Typhoid Ileal Perforation: A Review of a Rural District Hospital Experience


Author: Ojuka KD Affiliation: Department of Surgery, University of Nairobi P.O. Box 30197 -00100 Nairobi . Correspondence: Dr. Daniel Ojuka, Email address



Background: Enteric fever is endemic in developing countries and frequently complicates with ileal perforation. Surgical intervention for the perforation is the usual treatment but attendant rate of post-operative complications high. It is unclear what the spec- trum of enteric fever perforations is in rural hospital practice in Kenya, where the diagnosis most often in intraoperative.


Objective: To describe the surgical experience of typhoid perforations at a rural district hospital in Kenya.


Patient and Methods: This was a retrospective chart review of patients who underwent laparotomies for peritonitis at the Kapenguria District Hospital in Kenya between April 2007 and October 2009. Data abstracted from patient files included demographics, presenting symp-tom, duration of symptoms, investigations (Widal and/or stool culture), operative management, complications, length of stay, and death. Widal test was indicative when titer was 1:160 for “O” antigen or above. Anti - mesenteric longitudinal perforation was assumed to be a complication    of enteric fever. The data were analyzed using SPSS version 16.0.The results are presented in frequency tables, bar charts and pie charts.

Results: Of the 50 files retrieved with diagnosis of peritonitis, 21(42%) were found at operation to have had ileal perforations. Of these 15(71.5%) had resection and primary anastomosis, 2(9.5%) had refreshening and anastomosis (simple anastomosis) and 4 (19%) had ileostomy. Male to female ratio was 4:1, majority were aged 6-15 years (38.1%). Wound infection was 8(38.1%), enterocutaneous fistulae were 7(33. 3%), while 7(33.3%) required second laparotomy and 4(19%) were referred due to complications which could not be managed at this level. Mortality was 3(14.3%) and average length of stay was 17days.


Conclusion: Morbidity and mortality arising from typhoid ileal perfora-tion is high in this environment making it a major challenge in a resource poor environment. Prevention by use of protocols is highly recommended.



Typhoid fever is endemic in most developing countries and perforation is the most common surgical complica-tion seen in both children and adults (1, 2). It is usually a prolonged disease with bacteremia phase characterized by fever and chills during week one, reticuloendothelial involvement with rash , abdominal pain and prostration in week two, ulceration (which are longitudinal and on the antimesenteric border situated within 45cm of the ileocaecal valve) of Peyer’s patches with bleeding and perforation in the third week (1,3,4). Perforation could be the first sign of the disease (5). The incidence of perforation varies considerably, 15-33% in West Af-rica 12.96% in Nepal and 1-3% in Egypt and Iran (6, 7). Male sex, leucopenia, inadequate antimicrobial therapy and short duration of symptoms are some of the pre-dictors of enteric perforation (4). Surgical intervention is the main method of treatment with good results (8). Primary closure, excision and closure, resection and anastomosis, Ileostomy, ileotransverse anastomosis and ileotransverse anastomosis plus hemicolectomy are

proposed operative techniques treatments. The postop-erative complications are mainly wound sepsis, residual intra-abdominal abscess, wound dehiscence, fecal fistula and death. These have continued to be high in areas where typhoid fever is endemic (2, 8-13).


There is no study done to demonstrate the practice and the attendant complications in rural Kenya. This study sought to audit the experience in management of ty-phoid perforation in a rural district hospital in Kenya.


Patients and Methods

This was a retrospective chart review of surgical cases treated between April 2007 and October 2009. The data collected from the file included diagnosis at admission, tests done pre-operatively or post-operatively to confirm the diagnosis, presenting symptoms, duration of symp-toms, operative technique used, complications, need for second operation, need for referral, mortality and length of stay in the hospital. The Widal test indicated infection when the titer for antigen ‘O’ was 1:160 or above. The surgical finding of antimesenteric longitudinal perfora-


Click to view table 1




Click to view figure 1



Click to view figure 2


tion was used as confirming typhoid fever as the cause of the perforation (1, 2, 8). Data were analyzed using SPSS version 16.0. Results were presented on tables, bar charts and pie charts. Chi- square test was used to calculate the differences in proportions.





Fifty files of patients operated for laparotomy were re-trieved. Out of these, 21 had terminal ileal perforations. Of the twenty one patients, 17 were male, (male to fe-male ratio of 4:1), with age range of 6-61years and a mean of 21.29 years. Eighteen had admission diagnosis of general peritonitis with no known cause while three had preoperative diagnosis of typhoid fever and were admitted in medical wards before being diagnosed with peritonitis.


Only eight (38.1%) had Widal test performed, three of which were preoperative while five were post-operative. All tests were indicative of typhoid. In one patient, culture of excised tissue was performed and was indicative of salmonella infection. Twenty patients (95.2%) presented with fever (Table 1) and 14 (66.7%) presented with history of less than one week (Table 2).The number of perforations ranged from two to five with an average of The most often performed technique was resection and anastomosis (71.4%) (Figure 1). The complica-tions encountered were wound dehiscence (19.05%), enterocutaneous fistula (28.57%) and wound infection (33.3%) (Figure 2). Ten patients (47.6%) required reop-eration, while four (19%) were referred due to recurrent en- terocutaneous fistula. Three patients (14.2%) died while the average length of stay was 17 days.



Spontaneous ileal perforation is an important condition in condition in developing countries. Common cases involve typhoid fever and tuberculosis (13). While tu-berculosis perforations are transverse and in mesenteric border, typhoid ones are longitudinal and in antimesen teric border. The male preponderance seen in this study confirms what has been reported in West Africa (6, 8,14), Iran (4,8), Turkey (4, 6, 8) and the Indian subcontinent (15). The young age of presentation is also c o n s i s t e n t w i t h p r e v i o u s r e p o r t s (16).


The short  duration of symptoms in the study suggests a more virulent strain as reports indicate that most per-forations occur in the 2nd to 3rd week (17). Like in other reports (5), this mirrors our situation where 18 out of 21(85.7%) were diagnosed intraoperatively. The diagnosis was presumed due to the morphology of the perforation along the antim