Treating Surgical Infections in Low- and Middle-Income Countries: Source Control,Then What?

Surgical infections represent a significant burden of disease globally. Complicated intra-abdominal infections (IAI) and soft tissue infections comprise a large proportion of such disease in low- and middle-income countries (LMICs) [1, 2]. In addition, surgical site infections (SSIs) are common, with rates exceeding 20% in some studies [2–4].

Etiologies of soft tissue infections include diabetic foot infections, trauma, peripheral vascular disease, and necrotizing soft tissue infections. Globally, the most common intra-abdominal infectious processes are appendicitis (33%), post-operative infection (16%), and cholecystitis (15%) [5]. However, there is variability regionally and at different institutions [6, 7]. Common causes of peritonitis at a tertiary referral in Rwanda were complications of intestinal obstruction (39%), appendicitis (17%), and trauma (15%)[6]. In a referral hospital in Malawi, the most common etiologies for peritonitis were appendicitis (22%), intestinal volvulus (17%), perforated peptic ulcer disease (11%), and small bowel perforation (11%)[7]. In children with acute generalized peritonitis in Nigeria, the most common diagnoses were typhoid intestinal perforation (51%) and intestinal obstruction with or without gangrene (13%) [8]. In Tanzania, the most common causes of secondary peritonitis were appendicitis (24%), peptic ulcer disease (19%), ischemia (19%), and typhoid perforation (15%) [9]. In a hospital in India, the most common causes of peritonitis were peptic ulcer disease (45%), appendicitis (19%), typhoid perforation (12%), tuberculosis (10%), and trauma (8%) [10]. In New Delhi, India, the most common causes of peritonitis were perforated duodenal ulcer (26%), typhoid ileal perforation (26%), small bowel tuberculosis (10%), stomach perforation (9%), and appendicitis (5%) [11].

The principles of management for infections are source control and antibiotics [12]. Source control includes any procedure to eliminate the infectious source, control any ongoing infection, correct anatomic derangements, and restore normal physiology [13]. This is accomplished by surgical or interventional means. In LMICs, because of a shortage or lack of interventional radiology facilities and personnel, the predominant means is open surgery. For intra-abdominal processes, the usual technique is exploratory laparotomy, although laparoscopy may be available. Source control in soft tissue infections ranges from debridement to amputation [14].

Globally, five billion people lack safe, timely, and affordable surgical care, with the greatest burden falling on those from LMICs [15]. Emergency cases predominate, and patients often present in delayed fashion. This situation can be complicated by malnutrition, advanced disease, and various co-morbidities or co-existing disease processes. It is estimated that provision of basic emergency and essential surgical services could avert 1.5 million deaths each year [16].

Providing surgical care in low-resource settings is challenging on numerous fronts and requires particular efforts to provide material and human resources. Material resources include pre-operative tools to make the diagnosis accurately and efficiently, proper instruments to conduct the operation, and the necessary materials for post-operative management. The needs include monitoring equipment such as functioning blood pressure cuffs and pulse oximeters. Inadequate supplies for monitoring may result in delayed recognition of a deteriorating patient. Diagnostic imaging modalities such as radiography, ultrasonography, and computed tomography can aid in pre-operative diagnosis as well as post-operative monitoring of complications. If intravenous antibiotics are required, appropriate tubing must be available. This requires greater material resources and increases the overall costs of care. Material resources also include a ready supply of dressing supplies for incisions. Inadequate or infrequent incision care and dressing changes may increase the risk of SSI and delay healing.

Human resources are nurses and physicians for routine assessments, which contribute to the ability to identify patients at risk for clinical deterioration. This includes adequate staff at all levels of care, including in the emergency department, operating room, recovery room, intensive care unit, wards, and clinic. Access to consulting services and a wide range of specialists can improve care, especially for complex cases. Interventional radiologists and gastrointestinal endoscopists provide a valuable resource for minimally invasive interventions such as percutaneous drainage of abscesses or endoscopic retrograde cholangiopancreatography, which could minimize the need for surgical intervention. Other human resources infrequently found in LMICs are ancillary staff such as physical therapists, respiratory therapists, nutritionists, and social workers. There are limited data quantifying the impact of these staff individually. However, they play critical roles in developing a robust multi-disciplinary workforce, and further studies are needed to determine the impact of these ancillary services on improving surgical care in LMICs.

Antibiotic management of surgical infections is initially empiric, with the drug given then tailored to culture data. A randomized controlled trial of four days versus longer-duration antibiotic therapy showed no difference in outcomes in patients with complicated IAI after adequate source control [17]. Inadequate or inappropriate antibiotics are associated with higher costs and more adverse outcomes [18].

Gram-negative pathogens cause a large burden of surgical infections in LMICs. In Ethiopia, 73% of SSIs were caused by gram-negative bacilli [19]. The most common organisms isolated from SSIs in Uganda were Escherichia coli (24%) and Staphylococcus aureus (21%) [20]. Common causative pathogens in necrotizing soft tissue infections (NSTI) are Streptococcus pyogenes and E. coli [14]. In Uganda, 91% of NSTI were caused by gram-negative pathogens, with the most common organism isolated being E. coli [21].

Worldwide, Enterobacteriaceae predominate in IAI. Gram-negative infections (72%) are most common, with E. coli (41%) and Klebsiella spp. (11%) constituting the greatest burden [5]. Globally, in the Study for Monitoring Antimicrobial Resistance Trends (SMART) surveillance program, the most common organisms isolated from IAI were E. coli (48%), Klebsiella (13%), and Pseudomonas (10%) [22]. Common gram-negative bacilli in IAI in the Asia Pacific region were E. coli (46%), Klebsiella pneumonia (19%) and P. aeruginosa (10%) [23]. In patients with peritonitis in Nigeria, the most commonly isolated pathogens were E. coli (31%), S. aureus (17%), and Klebsiella (11%) [24]. In India, among patients with peritonitis, the most commonly isolated organisms were E. coli (44%) and Klebsiella (25%) [25].

Defining adequate and appropriate empiric antibiotic therapy requires knowledge of likely organisms and the local antibiotic susceptibility profile of these organisms. This requires a locally functioning laboratory that collects and collates these data. It can be difficult to provide appropriate empiric antibiotics in regions with limited culture data, especially if the rate of resistance to commonly available antibiotics is high. If there are no or limited means to obtain culture data, antibiotic regimens cannot be scaled back, resulting in many patients receiving an unnecessarily long duration of broad-spectrum antibiotics.

Because of resource limitations, there often are few antibiotics available in LMICs. There is a particularly low availability of injectable agents such as ampicillin, gentamicin, and penicillin in first-line hospitals in Africa and Asia [26]. There may be a great amount of variability in daily access options, which may result in patients receiving an inadequate duration or dose of antibiotics. Cost also plays a key role in antibiotic use and misuse. Although there are limited data on the relation between cost and antibiotic use, studies have shown an association between increased out-of-pocket expenditures and antimicrobial drug resistance in LMICs [27]. This may be because co-payments imposed by the private sector result in overuse of antibiotics. In addition, if patients or families are responsible for purchasing the antibiotics, they may be able to afford only a portion of the needed dose. All of this contributes to inappropriate use of antibiotics and an increased risk of antimicrobial drug resistance. The overuse of antibiotics is a contributor to the development of resistant pathogens and global antimicrobial resistance [28]. In Thailand, the strongest risk factor for acquiring multi-drug resistant gram-negative infection was prior antibiotic use [29].

Referral patterns may also contribute to the risk of antimicrobial drug resistance. Tiered referral patterns are used to optimize human resource management and provide care in a distributed manner [15]. Patients initially receive treatment and management at the local health center. They may then be referred to a district hospital and then to a referral hospital. In many LMICs, surgeons are concentrated in urban, referral centers. By the time a patient is seen by a surgeon, he or she likely has been treated at a health center and a district hospital. The patient also may have sought treatment and medication independently from a pharmacy. At each of these levels, it is likely the patient has received antibiotics. In pediatric patients, 25% of antibiotic purchases in LMICs were obtained without a prescription [26]. The surgeon therefore does not know which antibiotics or what doses and duration of treatment the patient has received. Antibiotic stewardship programs will need to look beyond hospital-level management to address these challenges.

Antimicrobial drug resistance is a global threat and associated with higher morbidity and mortality rates and costs [30]. The World Health Organization (WHO) has identified organisms of critical priority because of the high clinical burden, increasing rates of resistance, and availability of treatment options. Enterobacteriaceae with third-generation cephalosporin or carbapenem resistance are classified as an area of “critical priority” [31, 32]. These organisms contribute a substantial burden of disease both within the healthcare sector and within the community. In addition, there are limited antibiotic treatment options available or in the pipeline to address this growing burden [33]. This is not a problem isolated to LMICs. However, the problem is especially challenging in these regions because of the lack of alternative antibiotics.

Resistance by Enterobacteriaceae to third-generation cephalosporins and carbapenems has been noted in all WHO regions [31]. In pediatric patients, the median resistance rate of K. pneumoniae to ceftriaxone was 84% in Asia and 50% in Africa [34]. In Cambodia, 86% of pediatric patients were colonized by at least one third-generation cephalosporin-resistant organism, with 62% being colonized at the time of hospital admission and 23% during hospitalization [35]. In Tunisia, 43% of healthcare-associated IAI caused by E. coli were resistant to cefotaxime [36]. In Rwanda, 31% and 59% of E. coli and Klebsiella, respectively, were resistant to at least one of the third-generation cephalosporins [37]. In Ugandan patients with E. coli isolated from blood cultures, the resistance rates to ceftriaxone and imipenem were 67% and 19%, respectively; in Klebsiella isolated from blood cultures, resistance rates to ceftriaxone and imipenem were 85% and 20%, respectively [38].

Rates of extended-spectrum beta-lactamase (ESBL)-producing organisms are increasing globally. In IAI in the SMART study, 79% of E. coli and 70% of Klebsiella were ESBL producers [39]. In the Asia–Pacific region, the rates of ESBL production among E. coli and Klebsiella were 25% and 16%, respectively [40]. In Latin America, 30% of Klebsiella and E. coli isolates were ESBL producers [41]. In a rural hospital in Gabon, 40% of E. coli and Klebsiella isolates were ESBL-producing organisms [42]. In Vietnam, the rates of ESBL production in E. coli and Klebsiella isolates were 48% and 40%, respectively [43]. In Mexico, the rates of ESBL production in E. coli and Klebsiella were 54% and 39%, respectively [44]. In patients with peritonitis in India, 62% of E. coli and 74% of Klebsiella isolates were ESBL positive [25]. Among Enterobacteriaceae isolates in southeast Asia, 39% were positive for ESBL production, with the highest rates seen in Vietnam (55%) and Thailand (45%) [45]. In patients with SSIs at a referral hospital in Uganda, 81% of Enterobacteriaceae were ESBL producers [20].

Susceptibility to carbapenems remains high with >90% of Enterobacteriaceae sensitive in ertapenem [46]. However, there has been a decreasing trend in ertapenem activity against Enterobacteriaceae [46]. Among Enterobacteriaceae in a Latin American cohort, 6.6% were carbapenem resistant [47]. In Ethiopian children, the prevalence of carbapenemase-producing Enterobacteriaceae was 12% [48]. In Tanzania, carbapenemase genes were identified in 11% of K. pneumoniae and 8% of E. coli isolates [49]. In Latin America, the carbapenem resistance rate in Klebsiella isolates was 9% [50].

The challenge of antimicrobial resistance in surgical infections and the role of the surgeon in antibiotic stewardship have been noted [51, 52]. Continued efforts need to be focused on strengthening the surgical and healthcare workforce globally. Surgical care is cost effective, and strengthening surgical systems helps improve healthcare systems overall [15, 16]. Surgical care is essential for source control—the cornerstone of management of surgical infections. In addition to antibiotic stewardship, attention must be paid to infection control measures such as prevention of SSIs. In Ugandan patients with SSI, 79% of the gram-negative bacteria were multi-drug resistant [20]. High rates of SSI are secondary in part to lack of infection control and surveillance [2].

Guidelines being developed for the management of surgical infections need to include the particular circumstances of LMICs, addressing the most common surgical infections and providing management priorities that are locally appropriate and feasible. These measures need to factor cost into the equation, as this can be a significant factor limiting implementation of guidelines. Teams for guideline development and research priority formulations should include LMIC researchers.

Antibiotic stewardship needs to be emphasized and strengthened in LMICs [53]. Successful implementation strategies include tighter control of antibiotics, multi-disciplinary teams to assess indications for antibiotic use, and hospital-based teams for infection prevention and control. Governmental support and engagement are essential in a national antibiotic stewardship program. Corruption and conflict of interest may influence antibiotic use and misuse. Antibiotic stewardship programs often are instituted at the hospital level. However, with countrywide, tiered referral patterns, a broader antibiotic stewardship approach may be needed [54]. Procurement and supply chains should be strengthened to improve the antibiotic supply and the options available. Data on actual antibiotic use in LMICs are limited but could play an important role in understanding the development of antibiotic resistance [55]. A situation analysis in Vietnam found that with greater access to medical care, there has been an increase in injudicious use of antibiotics with further escalation in antimicrobial resistance [56]. Although policies are in place to regulate antibiotic use, tools for enforcement of such policies are lacking or insufficient [56].

Surgical infections require both source control and antibiotics. Strengthening surgical care globally is an important first step. However, with the growing rates of antimicrobial drug resistance and limited antibiotic options, attention needs to be directed toward innovative solutions for this challenge, focusing on appropriate antibiotic use and stewardship.