram-positive bacteria (21.4%, 51/238) (p-value < 0.0001, questionnaire and laboratory data were entered into χ2 = 49.219). Excel®, cleaned and exported to STATA software version In the univariate analysis, bacterial isolation from SSIs 11 (College Station, Texas, USA) for analysis according among patients was associated with age greater than 40 to the objectives of the study. Continuous variables were years, sex, ward type, operation (whether elective or emer- described as mean (± standard deviation). Categorical gency), duration from admission to operation, duration variables were described as proportion and were ana- from admission to discharge, type of antibiotic chemo- lyzed to compare the significance of difference in distri- prophylaxis given, duration of postoperative antibiotics bution by using Chi square test or Fischer’s exact test and outcome (see Table 2). Seni et al. BMC Research Notes 2013, 6:298 Page 4 of 7 http://www.biomedcentral.com/1756-0500/6/298 Figure 1 Proportion of bacterial isolates from patients with surgical site infections (N = 304). Other g- bacteria: Enterobacter cloacae (10), Proteus mirabilis (7), Morganella morganii (6), Providencia spp (5), Citrobacter freundii (4), Serratia marcescens (1) and Leclercia adecarboxylata (1); Other g + bacteria: Streptococcus pyogenes (3), Streptococcus agalactiae (1), and Streptococcus spp (4).Acinetobacter spp: Acinetobacter baumanii (48) and Acinetobacter baumanii-calcoaceticus complex (4); Klebsiella spp: Klebsiella pneumonia (35) and Klebsiella oxytoca (4); Enterococcus spp: Enterococcus faecalis (21) and Enterococcus faecium (2). Table 1 Antimicrobial resistance pattern among isolates from surgical site infections Antibiotics Bacteria isolates (N = 304) E.coli Klebsiella spp Acinetobacter P.aeruginosa Other g- S. aureus Enterococcus Other g + (n = 72) (n = 39) spp (n = 52) (n = 12) (n = 34) (n = 64) spp (n = 23) (n = 08) Ampicillin 100.0% 100.0% NA NA 97.1% 100.0% 30.0% 50.0% Piperacillin NA NA 100.0% 66.7% NA NA NA NA TZP 30.6% 38.5% 53.8% 16.7% 32.4% NA NA NA AMC 90.3% 100.0% NA NA 94.1% NA NA NA TMP-SMX 86.1% 92.3% 98.0% 100.0% 94.1% 89.1% NA 80.0% Tetracycline 72.2% 76.9% 65.4% 100.0% 82.4% 42.2% 74.0% 62.5% Ciprofloxacin 72.2% 66.6% 77.0% 16.7% 47.1% 29.7% 60.9% 50.0% Chloramphenicol 41.7% 71.8% NA NA 70.6% 15.6% 30.4% 37.5% ** Gentamicin 54.2% 76.9% 88.5% 16.7% 76.5% 18.8% 21.7% 50.0% Amikacin 4.2% 2.6% 32.7% 8.3% 0.0% NA NA NA Ceftriaxone 77.8% 92.3% NA NA 67.6% NA NA NA Ceftazidime 77.8% 87.2% 90.4% 25.0% 61.8% NA NA NA Cefepime 69.4% 87.2% 61.5% 16.7% 38.2% NA NA NA Imipenem 0.0% 0.0% 3.9% 0.0% 0.0% NA NA NA Erythromycin NA NA NA NA NA 46.9 % 65.2% 50.0% Clindamycin NA NA NA NA NA 40.6% NA 62.5% Vancomycin NA NA NA NA NA 0.0% 4.4% 0.0% * + × ** Piperacillin-tazobactam, Amoxicillin-clavulanate, Trimethoprim-sulphamethoxazole, High level resistance screening, NA: Not applicable, spp: species, g+: Gram-positive bacteria [Streptococcus pyogenes (3), Streptococcus agalactiae (1), and Streptococcus spp (4)] and g-: Gram-negative bacteria [Enterobacter cloacae (10), Proteus mirabilis (7), Morganella morganii (6), Providencia spp (5), Citrobacter freundii (4), Serratia marcescens (1) and Leclercia adecarboxylata (1)]. Seni et al. BMC Research Notes 2013, 6:298 Page 5 of 7 http://www.biomedcentral.com/1756-0500/6/298 Table 2 Association of clinico-demographic characteristics of patients with and without bacterial isolates from surgical site infections Variable Bacteria isolated No bacteria isolated OR [95% CI] P-value from SSI (n = 216) from SSI (n = 98) Age (years) ≤ 19 38 (67.9%) 18 (32.1%) 1 20 – 39 133 (63.6%) 76 (36.4%) 0.83(0 .42- 1.61) 0.5577 40 – 59 31 (91.2%) 3 (8.8%) 4.89( 1.24- 27.87) 0.0112 ≥ 60 14 (93.3%) 1 (6.7%) 6.63( 0.86- 295.67) 0.0478 Sex Male 61 (81.3%) 14 (18.7%) 1 Female 155 (64.9%) 84 (35.1%) 0.42(0 .21- 0.82) 0.0072 Ward type Obst & Gyn 115 (59.0%) 80 (41%) 1 General Surgery 89 (85.6%) 15 (14.4%) 4.13( 2.17- 8.22) 0.0000 Orthopedic 12 (80.0%) 3 (20.0%) 2.78(0.72- 15.79) 0.1085 Operation Elective 57 (83.8%) 11 (16.2%) 1 Emergency 159 (64.6%) 87 (35.4%) 0.35( 0.16- 0.73) 0.0025 Duration from admission to operation ≤ 1 day 147 (65.3%) 78 (34.7%) 1 > 1 days 69 (77.5%) 20 (22.5%) 1.83( 1.01- 3.42) 0.0356 Duration from admission to discharge ≤ 14days 48 (58.5%) 34 (41.5%) 1 > 14days 110 (82.7%) 23 (17.3%) 3.39( 1.73- 6.68) 0.0001 Antibiotic chemoprophylaxis CRO 124 (62.9%) 73 (37.1%) 1 * ** CRO +MZ 70 (76.9%) 21 (23.1%) 1.96( 1.08- 3.65) 0.0187 Others 22 (84.6%) 4 (15.4%) 3.24( 1.04- 13.37) 0.0289 Duration of Postoperative antibiotics ≤ 3days 48 (50.5%) 47 (49.5%) 1 > 3days 168 (76.7%) 51 (23.3%) 3.23( 1.87- 5.54) 0.0001 Outcome Improved 147 (72.1%) 57 (27.9%) 1 Died 10 (100.0%) 0 (0.0%) - 0.042 * ** × Ceftriaxone, Metronidazole, Ampicloxacillin, Ciprofloxacin & Gentamicin. On multivariate logistic regression analysis, longer dur- Discussion ation from admission to discharge, longer duration of Of the 314 patients with clinical SSIs enrolled in this postoperative antibiotics and outcome were statistically study; the mean age (29.7 ± 13.14 years), higher female associated with bacteria isolation from SSIs (see Table 3). proportion (76.1%) and preponderance of admission in Of ten patients who died, 70% (7/10) had ESBL producing obstetrics and gynecology wards (62.1%) was similar to Enterobacteriaceae, four of which were E. coli. another study [11], but the proportion of female in other studies were between 30% to 60% and wards of admis- sion varied reflecting the underlying surgical procedures [8,9,13]. Caesarean section and laparotomy accounted for Table 3 Multivariate logistic regression analysis for more than three quarters of all surgical procedures in this factors associated with bacterial isolation from SSIs study whereas other similar studies have also found these Variable OR 95% CI p-value procedures to be quite common [8,11,27]. The predomin- Age (years) 1.16 0.74 - 1.80 0.524 ance of SSIs cases in obstetrics and gynecology wards is Sex 1.41 0.49 - 4.02 0.522 quite alarming and thus, a need to institute stringent in- Wards 0.44 0.18 - 1.10 0.078 fection prevention and control measures in this setting, Operation 0.62 0.24 - 1.65 0.343 more especially in emergency surgeries which accounted Duration from admission to operation 0.89 0.40 - 1.97 0.773 for more SSIs cases as opposed to elective surgeries. As noted from other studies [8,9,11,28], S. aureus and Duration from admission to discharge 1.93 1.01 - 3.72 0.047 E. coli were the most common bacterial isolates from Antibiotic chemoprophylaxis 1.27 0.77 - 2.12 0.351 SSIs. However their sources remain unknown from the Duration of postoperative antibiotics 2.52 1.45 - 4.36 0.001 present study though other studies have documented Outcome 0.52 0.30 - 0.91 0.021 both endogenous and exogenous sources from hospital Seni et al. BMC Research Notes 2013, 6:298 Page 6 of 7 http://www.biomedcentral.com/1756-0500/6/298 environment could be potential niches [29,30]. Similar reported as negative cultures. This was because of lack of to other related studies [9,18,31], high level of resistance standardized in-house detection methods and lack of an- was found among commonly used antibiotics like ampi- aerobic detection panels in the Phoenix Automated instru- cillin, trimethoprim-sulphamethoxazole, and tetracycline ment (Becton-Dickson, Sparks Maryland) that we used. in both Gram-positive and Gram-negative bacteria. Gram- negative bacteria showed more resistance to gentamicin, Conclusion ciprofloxacin and chloramphenicol as compared to Gram- Most SSIs at Mulago National Hospital are due to MDR positive bacteria. With exception of Acinetobacter spp,all bacteria, these are significantly more among Gram-negative Gram-negative bacteria displayed low resistance rates to than Gram-positive bacteria. Isolation of MRSA and ESBL- piperacillin-tazobactam, amikacin and imipenem. There producing Enterobacteriaceae in higher proportions than were also low resistance rates of Acinetobacter spp and previously reported calls for enhanced antibiotic steward- P. aeruginosa to colistin. These findings are similar to ship including laboratory guided SSIs-therapy and strength- another study [17]. The present study has shown that ening of infection control surveillance by identifying the rates of resistance to erythromycin (46.9%) among S. sources of these MDR isolates. In the light of these findings, aureus was relatively low compared to that of Entero- there is a need to investigate whether there is clonal spread coccus spp (65.2%), with excellent performance of vanco- of the predominant bacteria within/or among surgical mycin on both S. aureus and Enterococcus spp. These wards at Mulago National Hospital. findings are in agreement to another study [18]. This study found more MDR among Gram-negative bacteria than Additional file Gram-positive bacteria; of these the proportion of MRSA (37.5%) among S. aureus isolates was more than previously Additional file 1: Appendix i. Questionnaire. reported (25.0% and 31.5%) from Uganda [9,21] and other countries [8,16,18]. Thus, while β-lactamase-resistant anti- Competing interests biotics such as cloxacillin could still be effective in this The authors declare that they have no competing interests. setting, they are likely to be ineffective against the 38% Authors’ contributions of isolates that were confirmed as MRSA. The fact that Conceived and designed the experiments: JS, DPK, and FB. Specimen we found no vancomycin resistance among S. aureus iso- collection: JS. Supervised the clinical component of research: PM. Performed lates shows that this drug remains the last resort in sys- the experiments: JS. Supervised the laboratory component of research: CFN, DPK, and FB. Analyzed the data: JS, DPK, AK, and FB. Contributed reagents, temic infections caused by MRSA in this setting. Previous materials and analysis tools: JS, CFN, HK, and MLJ. Wrote the manuscript: JS, studies [11,17] have shown low rates (14% to 22%) of CFN, DPK, PM, MLJ, HK, AK, and FB. All authors have read and approved the ESBL-producers among Enterobacteriaceae isolates but the final manuscript. present study and another similar study in the same region Acknowledgements [8] have shown ESBL-producers to account more than The authors would like to thank patients and health workers in all surgical wards; three quarter of Enterobacteriaceae. This can be attributed Emmanuel Aboce, Tonny Lugya, and Hannington Baluku for excellent technical to the empirical use of third generation cephalosporins assistance; Willy Ssengooba for statistical inputs, and all staffs in the Department of Medical Microbiology for their support. This work was funded by Catholic (usually in combination with another drug such as genta- University of Health and Allied Sciences Bugando, Mwanza-Tanzania to JS. micin) in almost all hospitalized patients and lack of anti- microbial resistance surveillance in surgical wards at Author details Department of Medical Microbiology, Makerere University College of Health Mulago National Hospital. This is a major threat to patient Sciences, P.O. Box 7072, Kampala, Uganda. Department of Surgery, Makerere care as ESBL production renders use of these ceftriaxone University College of Health Sciences, P.O. Box 7072, Kampala, Uganda. or ceftazidime useless. Absence of resistance to imepenem Department of Microbiology and Immunology, Catholic University of Health and Allied Sciences-Bugando, P.O. Box 1464, Mwanza, Tanzania. Department among these isolates is however a good finding, except that of Community Medicine, Catholic University of Health and Allied Sciences- imepenem remains an expensive drug. Bugando, P.O. Box 1464, Mwanza, Tanzania. Multivariate logistic regression analysis of clinical and Received: 23 November 2012 Accepted: 25 July 2013 demographic characteristics of patients with SSIs in this Published: 27 July 2013 study showed that longer duration from admission to dis- charge, longer duration of postoperative antibiotics and References outcome (death) were associated with bacteria isolation 1. Horan TC, Gaynes RP, Martone WJ, Jarvis WR, Emori TG: CDC definitions of nosocomial surgical site infections, 1992: a modification of CDC from SSIs. These findings have also been shown in other definitions of surgical wound infections. Am J Infect Control 1992, similar studies [5,11,27]. 20(5):271–274. 2. Mangram AJ, Horan TC, Pearson ML, Silver LC, Jarvis WR: Guideline for prevention of surgical site infection, 1999. Centers for Disease Control Limitation and Prevention (CDC) Hospital Infection Control Practices Advisory The study did not isolate strict anaerobes, which could Committee. Am J Infect Control 1999, 27(2):97–132. quiz 133–134; have increased the number of bacterial isolates currently discussion 196. Seni et al. BMC Research Notes 2013, 6:298 Page 7 of 7 http://www.biomedcentral.com/1756-0500/6/298 3. Emori TG, Gaynes RP: An overview of nosocomial infections, including the 23. Koneman EW, Allen SD, Janda WM, Schreckenberger PC, Winn WC: Color role of the microbiology laboratory. Clin Microbiol Rev 1993, 6(4):428–442. atlas and textbook of diagnostic microbiology. 5th edition. Philadelphia, 4. Fabiano G, Pezzolla A, Filograna MA, Ferrarese F: [Risk factors of surgical Pa: Lippincott, Williams & Wilkins Publishers; 1997. wound infection]. Ann Ital Chir 2004, 75(1):11–16. 24. CLSI: Perfomance standars for antimicrobial susceptibility testing; twenty first 5. Astagneau P, Rioux C, Golliot F, Brucker G: Morbidity and mortality information supplement, vol. CLSI document M100-S21. Wayne, PA: Clinical associated with surgical site infections: results from the 1997–1999 and Laboratory Standards Institute; 2011. INCISO surveillance. J Hosp Infect 2001, 48(4):267–274. 25. Livermore DM, Brown DF: Detection of beta-lactamase-mediated 6. Barwolff S, Sohr D, Geffers C, Brandt C, Vonberg RP, Halle H, Ruden H, resistance. J Antimicrob Chemother 2001, 48(Suppl 1):59–64. Gastmeier P: Reduction of surgical site infections after Caesarean delivery 26. D'Agata EM: Rapidly rising prevalence of nosocomial multidrug-resistant, using surveillance. J Hosp Infect 2006, 64(2):156–161. F-negative bacilli: a 9-year surveillance study. Infect Control Hosp Epidemiol 2004, 25(10):842–846. 7. Santos KR, Fonseca LS, Bravo Neto GP, Gontijo Filho PP: Surgical site 27. Razavi SM, Ibrahimpoor M, Sabouri Kashani A, Jafarian A: Abdominal infection: rates, etiology and resistance patterns to antimicrobials surgical site infections: incidence and risk factors at an Iranian teaching among strains isolated at Rio de Janeiro University Hospital. Infection hospital. BMC Surg 2005, 5:2. 1997, 25(4):217–220. 28. Kasatpibal N, Jamulitrat S, Chongsuvivatwong V: Standardized incidence 8. Mawalla B, Mshana SE, Chalya PL, Imirzalioglu C, Mahalu W: Predictors of rates of surgical site infection: a multicenter study in Thailand. Am J surgical site infections among patients undergoing major surgery at Bugando Infect Control 2005, 33(10):587–594. MedicalCentreinNorthwesternTanzania. BMC Surgery 2011, 11:21. 29. Wenzel RP, Perl TM: The significance of nasal carriage of Staphylococcus 9. Anguzu JR, Olila D: Drug sensitivity patterns of bacterial isolates from aureus and the incidence of postoperative wound infection. J Hosp Infect septic post-operative wounds in a regional referral hospital in Uganda. 1995, 31(1):13–24. Afr Health Sci 2007, 7(3):148–154. 30. Atata RF, Ibrahim YKE, Olurinola PF, Giwa A, Akanbi AA II, Sani AA: Clinical 10. Whitehouse JD, Friedman ND, Kirkland KB, Richardson WJ, Sexton DJ: The bacterial isolates from hospital environment as agents of surgical wound impact of surgical-site infections following orthopedic surgery at a nosocomial infections. Journal of Pharmacy & Bioresources 2010, 7(2):146–155. community hospital and a university hospital: adverse quality of life, 31. Adegoke AA, Mvuyo T, Okoh AI, Steve J: Studies on multiple antibiotic excess length of stay, and extra cost. Infect Control Hosp Epidemiol 2002, resistant bacterial isolated from surgical site infection. Scientific Research 23(4):183–189. and Essays 2010, 5(24):3876–3881. 11. Fehr J, Hatz C, Soka I, Kibatala P, Urassa H, Smith T, Mshinda H, Frei R, Widmer A: Risk factors for surgical site infection in a Tanzanian district doi:10.1186/1756-0500-6-298 hospital: a challenge for the traditional National Nosocomial Infections Cite this article as: Seni et al.: Antimicrobial resistance in hospitalized Surveillance system index. Infect Control Hosp Epidemiol 2006, 27(12):1401–1404. surgical patients: a silently emerging public health concern in Uganda. 12. Kirkland KB, Briggs JP, Trivette SL, Wilkinson WE, Sexton DJ: The impact of BMC Research Notes 2013 6:298. surgical-site infections in the 1990s: attributable mortality, excess length of hospitalization, and extra costs. Infect Control Hosp Epidemiol 1999, 20(11):725–730. 13. Engemann JJ, Carmeli Y, Cosgrove SE, Fowler VG, Bronstein MZ, Trivette SL, Briggs JP, Sexton DJ, Kaye KS: Adverse clinical and economic outcomes attributable to methicillin resistance among patients with Staphylococcus aureus surgical site infection. Clin Infect Dis 2003, 36(5):592–598. 14. Wassef MA, Hussein A, Abdul Rahman EM, El-Sherif RH: A prospective surveillance of surgical site infections: study for efficacy of preoperative antibiotic prophylaxis. African Journal of Microbiology Research 2012, 6(12):3072–3078. 15. Amare B, Abdurrahman Z, Moges B, Ali J, Muluken L, Alemayehu M, Yifru S, Sendek B, Belyhun YFM, et al: Postoperative surgical site bacterial infections and drug susceptibility patterns at Gondar University Teaching Hospital, Northwest Ethiopia. J Bacteriol Parasitol 2011, 2:126. 16. Throckmorton AD, Baddour LM, Hoskin TL, Boughey JC, Degnim AC: Microbiology of surgical site infections complicating breast surgery. Surg Infect (Larchmt) 2010, 11(4):355–359. 17. Hawser SP, Bouchillon SK, Hoban DJ, Badal RE: Epidemiologic trends, occurrence of extended-spectrum beta-lactamase production, and performance of ertapenem and comparators in patients with intra-abdominal infections: analysis of global trend data from 2002–2007 from the SMART study. Surg Infect (Larchmt) 2010, 11(4):371–378. 18. Kownhar H, Shankar EM, Vignesh R, Sekar R, Velu V, Rao UA: High isolation rate of Staphylococcus aureus from surgical site infections in an Indian hospital. J Antimicrob Chemother 2008, 61(3):758–760. 19. Krukerink M, Kievit J, de Mheen PJ M-v: Evaluation of routinely reported Submit your next manuscript to BioMed Central surgical site infections against microbiological culture results: a tool to and take full advantage of: identify patient groups where diagnosis and treatment may be improved. BMC Infect Dis 2009, 9:176. • Convenient online submission 20. Olaro C: Study to assess the risk factors for postoperative complications • Thorough peer review following abdominal surgery in Mulago Hospital. Kampala: Makerere University; 1999. • No space constraints or color figure charges 21. Ojulong J, Mwambu TP, Joloba M, Bwanga F, Kaddu-Mulindwa DH: Relative • Immediate publication on acceptance prevalence of methicilline resistant Staphylococcus aureus and its susceptibility pattern in Mulago Hospital, Kampala, Uganda. Tanzan J • Inclusion in PubMed, CAS, Scopus and Google Scholar Health Res 2009, 11(3):149–153. • Research which is freely available for redistribution 22. Kitara DL, Kakande I, Mugisa BD, Obol JH: The postoperative complications prediction in Mulago Hospital using POSSUM scoring system. East and Submit your manuscript at Central African Journal of Surgery 2010, 15(2):90–96. www.biomedcentral.com/submit http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Research Notes Springer Journals

Antimicrobial resistance in hospitalized surgical patients: a silently emerging public health concern in Uganda

Loading next page...
 
/lp/springer-journals/antimicrobial-resistance-in-hospitalized-surgical-patients-a-silently-0f9zIhIU5m

References (61)

Publisher
Springer Journals
Copyright
Copyright © 2013 by Seni et al.; licensee BioMed Central Ltd.
Subject
Biomedicine; Biomedicine general; Medicine/Public Health, general; Life Sciences, general
eISSN
1756-0500
DOI
10.1186/1756-0500-6-298
pmid
23890206
Publisher site
See Article on Publisher Site

Abstract

Background: Surgical site infections (SSIs) are difficult to treat and are associated with substantially longer hospital stay, higher treatment cost, morbidity and mortality, particularly when the etiological agent is multidrug-resistant (MDR). To address the limited data in Uganda on SSIs, we present the spectrum of bacteria isolated from hospitalized patients, the magnitude and impact of MDR bacterial isolates among patients with SSIs. Methods: A descriptive cross sectional study was conducted from September 2011 through April 2012 involving 314 patients with SSIs in the obstetrics & gynecology, general surgery and orthopedic wards at Mulago National Hospital in Kampala, Uganda. Wound swabs were taken and processed using standard microbiological methods. Clinico-demographic characteristics of patients were obtained using structured questionnaires and patients’ files. Results: Of the 314 enrolled patients with SSIs (mean age 29.7 ±13.14 years), 239 (76.1%) were female. More than half of the patients were from obstetrics and gynecology (62.1%, 195/314). Of 314 wound swabs taken, 68.8% (216/314) were culture positive aerobically, yielding 304 bacterial isolates; of which 23.7% (72/304) were Escherichia coli and 21.1% (64/304) were Staphylococcus aureus. More than three quarters of Enterobacteriaceae were found to be extended spectrum beta lactamase (ESBL) producers and 37.5% of S. aureus were Methicillin resistant S. aureus (MRSA). MDR occurred in 78.3% (238/304) of the isolates; these were more among Gram-negative bacteria (78.6%, 187/238) compared to Gram-positive bacteria (21.4%, 51/238), (p-value < 0.0001, χ2 = 49.219). Amikacin and imepenem for ESBL-producing Enterobacteriacea and vancomycin for MRSA showed excellent performance except that they remain expensive drugs in Uganda. Conclusion: Most SSIs at Mulago National Hospital are due to MDR bacteria. Isolation of MRSA and ESBL-producing Enterobacteriaceae in higher proportions than previously reported calls for laboratory guided SSIs- therapy and strengthening of infection control surveillance in this setting. Keywords: Antimicrobial resistance, Surgical patients, Uganda Background and rank third among all types of nosocomial infections in Surgical site infections (SSIs) are the infections involving the USA [2,3]. Infection rates among operated patients var- skin, subcutaneous tissue and organs/spaces opened or ies with hospital settings reflecting infection control prac- manipulated during an operation, occurring within 30 days tices as well as factors related to the agent, environment after the procedure or within one year if orthopedic im- and the host [4]; for example, infection rates reported were plant is in situ [1,2]. SSIs account for approximately 15.0% less than 3.0% in German and France, 16.9% in Rio de Janeiro, Brazil and 26.0% in Mwanza, Tanzania [5-8]. SSIs * Correspondence: senijj80@gmail.com can involve any surgical procedure ranging from obstetrics Department of Medical Microbiology, Makerere University College of Health & gynecology, general surgery and orthopedic [6,9,10]. Sciences, P.O. Box 7072, Kampala, Uganda The most predominant bacteria in hospital-acquired SSIs Department of Microbiology and Immunology, Catholic University of Health and Allied Sciences-Bugando, P.O. Box 1464, Mwanza, Tanzania are Staphylococcus aureus, Enterococcus spp, Pseudomonas Full list of author information is available at the end of the article © 2013 Seni et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Seni et al. BMC Research Notes 2013, 6:298 Page 2 of 7 http://www.biomedcentral.com/1756-0500/6/298 aeruginosa, Escherichia coli, and other Enterobacteriaceae; occurring within 30 days after the operative procedure of these, single bacterial isolates are common whereas 9.0% or within one year if orthopedic implant was in situ were to 27.0% of bacterial isolates from different surgical sites are included, whereas surgical patients with community- attributed to polymicrobial [7-9,11]. These infections pose acquired pyogenic infections such as abscess, furuncle and therapeutic challenges and are associated with substantially carbuncles; patients with infection of an episiotomy; and longer duration of hospital stay, increased hospital cost, patients with open fractures were excluded from the study. higher morbidity and mortality [5,12], particularly when the agents are Methicillin resistant S. aureus (MRSA), Extended Study clearance and ethical considerations spectrum beta lactamase (ESBL) producing Enterobacte- The study got ethical clearance from the Institutional riaceae and/or other agents collectively referred to as Review Board (IRB) of Makerere University College of multidrug-resistant (MDR) [11,13,14]. Studies from devel- Health Sciences (# REC REF 2011–183), Mulago Hos- oping countries have shown high level of resistance (ran- pital Research Committee (MREC #125) and the Uganda ging from 50 to 100%) to the commonly used antibiotics National Council for Science and Technology (UNCST) like ampicillin, trimethoprim – sulphamethoxazole, genta- (REF # HS 1080). A written informed consent from each micin, chloramphenicol and third generation cephalospo- patient/caretaker and assent for minors (11 to 17years) rins among S. aureus, E. coli,and P. aeruginosa [8,15] as were obtained whereas for each minor (<11 years), con- opposed to low rates of resistance ranging from 0-50% sent was obtained from his/her parent or caretaker. All in developed countries [16]. In both settings however, sub- patient information was kept confidential and anonym- stantial rates of resistance to oxacillin, erythromycin and ous using codes. clindamycin reported for S. aureus, ranged from 10-60% [8,9,15,16] whereas vancomycin (for S. aureus and other Data collection and laboratory procedures Gram-positive bacteria), amikacin, piperacillin-tazobactam Demographic and clinical characteristics from patients were and imepenem (for E. coli, P. aeruginosa and other Gram- collected using structured questionnaire and from patients’ negative bacteria) showed resistant rates of less than 25% files (see Additional file 1). The infected site was cleaned [17,18]. using normal saline and sterile gauze then, from each pa- It is well known that specific therapeutic options to tient, two wound swabs were collected using sterile cotton patients with SSIs are largely dependent on data from swabs in Amies transport media (Biolab, HUNGARY®). antimicrobial sensitivity tests generated by clinical la- boratories or sound epidemiological data from ongoing Isolate identification nosocomial infection surveillance [6,11,19]. Wound swabs were processed in the bacteriology labora- In Uganda, about 10% of the surgical procedures be- tory of the Department of Medical Microbiology, Makerere come septic accounting for an increasing morbidity and University College of Health Sciences, within 2 hours of mortality, with the commonest organism isolated being collection. The first wound swab was used to make Gram S. aureus [20-22]; however data on the spectrum of bac- stain smears while the second one was inoculated into teria isolated from hospitalized patients and their anti- blood agar, MacConkey agar, and mannitol-salt agar and microbial susceptibility patterns to guide SSI-therapy in incubated at 35-37°C for 24–48 hours. Identification of Mulago National Hospital remains scanty. Furthermore, bacteria was based on conventional physiological and bio- the magnitude and impact of MDR bacteria from SSIs chemical methods such as Gram stain, catalase reaction, are unknown. Thus, this study aimed at addressing these coagulase test, DNase test, hemolytic activity on sheep areas. Data herein will be crucial in guiding SSIs-therapy blood agar plates, bacitracin, optochin and trimethoprim- and will form a baseline for nosocomial SSIs surveillance. sulphamethoxazole (SXT) antimicrobial identification disks and bile esculin test for Gram-positive bacteria. Gram- Methods negative bacteria were identified based on colony morph- Study design and sampling process ology on blood agar and MacConkey agar, followed by This descriptive cross-sectional study was conducted at biochemical reactions namely oxidase, triple sugar iron Mulago National Hospital in Kampala, Uganda. The hos- (TSI), sulphur indole and motility (SIM), citrate, and ure- pital is located on Mulago Hill in the northern part of the ase tests [23]. city of Kampala and is the largest hospital in Uganda with an estimated 1,500 beds. Drug susceptibility tests The study was conducted for a period of 8 months Following identification of the bacterial isolates, a from September 2011 to April 2012 and involved 314 standard disc diffusion technique for drug susceptibility patients with clinical SSIs who consented to participate. test (DST) was performed as recommended by Clinical The patients were from obstetrics & gynecology, general and Laboratory Standard Institute (CLSI) [24]. For Gram- surgery and orthopedic wards. All patients with SSIs positive bacteria, discs (Biolab®, HUNGARY) tested were Seni et al. BMC Research Notes 2013, 6:298 Page 3 of 7 http://www.biomedcentral.com/1756-0500/6/298 ampicillin (10 μg), oxacillin (1 μg), trimethoprim-sul- where appropriate. To determine factors associated with phamethoxazole (1.25/23.75 μg), tetracycline (30 μg), bacteria isolation from SSI, we used univariate followed ciprofloxacin (5 μg), chloramphenicol (30 μg), gentami- by multivariate logistic regression analysis. At univariate cin (10 μg) [high level gentamicin (120 μg) for Entero- level all factors which had a p-value of less than 0.05 coccus spp], erythromycin (15 μg), clindamycin (2 μg), were subjected to multivariate analysis. The strength of and vancomycin (30 μg). For Gram-negative bacteria association between factors and outcome was measured discs (Biolab®, HUNGARY) tested included ampicillin using odds ratio with respective 95% confidence interval. (10 μg), piperacillin(100 μg), piperacillin-tazobactam Factors with p-value of less than 0.05 on multivariate lo- (100/10 μg), amoxicillin-clavulanic acid (20/10 μg), gistic regression analysis were considered as independent trimethoprim-sulphamethoxazole (1.25/23.75 μg), tetra- association of bacteria isolation from SSI. cycline (30 μg), ciprofloxacin (5 μg), chloramphenicol (30 μg), gentamicin (10 μg), amikacin (30 μg), ceftriax- one (30 μg), ceftazidime (30 μg), cefepime (30 μg), and Results imipenem (10 μg). These were incubated at 35-37°C for This study enrolled 314 patients with clinical SSIs. Among 24 hours. these, 239 (76.1%) were female. The overall mean age was Isolates which were not identifiable by the standard 29.7 ± 13.14 years (minimum 12 and maximum 83 years). conventional methods, colistin DST for all Acinetobacter More than half of the patients were from obstetrics and spp and P. aeruginosa as well as vancomycin DST for all gynecology wards, 62.1% (195/314), whereas 33.1% (104/ S. aureus were confirmed using the Phoenix Automated 314) and 4.8% (15/314) were from general surgery and instrument® (Becton-Dickson, Sparks Maryland) as per orthopedic wards respectively. The most common surgical manufacturer’s instruction. procedures were caesarean section 46.2% (145/314) and For determining inducible clindamycin resistance, clin- laparotomy 42.7% (134/314); open reduction and internal damycin disk (2 μg) and erythromycin disk (15 μg) were fixation (ORIF) accounted for 3.5% (11/314) while other placed side by side approximately 15-26 mm apart. Flat- surgical procedures contributed 7.6% (24/314). tening of the zone of inhibition adjacent to the eryth- Of the 314 non-repeat wound swabs collected, 216 romycin disk was regarded as a positive D-test. As (68.8%) were culture positive aerobically. The most pre- recommended by CLSI, isolates were screened for ESBL dominant bacterial isolates were E. coli, 23.7% (72/304) production using the double disc method and MRSA was and S. aureus, 21.1% (64/304) (see Figure 1). Single bac- identified by the use of cefoxitin disc (30 μg) [24,25]. terial isolates were recovered from 137 (63.4%) patients MDR was defined as an isolate with resistance to three or whereas 79 (36.6%) had polymicrobial infections. more antimicrobial classes [26]. All P. aeruginosa isolates were sensitive to colistin, Results on isolate identity and antimicrobial suscepti- whereas 3.9% (2/52) of Acinetobacter spp were resistant bility patterns were promptly reported to the attending to colistin. All Gram-negative bacteria were susceptible doctor for patient care. to imipemen, except 3.9% (2/52) of Acinetobacter spp. All S. aureus isolates were sensitive to vancomycin, Quality control whereas one Enterococcus spp [4.4% (1/23)] was resistant Reference strains S. aureus ATCC 25923 and Staphylo- to vancomycin i.e. VRE (see Table 1). coccus epidermidis ATCC 12228 for Gram-positive bac- More than three quarters of Enterobacteriaceae were teria and E. coli ATCC 25922 and P. aeruginosa ATCC phenotypically found to be ESBL producers [E. coli,79.2% 27853 for Gram-negative bacteria were used to quality- (57/72); Klebsiella spp, 92.3% (36/39) and other Entero- control microbiological procedures such as Gram staining , bacteriaceae, 73.5% (25/34)]. The proportion of MRSA growth of bacteria on respective media, microscopy, bio- among S. aureus isolates was 37.5% (24/64). chemical identification tests and drug susceptibility testing. Of the 304 isolates recovered from SSIs, 78.3% (238/304) were MDR, these were found significantly more among Data analysis Gram-negative bacteria (78.6%, 187/238) compared to Variables from the clinical and demographic data in the Gram-positive bacteria (21.4%, 51/238) (p-value < 0.0001, questionnaire and laboratory data were entered into χ2 = 49.219). Excel®, cleaned and exported to STATA software version In the univariate analysis, bacterial isolation from SSIs 11 (College Station, Texas, USA) for analysis according among patients was associated with age greater than 40 to the objectives of the study. Continuous variables were years, sex, ward type, operation (whether elective or emer- described as mean (± standard deviation). Categorical gency), duration from admission to operation, duration variables were described as proportion and were ana- from admission to discharge, type of antibiotic chemo- lyzed to compare the significance of difference in distri- prophylaxis given, duration of postoperative antibiotics bution by using Chi square test or Fischer’s exact test and outcome (see Table 2). Seni et al. BMC Research Notes 2013, 6:298 Page 4 of 7 http://www.biomedcentral.com/1756-0500/6/298 Figure 1 Proportion of bacterial isolates from patients with surgical site infections (N = 304). Other g- bacteria: Enterobacter cloacae (10), Proteus mirabilis (7), Morganella morganii (6), Providencia spp (5), Citrobacter freundii (4), Serratia marcescens (1) and Leclercia adecarboxylata (1); Other g + bacteria: Streptococcus pyogenes (3), Streptococcus agalactiae (1), and Streptococcus spp (4).Acinetobacter spp: Acinetobacter baumanii (48) and Acinetobacter baumanii-calcoaceticus complex (4); Klebsiella spp: Klebsiella pneumonia (35) and Klebsiella oxytoca (4); Enterococcus spp: Enterococcus faecalis (21) and Enterococcus faecium (2). Table 1 Antimicrobial resistance pattern among isolates from surgical site infections Antibiotics Bacteria isolates (N = 304) E.coli Klebsiella spp Acinetobacter P.aeruginosa Other g- S. aureus Enterococcus Other g + (n = 72) (n = 39) spp (n = 52) (n = 12) (n = 34) (n = 64) spp (n = 23) (n = 08) Ampicillin 100.0% 100.0% NA NA 97.1% 100.0% 30.0% 50.0% Piperacillin NA NA 100.0% 66.7% NA NA NA NA TZP 30.6% 38.5% 53.8% 16.7% 32.4% NA NA NA AMC 90.3% 100.0% NA NA 94.1% NA NA NA TMP-SMX 86.1% 92.3% 98.0% 100.0% 94.1% 89.1% NA 80.0% Tetracycline 72.2% 76.9% 65.4% 100.0% 82.4% 42.2% 74.0% 62.5% Ciprofloxacin 72.2% 66.6% 77.0% 16.7% 47.1% 29.7% 60.9% 50.0% Chloramphenicol 41.7% 71.8% NA NA 70.6% 15.6% 30.4% 37.5% ** Gentamicin 54.2% 76.9% 88.5% 16.7% 76.5% 18.8% 21.7% 50.0% Amikacin 4.2% 2.6% 32.7% 8.3% 0.0% NA NA NA Ceftriaxone 77.8% 92.3% NA NA 67.6% NA NA NA Ceftazidime 77.8% 87.2% 90.4% 25.0% 61.8% NA NA NA Cefepime 69.4% 87.2% 61.5% 16.7% 38.2% NA NA NA Imipenem 0.0% 0.0% 3.9% 0.0% 0.0% NA NA NA Erythromycin NA NA NA NA NA 46.9 % 65.2% 50.0% Clindamycin NA NA NA NA NA 40.6% NA 62.5% Vancomycin NA NA NA NA NA 0.0% 4.4% 0.0% * + × ** Piperacillin-tazobactam, Amoxicillin-clavulanate, Trimethoprim-sulphamethoxazole, High level resistance screening, NA: Not applicable, spp: species, g+: Gram-positive bacteria [Streptococcus pyogenes (3), Streptococcus agalactiae (1), and Streptococcus spp (4)] and g-: Gram-negative bacteria [Enterobacter cloacae (10), Proteus mirabilis (7), Morganella morganii (6), Providencia spp (5), Citrobacter freundii (4), Serratia marcescens (1) and Leclercia adecarboxylata (1)]. Seni et al. BMC Research Notes 2013, 6:298 Page 5 of 7 http://www.biomedcentral.com/1756-0500/6/298 Table 2 Association of clinico-demographic characteristics of patients with and without bacterial isolates from surgical site infections Variable Bacteria isolated No bacteria isolated OR [95% CI] P-value from SSI (n = 216) from SSI (n = 98) Age (years) ≤ 19 38 (67.9%) 18 (32.1%) 1 20 – 39 133 (63.6%) 76 (36.4%) 0.83(0 .42- 1.61) 0.5577 40 – 59 31 (91.2%) 3 (8.8%) 4.89( 1.24- 27.87) 0.0112 ≥ 60 14 (93.3%) 1 (6.7%) 6.63( 0.86- 295.67) 0.0478 Sex Male 61 (81.3%) 14 (18.7%) 1 Female 155 (64.9%) 84 (35.1%) 0.42(0 .21- 0.82) 0.0072 Ward type Obst & Gyn 115 (59.0%) 80 (41%) 1 General Surgery 89 (85.6%) 15 (14.4%) 4.13( 2.17- 8.22) 0.0000 Orthopedic 12 (80.0%) 3 (20.0%) 2.78(0.72- 15.79) 0.1085 Operation Elective 57 (83.8%) 11 (16.2%) 1 Emergency 159 (64.6%) 87 (35.4%) 0.35( 0.16- 0.73) 0.0025 Duration from admission to operation ≤ 1 day 147 (65.3%) 78 (34.7%) 1 > 1 days 69 (77.5%) 20 (22.5%) 1.83( 1.01- 3.42) 0.0356 Duration from admission to discharge ≤ 14days 48 (58.5%) 34 (41.5%) 1 > 14days 110 (82.7%) 23 (17.3%) 3.39( 1.73- 6.68) 0.0001 Antibiotic chemoprophylaxis CRO 124 (62.9%) 73 (37.1%) 1 * ** CRO +MZ 70 (76.9%) 21 (23.1%) 1.96( 1.08- 3.65) 0.0187 Others 22 (84.6%) 4 (15.4%) 3.24( 1.04- 13.37) 0.0289 Duration of Postoperative antibiotics ≤ 3days 48 (50.5%) 47 (49.5%) 1 > 3days 168 (76.7%) 51 (23.3%) 3.23( 1.87- 5.54) 0.0001 Outcome Improved 147 (72.1%) 57 (27.9%) 1 Died 10 (100.0%) 0 (0.0%) - 0.042 * ** × Ceftriaxone, Metronidazole, Ampicloxacillin, Ciprofloxacin & Gentamicin. On multivariate logistic regression analysis, longer dur- Discussion ation from admission to discharge, longer duration of Of the 314 patients with clinical SSIs enrolled in this postoperative antibiotics and outcome were statistically study; the mean age (29.7 ± 13.14 years), higher female associated with bacteria isolation from SSIs (see Table 3). proportion (76.1%) and preponderance of admission in Of ten patients who died, 70% (7/10) had ESBL producing obstetrics and gynecology wards (62.1%) was similar to Enterobacteriaceae, four of which were E. coli. another study [11], but the proportion of female in other studies were between 30% to 60% and wards of admis- sion varied reflecting the underlying surgical procedures [8,9,13]. Caesarean section and laparotomy accounted for Table 3 Multivariate logistic regression analysis for more than three quarters of all surgical procedures in this factors associated with bacterial isolation from SSIs study whereas other similar studies have also found these Variable OR 95% CI p-value procedures to be quite common [8,11,27]. The predomin- Age (years) 1.16 0.74 - 1.80 0.524 ance of SSIs cases in obstetrics and gynecology wards is Sex 1.41 0.49 - 4.02 0.522 quite alarming and thus, a need to institute stringent in- Wards 0.44 0.18 - 1.10 0.078 fection prevention and control measures in this setting, Operation 0.62 0.24 - 1.65 0.343 more especially in emergency surgeries which accounted Duration from admission to operation 0.89 0.40 - 1.97 0.773 for more SSIs cases as opposed to elective surgeries. As noted from other studies [8,9,11,28], S. aureus and Duration from admission to discharge 1.93 1.01 - 3.72 0.047 E. coli were the most common bacterial isolates from Antibiotic chemoprophylaxis 1.27 0.77 - 2.12 0.351 SSIs. However their sources remain unknown from the Duration of postoperative antibiotics 2.52 1.45 - 4.36 0.001 present study though other studies have documented Outcome 0.52 0.30 - 0.91 0.021 both endogenous and exogenous sources from hospital Seni et al. BMC Research Notes 2013, 6:298 Page 6 of 7 http://www.biomedcentral.com/1756-0500/6/298 environment could be potential niches [29,30]. Similar reported as negative cultures. This was because of lack of to other related studies [9,18,31], high level of resistance standardized in-house detection methods and lack of an- was found among commonly used antibiotics like ampi- aerobic detection panels in the Phoenix Automated instru- cillin, trimethoprim-sulphamethoxazole, and tetracycline ment (Becton-Dickson, Sparks Maryland) that we used. in both Gram-positive and Gram-negative bacteria. Gram- negative bacteria showed more resistance to gentamicin, Conclusion ciprofloxacin and chloramphenicol as compared to Gram- Most SSIs at Mulago National Hospital are due to MDR positive bacteria. With exception of Acinetobacter spp,all bacteria, these are significantly more among Gram-negative Gram-negative bacteria displayed low resistance rates to than Gram-positive bacteria. Isolation of MRSA and ESBL- piperacillin-tazobactam, amikacin and imipenem. There producing Enterobacteriaceae in higher proportions than were also low resistance rates of Acinetobacter spp and previously reported calls for enhanced antibiotic steward- P. aeruginosa to colistin. These findings are similar to ship including laboratory guided SSIs-therapy and strength- another study [17]. The present study has shown that ening of infection control surveillance by identifying the rates of resistance to erythromycin (46.9%) among S. sources of these MDR isolates. In the light of these findings, aureus was relatively low compared to that of Entero- there is a need to investigate whether there is clonal spread coccus spp (65.2%), with excellent performance of vanco- of the predominant bacteria within/or among surgical mycin on both S. aureus and Enterococcus spp. These wards at Mulago National Hospital. findings are in agreement to another study [18]. This study found more MDR among Gram-negative bacteria than Additional file Gram-positive bacteria; of these the proportion of MRSA (37.5%) among S. aureus isolates was more than previously Additional file 1: Appendix i. Questionnaire. reported (25.0% and 31.5%) from Uganda [9,21] and other countries [8,16,18]. Thus, while β-lactamase-resistant anti- Competing interests biotics such as cloxacillin could still be effective in this The authors declare that they have no competing interests. setting, they are likely to be ineffective against the 38% Authors’ contributions of isolates that were confirmed as MRSA. The fact that Conceived and designed the experiments: JS, DPK, and FB. Specimen we found no vancomycin resistance among S. aureus iso- collection: JS. Supervised the clinical component of research: PM. Performed lates shows that this drug remains the last resort in sys- the experiments: JS. Supervised the laboratory component of research: CFN, DPK, and FB. Analyzed the data: JS, DPK, AK, and FB. Contributed reagents, temic infections caused by MRSA in this setting. Previous materials and analysis tools: JS, CFN, HK, and MLJ. Wrote the manuscript: JS, studies [11,17] have shown low rates (14% to 22%) of CFN, DPK, PM, MLJ, HK, AK, and FB. All authors have read and approved the ESBL-producers among Enterobacteriaceae isolates but the final manuscript. present study and another similar study in the same region Acknowledgements [8] have shown ESBL-producers to account more than The authors would like to thank patients and health workers in all surgical wards; three quarter of Enterobacteriaceae. This can be attributed Emmanuel Aboce, Tonny Lugya, and Hannington Baluku for excellent technical to the empirical use of third generation cephalosporins assistance; Willy Ssengooba for statistical inputs, and all staffs in the Department of Medical Microbiology for their support. This work was funded by Catholic (usually in combination with another drug such as genta- University of Health and Allied Sciences Bugando, Mwanza-Tanzania to JS. micin) in almost all hospitalized patients and lack of anti- microbial resistance surveillance in surgical wards at Author details Department of Medical Microbiology, Makerere University College of Health Mulago National Hospital. This is a major threat to patient Sciences, P.O. Box 7072, Kampala, Uganda. Department of Surgery, Makerere care as ESBL production renders use of these ceftriaxone University College of Health Sciences, P.O. Box 7072, Kampala, Uganda. or ceftazidime useless. Absence of resistance to imepenem Department of Microbiology and Immunology, Catholic University of Health and Allied Sciences-Bugando, P.O. Box 1464, Mwanza, Tanzania. Department among these isolates is however a good finding, except that of Community Medicine, Catholic University of Health and Allied Sciences- imepenem remains an expensive drug. Bugando, P.O. Box 1464, Mwanza, Tanzania. Multivariate logistic regression analysis of clinical and Received: 23 November 2012 Accepted: 25 July 2013 demographic characteristics of patients with SSIs in this Published: 27 July 2013 study showed that longer duration from admission to dis- charge, longer duration of postoperative antibiotics and References outcome (death) were associated with bacteria isolation 1. Horan TC, Gaynes RP, Martone WJ, Jarvis WR, Emori TG: CDC definitions of nosocomial surgical site infections, 1992: a modification of CDC from SSIs. These findings have also been shown in other definitions of surgical wound infections. Am J Infect Control 1992, similar studies [5,11,27]. 20(5):271–274. 2. Mangram AJ, Horan TC, Pearson ML, Silver LC, Jarvis WR: Guideline for prevention of surgical site infection, 1999. Centers for Disease Control Limitation and Prevention (CDC) Hospital Infection Control Practices Advisory The study did not isolate strict anaerobes, which could Committee. Am J Infect Control 1999, 27(2):97–132. quiz 133–134; have increased the number of bacterial isolates currently discussion 196. Seni et al. BMC Research Notes 2013, 6:298 Page 7 of 7 http://www.biomedcentral.com/1756-0500/6/298 3. Emori TG, Gaynes RP: An overview of nosocomial infections, including the 23. Koneman EW, Allen SD, Janda WM, Schreckenberger PC, Winn WC: Color role of the microbiology laboratory. Clin Microbiol Rev 1993, 6(4):428–442. atlas and textbook of diagnostic microbiology. 5th edition. Philadelphia, 4. Fabiano G, Pezzolla A, Filograna MA, Ferrarese F: [Risk factors of surgical Pa: Lippincott, Williams & Wilkins Publishers; 1997. wound infection]. Ann Ital Chir 2004, 75(1):11–16. 24. CLSI: Perfomance standars for antimicrobial susceptibility testing; twenty first 5. Astagneau P, Rioux C, Golliot F, Brucker G: Morbidity and mortality information supplement, vol. CLSI document M100-S21. Wayne, PA: Clinical associated with surgical site infections: results from the 1997–1999 and Laboratory Standards Institute; 2011. INCISO surveillance. J Hosp Infect 2001, 48(4):267–274. 25. Livermore DM, Brown DF: Detection of beta-lactamase-mediated 6. Barwolff S, Sohr D, Geffers C, Brandt C, Vonberg RP, Halle H, Ruden H, resistance. J Antimicrob Chemother 2001, 48(Suppl 1):59–64. Gastmeier P: Reduction of surgical site infections after Caesarean delivery 26. D'Agata EM: Rapidly rising prevalence of nosocomial multidrug-resistant, using surveillance. J Hosp Infect 2006, 64(2):156–161. F-negative bacilli: a 9-year surveillance study. Infect Control Hosp Epidemiol 2004, 25(10):842–846. 7. Santos KR, Fonseca LS, Bravo Neto GP, Gontijo Filho PP: Surgical site 27. Razavi SM, Ibrahimpoor M, Sabouri Kashani A, Jafarian A: Abdominal infection: rates, etiology and resistance patterns to antimicrobials surgical site infections: incidence and risk factors at an Iranian teaching among strains isolated at Rio de Janeiro University Hospital. Infection hospital. BMC Surg 2005, 5:2. 1997, 25(4):217–220. 28. Kasatpibal N, Jamulitrat S, Chongsuvivatwong V: Standardized incidence 8. Mawalla B, Mshana SE, Chalya PL, Imirzalioglu C, Mahalu W: Predictors of rates of surgical site infection: a multicenter study in Thailand. Am J surgical site infections among patients undergoing major surgery at Bugando Infect Control 2005, 33(10):587–594. MedicalCentreinNorthwesternTanzania. BMC Surgery 2011, 11:21. 29. Wenzel RP, Perl TM: The significance of nasal carriage of Staphylococcus 9. Anguzu JR, Olila D: Drug sensitivity patterns of bacterial isolates from aureus and the incidence of postoperative wound infection. J Hosp Infect septic post-operative wounds in a regional referral hospital in Uganda. 1995, 31(1):13–24. Afr Health Sci 2007, 7(3):148–154. 30. Atata RF, Ibrahim YKE, Olurinola PF, Giwa A, Akanbi AA II, Sani AA: Clinical 10. Whitehouse JD, Friedman ND, Kirkland KB, Richardson WJ, Sexton DJ: The bacterial isolates from hospital environment as agents of surgical wound impact of surgical-site infections following orthopedic surgery at a nosocomial infections. Journal of Pharmacy & Bioresources 2010, 7(2):146–155. community hospital and a university hospital: adverse quality of life, 31. Adegoke AA, Mvuyo T, Okoh AI, Steve J: Studies on multiple antibiotic excess length of stay, and extra cost. Infect Control Hosp Epidemiol 2002, resistant bacterial isolated from surgical site infection. Scientific Research 23(4):183–189. and Essays 2010, 5(24):3876–3881. 11. Fehr J, Hatz C, Soka I, Kibatala P, Urassa H, Smith T, Mshinda H, Frei R, Widmer A: Risk factors for surgical site infection in a Tanzanian district doi:10.1186/1756-0500-6-298 hospital: a challenge for the traditional National Nosocomial Infections Cite this article as: Seni et al.: Antimicrobial resistance in hospitalized Surveillance system index. Infect Control Hosp Epidemiol 2006, 27(12):1401–1404. surgical patients: a silently emerging public health concern in Uganda. 12. Kirkland KB, Briggs JP, Trivette SL, Wilkinson WE, Sexton DJ: The impact of BMC Research Notes 2013 6:298. surgical-site infections in the 1990s: attributable mortality, excess length of hospitalization, and extra costs. Infect Control Hosp Epidemiol 1999, 20(11):725–730. 13. Engemann JJ, Carmeli Y, Cosgrove SE, Fowler VG, Bronstein MZ, Trivette SL, Briggs JP, Sexton DJ, Kaye KS: Adverse clinical and economic outcomes attributable to methicillin resistance among patients with Staphylococcus aureus surgical site infection. Clin Infect Dis 2003, 36(5):592–598. 14. Wassef MA, Hussein A, Abdul Rahman EM, El-Sherif RH: A prospective surveillance of surgical site infections: study for efficacy of preoperative antibiotic prophylaxis. African Journal of Microbiology Research 2012, 6(12):3072–3078. 15. Amare B, Abdurrahman Z, Moges B, Ali J, Muluken L, Alemayehu M, Yifru S, Sendek B, Belyhun YFM, et al: Postoperative surgical site bacterial infections and drug susceptibility patterns at Gondar University Teaching Hospital, Northwest Ethiopia. J Bacteriol Parasitol 2011, 2:126. 16. Throckmorton AD, Baddour LM, Hoskin TL, Boughey JC, Degnim AC: Microbiology of surgical site infections complicating breast surgery. Surg Infect (Larchmt) 2010, 11(4):355–359. 17. Hawser SP, Bouchillon SK, Hoban DJ, Badal RE: Epidemiologic trends, occurrence of extended-spectrum beta-lactamase production, and performance of ertapenem and comparators in patients with intra-abdominal infections: analysis of global trend data from 2002–2007 from the SMART study. Surg Infect (Larchmt) 2010, 11(4):371–378. 18. Kownhar H, Shankar EM, Vignesh R, Sekar R, Velu V, Rao UA: High isolation rate of Staphylococcus aureus from surgical site infections in an Indian hospital. J Antimicrob Chemother 2008, 61(3):758–760. 19. Krukerink M, Kievit J, de Mheen PJ M-v: Evaluation of routinely reported Submit your next manuscript to BioMed Central surgical site infections against microbiological culture results: a tool to and take full advantage of: identify patient groups where diagnosis and treatment may be improved. BMC Infect Dis 2009, 9:176. • Convenient online submission 20. Olaro C: Study to assess the risk factors for postoperative complications • Thorough peer review following abdominal surgery in Mulago Hospital. Kampala: Makerere University; 1999. • No space constraints or color figure charges 21. Ojulong J, Mwambu TP, Joloba M, Bwanga F, Kaddu-Mulindwa DH: Relative • Immediate publication on acceptance prevalence of methicilline resistant Staphylococcus aureus and its susceptibility pattern in Mulago Hospital, Kampala, Uganda. Tanzan J • Inclusion in PubMed, CAS, Scopus and Google Scholar Health Res 2009, 11(3):149–153. • Research which is freely available for redistribution 22. Kitara DL, Kakande I, Mugisa BD, Obol JH: The postoperative complications prediction in Mulago Hospital using POSSUM scoring system. East and Submit your manuscript at Central African Journal of Surgery 2010, 15(2):90–96. www.biomedcentral.com/submit

Journal

BMC Research NotesSpringer Journals

Published: Jul 27, 2013

There are no references for this article.