Urinary catheters are passed to permit drainage of urine . They may have diagnostic or therapeutic uses . Globally, about two-thirds of urinary tract catheterization among adults is for therapeutic reasons in order to relief bladder outlet obstruction due to benign prostatic obstruction    . The risk of bacterial Urinary Tract Infections (UTI) is dependent on the patient’s susceptibility, the quality of catheter and how long catheter has been in place  .
Catheter Associated Bacteria Urinary Tract Infection (CABUTI) occurs in at least 40% of hospital-acquired infections . CABUTI has been associated with substantial morbidity in acute care settings and extended care facilities at rates of 20% and 50% respectively . CABUTI rates vary widely from up to 5% for single brief catheterization to 100% for indwelling catheters over a duration of 4 days . Female, advanced age and the critically ill are known risk factors .
In clinical practice, varied microorganisms may be associated with CABUTI and these include: Escherichia coli, Klebsiella spp., Proteus, Enterococci, Pseudomonas, Enterobacter, Serratia and Candida  . Within a couple of days after insertion of catheter, bacteria may migrate to bladder from biofilms formed on the surface of indwelling catheter  . Commonly, biofilms are initially caused by a single species of bacteria and may eventually become polymicrobial and resistant to various antimicrobial agents especially following long-term catheterization  . Worldwide, antimicrobial resistance due to CABUTI contributes substantially to a rise in morbidity and mortality as well as high cost of health care delivery. The epidemiology remains variable and health facility dependent .
Understanding the local antibiotic susceptibility pattern in our setting could enable practitioners to select the appropriate medication necessary for effective treatment. Therefore, this study was carried out to determine the prevalence of CABUTI and the antimicrobial susceptibility pattern at Tamale Teaching Hospital.
2.1. Study Type
This is a cross-sectional study conducted among patients who had catheter inserted into urinary bladder from November 2015 to April 2016.
2.2. Study Site
Tamale Teaching Hospital, an 800 bed capacity tertiary hospital in Tamale, Northern Region, Ghana.
2.3. Patients’ Recruitment and Specimen Collection
From November 2015 to April 2016, a cross sectional study was undertaken at the urology clinic of the Tamale Teaching Hospital of Ghana. The eligibility criteria included patients who had catheter in situ and who consented to be part of the study. We excluded, immunosuppressed patients, non-catheterized patients, those who had confirmed UTI just preceding this study, those taking antibiotic prophylaxis prior to catheterization and those who declined consent.
Eligible patients who consented to be part of the study were assigned unique Identification (Id) numbers. Data was recorded on a well-designed sheet. Data fields included: age, sex, address marital status and indication for catheterization.
A spigot was placed at tip of catheter and opened when the patient experienced the sensation to void, associated with a suprapubic mass, which was indicative of a full bladder. The spigot was then removed to allow about 10 - 20 ml of urine to flow through and drop off. This was to ensure clean urine was obtained devoid of contamination. Urine collection was done under aseptic conditions into a sterile, dry, leak-proof container. About 2 - 5 ml was collected from the tip of catheter. The containers were labeled with the patient’s identification number, age, sex, date and the time of collection. The urine specimen was transported together with the data collection form and delivered to the bacteriology laboratory for culture, biochemical tests, isolation and antibiotic susceptibility tests.
Using a sterile calibrated wire loop and under aseptic conditions, about 0.01 ml of urine was inoculated onto a prepared agar plate of Cystine Lactose Electrolyte Deficient (CLED). The plate was incubated under aerobic conditions at 37˚C for 24 hours and observed for bacteria growth. Significant growth of >105 bacteria/ml of catheter urine was interpreted as a colony of bacteria with a viable count . Bacteria colonies were identified using colony growth characteristics and Gram staining as well as standard biochemical testing procedures which included indole, urea, triple sugar Iron (TSI), motility and citrate tests were all carried out in accordance with Monica Chessbrough .
An emulsification was made in bijou bottle containing 5 ml peptone water with the pure colonies, until the turbidity was equal to the 0.5 McFarland standards. An approximately 200 µl/loopful of the suspension was dispensed to the center of 25 ml Muller-Hinton culture plate and seeded carefully with the sterile swab stick in three directions to obtain even growth on the Muller-Hinton agar surface, allowing the moisture to be absorbed for at least 15minutes. Using the disc diffusion method of antimicrobial susceptibility test, the urine antibiotics multi-discs (manufactured by Axiom Laboratories, India) were applied firmly to the surface of the Mueller-Hinton agar plate. The antibiotics multidisc comprised of Ampicillin (AMP, 20 mcg), Ceftizoxime (CL, 30 mcg), Ciprofloxacin (CP, mcg), Amikacin (AMK, 30 mcg), Cotrimoxazole (BA, 25 mcg), Cephalexin (PR, 30 mcg), Tetracycline (TE, 30 mcg), Levofloxacin (LE, 5 mcg), Ofloxacin (OF, 5 mcg), Norfloxacin (NX, 10 mcg), Chloramphenicol (CH, 30 mcg), Sparfloxacin (SC, 5 mcg), Gentamicin (GEN, 30 mcg), Ceftriaxone (CTR, 30 mcg), Cefuroxime (30 mcg).
The set-up was incubated aerobically at 37˚C for 18 - 24 hrs, after which it was inspected for bacteria growth and growth inhibition. The diameter of the zone of growth inhibition around each antimicrobial agent was measured and compared with the NCCLS interpretive table, NCCLS, 1997 to determine bacterial sensitivity or resistance to each of the antimicrobial agents used . Standard commercial bacteria strains comprising of Staphylococcus aureus NCTC 6571, Escherichia coli NCTC 10418 and Pseudomonas aeruginosa NCTC 10662 were used as control.
Data was entered into Microsoft excel spreadsheet windows 7 and checked for data entry errors. Data analysis was carried out using IBM SPSS version 21 statistical package. Associations between variables were determined with level of significance set at p < 0.05.
One hundred and twenty-two patients participated in the study. There were 73 (59.8%) males and 49 (40.2%) females. The median age was 42.5 (range 33 - 65) years. There were 37 (30.3%) participants in the modal age group 51 - 60 years (Table 1). The highest number of bacterial isolates 22 (25.0%) out of the 88 positive culture results were in age group 31 - 40 years. Frequency of urine culture isolates and age were not statistically significant (p = 0.35) (Table 2).
Significant bacterial growth was obtained in 88 (72.1%) of the urine sample cultured, of which males constituted 48 (54.5%) and females 40 (45.5%). This was not statistically significant (p = 0.06). The relationship between sensitivity
Table 1. The socio-demographic characteristics of patients with catheter (n = 122).
Table 2. Urine culture results among various age groups.
df: degrees of freedom.
pattern of bacteria to antibiotics and sex category of the patient was not significant (p = 0.06).
Thirteen different organisms were identified from 88 (72.1%) culture positives of which 81 (92.1%) were Gram-negative bacteria, 6 (6.8%) Gram-positive bacteria and in 1 (1.1%) case of Candida albicans was isolated. The predominant bacteria isolates were Escherichia (E.) coli 41 (46.6%), Klebsiella spp. 18 (20.6%), Pseudomonas spp. 10 (11.4%), Enterobacter 6 (6.8%) and Staphylococcus aureus 5 (5.8%). Other organisms isolated 8 (8.8%) include: Citrobacterdiversus, morgani, Enterococcus spp., Klebsiellaoxytoca, Streptococcus spp., Proteus mirabilis and Candida albicans (Table 3).
All isolates demonstrated sensitivity to Amikacin. Pseudomonas aeroginosa demonstrated the least sensitivity to Amikacin (70%) while Enterobacter spp. and Staphylococcus aureus were the most sensitive (100%). Gentamycin and cefuroxime showed the least sensitivity pattern; as they were sensitive to only one isolate each of Escherichia coli. The remaining drugs on the antibiotics multidisc showed variable sensitivity pattern (Table 4).
Among the isolates, the best sensitivity pattern to majority of the antimicrobials was observed for Escherichia coli. Escherichia coli isolates demonstrated sensitivity to Amikacin 35 (85.4%), Levofloxacin 17 (41.5%) and Ciprofloxacin 16 (39%) in descending order. None of the isolates was susceptible to Ampicillin. Klebsiella spp. 15 (83.3%) exhibited sensitivity to Amikacin but resistant to Ampicillin, Gentamicin, Cefuroxime and Cephalexin (Table 4).
Generally, the best susceptibility of bacteria isolates to antibiotics was observed in the following antimicrobials: Amikacin 73 (83.0%), Levofloxacin 34 (38.6%), Ciprofloxacin 26 (29.5%), Ceftizoxime 23 (26.1%), Orfloxacin 22 (25.0%) and Chloramphinicol 21 (23.9%) respectively in descending order. Overall, decreased susceptibility of bacteria to antibiotics was observed with Sparfloxacin 13 (14.8%), Cephalexin 8 (9.1%), Norfloxacin 7 (8.0%), Cotrimoxazole 7 (8.0%), Gentamicin 3 (3.4%), Ampicillin 3 (3.4%) and Cefuroxime 1 (1.1%), respectively (Table 5).
Table 3. Pattern of uropathogens isolated.
Table 4. Antimicrobial susceptibility patterns to most common bacterial isolates.
Table 5. Overall susceptibility of uropathogens.
The prevalence of CABUTI and antimicrobial susceptibility patterns among patients may vary from one setting to the other. In the United Kingdom, Wazait and associate found catheter associated urinary tract infection to be 35.5%. Koshariya and colleagues reported the prevalence of CABUTI in India to be 27% . In Nigeria, Taiwo et al. reported the prevalence of CABUTI of 13.3% and 98.8% when bladder catheter was in situ at less than 7 days or more than 7 days respectively . In this study, the prevalence of CABUTI was 72.1%. This could be due to contamination of urine by bowel flora or catheters were passed without adherence to strict aseptic protocols. Also, this high prevalence may be due to prolonged catheterization as the duration of catheterization prior to sample collection was not established by this study. Among females, the prevalence of CABUTI was 45.5%. There exist anatomical variations, between female and male urethra and meatus. The female urethra is short and has a meatus closer to the anal opening. This poses a risk for females to contract CABUTI.
4.1. CABUTI Uropathogens Identification
Urine culture test has been used over decades for diagnosing patients who have UTI. The identification of CABUTI uropathogens and the antibiotic susceptibility patterns enable practitioners select the appropriate antibiotics for treatment. Escherichia coli, the leading uropathogen in urine cultures and other Enterobacteriaceae, account for approximately 75% of all uropathogens   . Included in the top five uropathogens were Escherichia coli 30.5%, Klebsiella pneumoniae 30.5%, Pseudomonas aeruginosa 16.6% and Candida spp. 16.6% as reported by Kazi and colleaques . This present study revealed that Escherichia coli 46.6%, Klebsiella spp. 20.6%, Pseudomonas spp. 11.4%, Enterobacter 6.8% and Staphylococcus aureus 5.8% respectively were the most prevalent CABUTI uropathogens at the Tamale Teaching Hospital. The least prevalent CABUTI uropathogens were Citrobacter diversus, Morganella morgani, Enterococcus spp., Klebsiellao xytoca, Streptococcus spp., Proteus mirabilis and Candida albicans each constituting 1 (1.1%). These organisms are mainly endogenous bowel flora. Thus, poor personal hygiene or non-adherence to aseptic technique during catheterization could account for the higher prevalence of these organisms in the urine of our study participants. It has been established that CABUTI is one of the health care associated infections that may be contracted through contact with contaminated equipment or solutions and from other patients or hospital staff   .
4.2. Uropathogen Antibiotics Susceptibility
Antibiotic use by patients prior to presentation of urine samples could significantly alter microbial yield and consequently prediction of infection rates because they suppress the endogenous bacteria flora  . Multiple studies have demonstrated resistance to a host of antibiotics including ampicillin, chloramphenicol, cotrimoxazole, gentamicin, cefuroxime   . This phenomenon of drug resistance differs from one place to another. Though there is antimicrobial resistance to a large extend, some studies show there were susceptibility of uropathogens to some antibiotics. In Ghana, Gyansa-Lutterodt and associates found high susceptibility of uropathogens to Nitrofuantoin and Gentamicin at the Police Hospital, while Gyasi-Sarpong et al reported susceptibility of uropathogens to ciprofloxacin, nalidixic acid, cefuroxime, ceftriaxone and cefotaxime at the Komfo Anokye Teaching Hospital  . In order of decreasing susceptibility, we report that Amikacin 83.0%, Levofloxacin 38.6% and Ciprofloxacin 29.5% were the antibiotics found to be most suitable in treatment of urinary tract infection at the Tamale Teaching Hospital. The existence of substantial resistance of bacterial isolates to antibiotics is therefore implied. This might be as a result of indiscriminate usage of these antibiotics resulting in resistance among the bacterial isolates. The bacterial isolates showed least susceptibility to Cefuroxime 1.1%, Ampicillin 3.4% and Gentamicin 3.4% in order of increasing susceptibility. Other antibiotics with least sensitivity were Co-trimoxazole 8.0%, Norfloxacin 8.0%, Cephalexin 9.1%, Sparfloxacin 14.8%, Chloramphenicol 23.9% and Ofloxacin 25.0% in order of increasing susceptibility. It is important for practicing clinicians to appraise their knowledge on the local antibiotics’ susceptibility patterns so as to effectively treat CABUTI.
This study had some limitations. Firstly, the duration of catheterization preceding sample collection for urine culture and sensitivity was not established. Secondly, the participants were not grouped into either catheter associated bacteriuria (asymptomatic) and catheter associated urinary tract infection (symptomatic). For catheter associated bacteriuria (asymptomatic), no treatment is usually needed. This is an important point to consider in determining antibiotic susceptibility and recommending treatment.
Catheter-associated bacterial urinary tract infection is prevalent at the Tamale Teaching Hospital. Micro bacterial isolates demonstrated substantial decrease in susceptibility to antibiotics commonly used. Understanding the local antibiotic susceptibility pattern could guide the choice of antibiotics used in treating catheter-associated bacterial urinary tract infection.
The authors are grateful to the Head of Department of the Tamale Teaching Hospital Laboratory for providing the laboratory space towards this work.