Surgical Site Infection after Digestive Surgery: Diagnosis and Treatment in a Context of Limited Resources

Patients and Methods

The aim of this study was to determine the incidence, risk factors, and bacteriologic profile of SSIs in the Department of Digestive Surgery at Tenkodogo Regional Hospital in Burkina Faso. We performed a prospective study that covers the period from January 1, 2016 to December 31, 2016. Patient preparation was based on the World Health Organization recommendations. The following were considered. A pre-operative shower was given only to patients who underwent scheduled surgery. Depilation was performed on the inguinal and abdominal incision sites using an electric razor. Polyvidone iodine was used in the skin preparation procedure. Disinfection of the surgeon's hands was performed by washing with soap followed by hydro-alcoholic friction. The anesthetic protocol included ceftriaxone administered 30 minutes before the surgical incision was made. The dose of ceftriaxone was 2 g for adults and 50 mg/kg for children. A sterile dressing was placed on all incisions at the end of the procedure.

All patients who underwent surgery of the digestive tract during this period were included in this study and followed. The initial surgeries were classified according to Altemeier classification [10]. For each patient included, we estimated the annual income. This led to the identification of patients living below and above the poverty line. When a person had an income of less than $1.25 per day, he or she was determined to live below the poverty line.

After the surgery, the patients were hospitalized. The first inspection of the surgical site was performed on the third day after surgery, and the appearance of the incision was noted. Then, the dressing was changed every three days until healing. These dressings were made in an annex infirmary of the hospital. In each external dressing, the aspects of the incision were noted on the patient's clinical record. When there were signs of an SSI, the surgeon responsible for the patient was alerted and the patient was seen immediately. After re-examination of the patient it was the surgeon who made the diagnosis of SSI. Patients whose follow-up was complicated by SSI were then identified. Thus, patients included in the study were divided into two groups: patients with SSI and patients without SSI. We compared the data of these two groups of patients.

Surgical site infection was diagnosed as outlined in the CDC definition of nosocomial surgical site infections [1]. A bacteriologic study was performed for all patients with SSI. The sampling technique for the bacteriologic study depended on the type of SSI. In the case of intra-abdominal suppuration, pus was collected for each operation using a sterile syringe. In the case of superficial suppuration of the incision, pus was collected directly in a sterile syringe. In the absence of free pus, an incision swab was performed with a sterile swab provided by the laboratory.

The pus was packaged aseptically and sent to the laboratory within a maximum of 10 minutes. The samples were gram-stained and examined for micro-organisms after culture of the samples on aerobic and anaerobic blood agar, as well as in cooked aerobic and anaerobic broth. The antibiogram was performed on agar medium. If the cultures were negative after three days, they were maintained for up to eight days. The 2011 recommendations of the Comite de l'Antibiogramme de la Societe Francaise de Microbiologie (CASFM) were used to interpret the results of the antibiogram. For each included patient, epidemiologic, clinical, and bacteriologic characteristics were collected using a pre-established survey form. Data analysis was performed on a microcomputer using the EPI INFO software (Centers for Disease Control and Prevention, Atlanta, Georgia). The χ² test was used for comparison of the data with a significance level of p = 0.05.

Results

Discussion

The incidence of SSI in our study was 11.8% (114/964). This result is similar to that found by other authors who published studies on patients in sub-Saharan Africa [11,12]. Our study confirms the assertion by the World Health Organization that the prevalence of SSI is higher in low-income countries [3]. In fact, the incidence of SSI in surgery departments is approximately 6% in Europe [3,13]. The difference of this incidence with ours could be explained by better preparation of the patients and by the best hygienic conditions in European hospitals. It should be noted, however, that the types of surgery performed in surgical departments differ from one hospital to another, which makes comparison of the frequencies difficult. Additionally, the incidence of SSI differs from one surgical procedure to another.

The percentage of SSIs is generally higher after interventions in the abdomen and perineum [13]. Similar to other authors from Africa, we found a substantial predominance of SSI in women. This finding is quite different from what has been reported in Western countries, where there is no relative risk related to gender [14, 15]. The higher prevalence of SSI among females could be explained by the level of poverty. In sub-Saharan Africa, the best-paid jobs are held by males. Girls are less educated than boys. As a result, the income for females is lower than that of males. We also highlighted other socio-demographic risk factors, such as geographical origin and poverty line. These demographic factors are almost exclusively reported in African series [11,12]. In sub-Saharan Africa, patients in precarious financial situations often opt for traditional medicine and self-medication. They are likely to have experienced complications by the time they arrive at the hospital. These socio-demographic factors are not found in the European and Asian series because social security programs ensure access to care for all [13].

In our study, the incidence of SSI was higher in emergency surgery compared with scheduled surgery. Our study has been characterized by various emergency surgical procedures, such as intestinal perforations, gastric perforation, and intestinal obstruction. This multitude of emergency surgeries makes it difficult to compare the incidence of SSI with Western countries or emergency interventions that are less varied [14,15]. The incidence was also higher in contaminated surgery than in clean surgery. These risk factors, which are related to the urgency of the intervention and the initial contamination of the surgical site, are universally found in both Africa and Europe [16,17]. Other risk factors described in the literature, such as depilation, antibiotic prophylaxis, length of pre-operative stay, and duration of surgery were not evaluated in our study. In 2016, the World Health Organization developed recommendations for the prevention of SSIs. However, these recommendations are not all applied in Burkina Faso. Full implementation of these recommendations would reduce the prevalence of SSIs in the African context substantially. The culture was negative in 54 of our patients, which equates to half of the cultures realized. These findings are apparent in many studies of SSI in sub-Saharan Africa where causative organisms are isolated in approximately 54% of cases [2]. This situation could be explained by sampling and sample conditioning techniques in Africa. It could also be explained by the precarious working conditions of the laboratories in our context.

Escherichia coli was the most implicated microbe in our study. In Europe, South America, and parts of Africa, a predominance of Staphylococcus aureus and Klebsiella aerogenes is seen [18,19]. The predominance of Escherichia coli in our context can be explained by the frequency of abdominal surgery, especially peritonitis, caused by digestive perforation. At our hospital, most cases of peritonitis were caused by ileal perforation, which is a frequent complication of typhoid fever in a tropical environment. Typhoid fever is an endemic infectious disease in the rural environments of Burkina Faso. This variety of peritonitis leads to an opening in the digestive tract, which in turn results in massive contamination of the peritoneal cavity and the surgical site by intestinal flora, particularly Escherichia coli. The predominance of non-MDR bacteria, especially non-MDR Staphylococcus aureus, suggests contamination from the bacterial flora of the patient or surgeon. The resistance of bacteria to ceftriaxone is increasingly reported in sub-Saharan Africa. This could be explained by the misuse of this antibiotic with respect to indications, duration, and dose. It would therefore be necessary to review the entire chain of preparation of the surgical site, particularly through the introduction of this practice to our department, and particularly the practice of non-depilation of patients.

Conclusion

Our study found that the frequency of SSIs was high in rural Burkina Faso. The microbes involved were mostly endogenous bacteria. The unique aspect of our situation is the existence of socio-demographic risk factors such as gender, geographical origin, and socio-economic status. Other risk factors that are reported commonly in the literature, such as the adequacy of pre-operative antibiotic prophylaxis and the absence of pre-operative depilation, should be evaluated to help reduce the frequency of post-operative infections in Africa.