Bowel perforation secondary to illegally induced abortion though rare and uncommon in developed world is a significant and major cause of maternal morbidity and mortality in countries like Tanzania where abortion laws are still restrictive and most abortions are performed clandestinely and illegally by unqualified personnel
[3, 15]. The incidence of abortion-related complications such as bowel injuries has been reported in most developing countries to be increasing at an alarming rate
. The rate of bowel perforation as a complication of induced abortion has been reported in literature to range from 5% to 18% of all abortion-related complications
[12–14]. However, as in other iatrogenic surgical problems, many cases may have been unreported because of its medico-legal implications
[9, 23]. In this study, the rate of 4.2% of bowel perforations may actually be an underestimate and the magnitude of the problem may not be apparent because many cases are not reported for fear of been arrested by police. Several other cases may also have been treated in private hospitals which were not included in the present study. Exclusion of large number of patients in this study as a result of lack of enough data may have also contributed to the underestimation of the magnitude of the problem.
In keeping with other studies
[2, 9, 24, 25], majority of our patients who underwent induced abortion were young, secondary school students/leavers, unmarried, nulliparous, unemployed and most of them were dependent member of the family. This finding is contrary to what was observed by Rehman et al.
 who reported that most of the women were married and had five or more children. The majority of patients in the present study presented themselves for abortion when the pregnancy was advanced and, therefore requiring relatively more complicated termination procedure which only a specialist may handle. But because of socio-economic, cultural and law restrictive reasons most of these women fear of revealing their pregnancy and as a result fall prey to unqualified and inexperienced people who perform such illegal procedures under substandard unhygienic places.
The majority of patients in this study came from urban areas, which is in agreement with other studies done elsewhere
[3–5, 9, 11, 15–17]. Previous studies have shown that premarital sexual intercourse is practiced much in urban than in rural areas probably because of increasing urbanization that broke down cultural barriers and predisposed to increased sexuality
. This needs to be studied further so that effective intervention strategies for positive behavioral change will be mounted.
In this study, the rate of contraceptive use was as low as 14.7% which is comparable with other studies done in developing countries
[4, 24, 28–30]. Low contraceptive uptake may be due to fear about the safety of contraceptives, lack of knowledge about family planning, religious believes and lack of access to services. This calls for proper training and continuing education for awareness on abortion and its complications.
In the present study, more than 70% of patients had procured the abortion in the 2nd trimester which is consistent with other studies
[29, 30], but at variant with Enabudoso et al.
 in which women sought abortion in the first trimester. Ignorance and inability to take quick decision regarding termination of an unwanted pregnancy compel a large number of women to seek illegally induced abortion in the second trimester from unauthorized person in unrecognized places. The delay in procuring abortion in the present study may be due to the restrictive abortion laws in this country, the secrecy associated with abortion and the religious and social norms that do not accommodate abortion practice. Also, lack of financial support may have contributed to delay in procuring abortion.
Women's reasons for seeking abortion were discussed in several studies
[9, 24, 29–31]. These included inappropriate timing of the pregnancy, fear of expulsion from school, financial difficulties, and uncertainties about the partner. In this study, fear of expulsion from school was the most common reason for terminating pregnancy.
As reported by many authors
[15, 17, 30, 31], majority of patients in the present study presented late in poor general condition. This was found to be the most important factor influencing the outcome of surgical procedure as also emphasized by a number of authors
[9, 15, 30]. In resource-poor countries, difficulties in diagnosis, lack of awareness of the disease and delayed referral to tertiary hospital often result in delayed presentation to a hospital
Surgical intervention is considered to be the gold standard treatment for patients with bowel perforation following induced abortion
. In this study, all patients underwent surgical treatment which is in keeping with other studies
[9, 11, 16–20, 26, 32, 33]. One of the many factors affecting the surgical outcome in patients with bowel perforation is time interval from perforation to laparotomy
[9, 15]. Early surgery can minimize the complications while delayed surgery leads to severe peritonitis and septic shock. In the present study, the majority of patients were operated more than 24 h after the onset of illness. Similar observation was reported by other studies done in developing countries
[4, 9, 30]. Delayed definitive surgery in the present study may be attributed to late presentation due to lack of accessibility to health care facilities, lack of awareness of the disease as a result some patients with bowel perforation following induced abortion may decide to take medications in the pre-hospital period with hope that the symptoms will abate. It is also possible that some clinicians managing the patients initially may not have considered perforation as a possible diagnosis leading to delayed referral to tertiary care hospital.
In keeping with other studies
[9, 16–20], the ileum and the sigmoid colon were the most common parts of the bowel affected. The relative fixity of these portions of the bowel has been suggested as a possible reason for this.
Early surgical interference is the optimal treatment option for perforation. However, the type of surgery to be applied is controversial
. The surgical management of small intestinal injuries is fairly straightforward with minimal sequalae. Our practice in managing these patients is a simple closure in solitary perforations and segmental intestinal resection and primary anastomosis in multiple perforations or gangrenous bowel. The management of large bowel injury is more controversial
[9, 18]. This is more so when the left colon is involved. A simple colostomy has been reported to be the safest approach in the management of these injuries. Other options include primary repair, resection and primary anastomosis, and repair with a proximal protective colostomy. A simple colostomy is easier and faster to accomplish in these poor surgical risk patients. However, the major drawback of colostomy is the need for a second operation to restore intestinal continuity, the specialized care before closure and the attendant cost which reduces its popularity
[34, 35]. The challenge is even more conspicuous in a developing country like Tanzania where resources for caring of patients with colostomy are limited. The management of stoma remains difficult in developing countries because of the shortage of suitable equipment in this respect and peristomal ulceration remains a major problem
. Experiences in our centre are primary repair and resection and primary anastomosis in case of viable bowel, whereas colostomy is reserved after resection of a gangrenous large bowel.
The overall complications rate in this series was 47.1% which is higher compared to what was reported by Thapa et al.
. High complications rate was also reported by Saleem & Fikree
 in Pakistan. This difference in complication rates can be explained by differences in antibiotic coverage, meticulous preoperative care and proper resuscitation of the patients before operation, improved anesthesia and somewhat better hospital environment. As reported by Rehman et al.
, surgical site infection was the most common postoperative complication in our study. High rate of surgical site infection in the present study may be attributed to contamination of the laparotomy wound during the surgical procedure.
In this study, mortality rate was 10.3% which is higher than that reported by Bhutta et al.
. High mortality rate in this study is attributed to high gestational age at termination of pregnancy, late presentation, delayed surgical treatment and postoperative complications.
The overall median length of hospital stay was 18 days , a figure which is lower than that reported by Rehman et al.
. Our overall median length of hospital stay was significantly long in patients who developed complications postoperatively. Prolonged length of hospitalization results in consumption of large amounts of healthcare resources such as personnel, theatre space, medications, and hospital beds.
Self-discharge against medical advice is a recognized problem in our setting and this is rampant, especially amongst patients with complications of illegally induced abortions
. Similarly, poor follow up visits after discharge from hospitals remain a cause for concern. These issues are often the results of poverty, long distance from the hospitals and ignorance.