Introduction

A woman’s sexual and reproductive health rights include the ability to have a satisfying and safe sex life while retaining the liberty to prevent pregnancy if it is undesirable [1]. Yet, in Nigeria, the unmet contraceptive need for married women and sexually active unmarried women are 16.1% and 21.8%, respectively [2]. Contraception is not always available or acceptable, and even when used, there are occasional failures associated with its use, resulting in unplanned pregnancies, for which women at times seek abortions. Even when contraception is available, it is apparent that a lot of advocacy is still required: some women consider contraception to be unsafe, avoid using it, and paradoxically consider seeking an abortion for an unwanted pregnancy to be safer than preventing it with a family planning method [3,4]. Women in countries like Nigeria, which have restrictive abortion laws, end up seeking illegal abortions, which unfortunately are often unsafe, because those medically qualified to do them usually do not [5].

Despite the clandestine nature of these abortions, the prevalence in Nigeria remains rather high. In 2012, 56% of unwanted pregnancies ended up in abortions; estimated to be about 1,250,000 abortions [6]. Figures range widely, depending on the population studied: 3.25% of university undergraduates, [7]; 33.5% of women surveyed from the community [8] and 62% of women surveyed at a health care facility [9] had previously had an induced abortion. Seventy-eight per cent of women who were attending an antenatal clinic gave a history of a previous induced abortion [10]. Reported complications rates are expectedly high: post-abortion complications requiring post-abortion care (PAC) occur in 41.4%–72.7% of abortions [9,11,12], yet many do not access this PAC [12]. Two local medical facility-based studies of maternal deaths found that 11.5% and 30.8%, respectively could be attributed to abortion [11,13].

Materials and methods

This was a cross-sectional study carried out at nine selected secondary and tertiary hospitals which offer post-abortion care in South-West Nigeria. Participating hospitals were Lagos State University Teaching Hospital (LREC/10/06/228); Lagos Island Maternity Hospital (LREC/10/06/228); Sacred Heart Catholic Hospital, Abeokuta (UI/UCH/11/0258); Adeoyo Maternity Teaching Hospital, Ibadan (AD13/479/257); Ring Road State Hospital, Ibadan (AD13/479/257); State Specialist Hospital, Akure (UI/UCH/11/0258); Obafemi Awolowo University Teaching Hospitals Complex, Ile-Ife (ERC/2012/12/01); Obafemi Awolowo University Teaching Hospitals Complex, Ilesa (ERC/2012/12/01) and University College Hospital, Ibadan (UI/UCH/11/0258). Research approval was obtained from the Institutional Research Boards covering all the participating centers (approval reference numbers inserted in parentheses) and from the WHO Ethics Review Committee.

Hospital medical records were the source from which, over a nine-year period from 2006 to 2014, data were retrieved on clinical presentations, methods used in induced abortion, treatment and outcome of management of patients with abortion at each facility, using a 58-item instrument. These records were identified from all possible points of care: emergency room, gynecology clinic, ward and theatre registers, to reduce potential bias on the subjects recruited. There was no pre-specified sample size; all eligible women with abortion complications were included. The independent variable was the method of abortion used. Pelvic instrumentation performed by manual vacuum aspiration or dilatation and curettage was termed as ‘surgical abortions’, as differentiation between them (from the patients’ account) was anticipated to be difficult. Participants mainly identified misoprostol as ‘Cytotec’ or by its polygonal tablet form. Dependent variables were complications of abortions, the need for additional treatment and maternal mortality. Data were analyzed with IBM SPSS Statistics 20 (IBM, Armonk, NY, USA). Statistical significance was set at p<0.05 and tested with chi-square, Fisher’s exact and Χ² for the trend tests.

It was anticipated that some variables might be missing from the data sources, on account of the retrospective design. However, the retrospection was necessary to assess the trends in the previous years, as misoprostol became more available to abortion seekers.

Results

There were 2,463 cases of abortions during the period of review. Information on the history of interference with the pregnancy was not available in 10 cases, leaving 2,453 cases for consideration. Of these, 1,754 (71.5%) were spontaneous abortions; final analyses were therefore based on the remaining 699 (28.5%) women with induced abortions.

Table 1 shows the methods used in procuring the abortions. Most of the women (484; 69%) had surgical abortions, followed by misoprostol use. Some of the other methods used included: concoctions made up of lime, gin, aspirin and other non-steroidal anti-inflammatory drugs; oxytocics and other injections, oral or vaginal herbal preparations, blunt abdominal trauma, and mixtures of battery water (which may contain some sulphuric acid), detergents or household bleach. No one had used mifepristone.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Methods used to procure abortion.

https://doi.org/10.1371/journal.pone.0209415.t001

Table 2 shows that over the study period, the proportion of women using misoprostol for induced abortions rose relative to the other methods (Χ² trend: 30.96, P <0.001). This observed increase in misoprostol abortion, during the study period was a linear increase (Χ²_GOF 10.19; p = 0.178).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Trends in methods of induced abortions.

https://doi.org/10.1371/journal.pone.0209415.t002

Compared to patients who used misoprostol, patients who procured abortion with other methods (and specifically, surgical abortion) had disproportionately more injuries to the genital tract, were more likely to have offensive vaginal discharge, pelvic or intra-abdominal abscesses and peritonitis and were more likely to present with high-grade fever and septicemia, or develop severe medical complications. Between these two groups of patients, there was no significant difference in the occurrence of retained products of conception, complications like severe anemia or hypotension, or mortality (Table 3).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Comparison of clinical features and complications of abortions with misoprostol use to other methods, and specifically to surgical abortions.

https://doi.org/10.1371/journal.pone.0209415.t003

Table 4 summarizes hospital interventions following the different methods used to procure an abortion. Compared to patients who used misoprostol to procure abortion, patients who had a surgical abortion were more likely to be admitted as in-patients, have prolonged course of antibiotics, were more likely to have a laparotomy, drainage of abscesses, and were more likely to require additional specialist care.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Hospital interventions required following misoprostol use versus other methods, and specifically surgical abortions.

https://doi.org/10.1371/journal.pone.0209415.t004

Discussion

Almost three out of ten women managed for abortion at nine of the hospitals in south-west Nigeria had induced abortions. It is likely that this proportion is actually higher: some women are not likely to admit to interference, as abortions are restricted and they could be prosecuted, or at best, stigmatized [17]. Many of the local methods volunteered in this survey have long been used to induce abortions, with variable levels of success. Misoprostol is relatively new on the list, but as the trend shows, it is becoming increasingly popular.

It is not expected that presentations such as reproductive tract injuries would be associated with a medical abortion method such as misoprostol use, as found in this study. Other presentations like fever (and complications like infectious morbidity) were significantly less associated with misoprostol than with other methods and specifically, surgical abortions. It is reported that misoprostol abortions are not always complete [18–21]. This theoretically makes infectious morbidity a likely complication; however, the study did not corroborate this. Incomplete abortion was a common complication in this study, irrespective of the method used. These findings are in favor of use of misoprostol over surgical abortion, in agreement with other studies [21].

Studies comparing misoprostol to manual vacuum aspiration (MVA) for incomplete abortions have found that patients had less pain (though they bled somewhat more) and were more likely to choose misoprostol again in the same situation, and recommend it to a friend [19,20,22]. Completeness of abortion with misoprostol, however, is reported to be 92.2% to 98.8% compared to 100% with MVA [18–21]: this is imperfect, but acceptably high nonetheless. Singh [23] concluded from hospital studies that misoprostol reduces the severity in abortion complications. The finding of a relatively large percentage of women presenting with incomplete abortions to the selected hospitals in S.W. Nigeria may be a result of a number of factors. Firstly, the study is hospital-based so does not reflect the actual proportion of women who used misoprostol who subsequently presented for an incomplete abortion. Without a well-designed community-based study, it is not possible to estimate the use effectiveness of medical abortion in Nigeria, where abortion generally occurs outside a hospital setting. However, because women who wish to induce abortion, and who employ misoprostol, may be unsure of the outcomes of the use of the drug, this may result in them presenting at a hospital to ensure that the abortion is complete. This is also likely to occur because of the increased bleeding associated with misoprostol. Better methods of communicating the effects of the use of misoprostol use may result in decreases in the number of misoprostol users who subsequently seek surgical procedures to ensure that an abortion has been completed.

The WHO guidelines recommend that both surgical and medical abortions are safe and effective [24]. Abortions are safe “if they are done with a method recommended by WHO that is appropriate to the pregnancy duration and if the person providing or supporting the abortion is trained” [25]. However, as surgical abortions are usually procured illegally in Nigeria, they are often unsafe. An unsafe abortion occurs “when a pregnancy is terminated either by persons lacking the necessary skills or in an environment that does not conform to minimal medical standards, or both” [5]. The post-abortion care required for the studied subjects was significantly more elaborate when surgical abortions were procured, with associated morbidity and greater cost of care. It is also important to note that surgical complications are more common with dilatation and curettage (D & C), than the less traumatic MVA. D & C is more likely to be used in an illegal abortion. The study was unable to distinguish between the two methods, as patients were not able to report the types of instruments used for their procedures.

Most of the local studies in literature evaluated misoprostol’s effectiveness in treating incomplete abortions [19–21]. No studies assessing the effectiveness or safety of inducing abortions with misoprostol were found; Okonofua et al. [26] prospectively evaluated legally induced abortions with a combination of mifepristone and misoprostol. This makes the index study of some value, despite its limitations of being a retrospective, records-based study.

Other limitations of the study include missing data (as data sources were retrospectively-accessed medical records), and the possibility that some instances of misoprostol use may have been missed, as it depended on the woman’s ability to name or describe it. Also, the study assesses women who require health services following induced abortion. As previously discussed, these participants’ proportion among all women who procured abortions is not known, therefore, the comparisons made between misoprostol and surgical methods are limited. One cannot determine how many of either group did not need to seek post-abortion care.

It can be concluded that, in this study, severe complications, including infectious morbidity, are more likely to occur with surgical abortions than with misoprostol, and that misoprostol use has increased in recent years. It may be cautiously implied that misoprostol may be a safer method for abortion induction than traditional methods of abortion and surgical evacuation obtained in an illegal context. Its use for legally indicated abortion should be encouraged.

Supporting information

Acknowledgments

The authors wish to thank the research assistants and data managers who worked on this project.