FormalPara Key Summary Points

Why carry out the study?

This retrospective study aimed to determine the incidence of fever following misoprostol administration at a tertiary care hospital in Saudi Arabia and assess unnecessary antibiotic prescription in this context.

The incidence of fever post-misoprostol administration was found to be 8%.

What was learned from the study?

Patients who developed fever had a higher gestational age and received higher doses of misoprostol, but no significant associations were found between other patient variables and fever incidence.

Antibiotic therapy was administered to almost half of the patients who developed fever, but it was determined to be unnecessary in all cases.

The study highlights the need for further research to explore strategies for reducing unnecessary antibiotic use in the context of misoprostol-induced fever in medical abortion.

Introduction

Medication abortion exhibits a remarkably low incidence of infection primarily due to its non-invasive nature, eliminating the need for uterine instrumentation.

Misoprostol, an indispensable drug for abortion induction and postpartum hemorrhage management, has been associated with pyrexia. However, misoprostol-induced fever is mechanistic in nature and independent of infection-induced fever. The pharmacological action of misoprostol as a prostaglandin E1 analogue, acting upon the central thermoregulatory centers, elucidates its adverse effect of fever [1].

The reported incidence of misoprostol-induced fever has demonstrated variability across studies and countries, with a range of 20–40% observed in two large multinational trials [2, 3]. Furthermore, several cohort studies have reported even higher rates [4, 5].

A study conducted in Ecuador reported that 35% of patients developed fever following sublingual administration of misoprostol for postpartum hemorrhage. Notably, peak temperatures were observed between 1 h and 2 h post-treatment and gradually subsided over a span of 3 h. No significant risk factors associated with fever development were identified within the patient cohort [4]. Similarly, in a prospective cohort study conducted in Argentina, a substantial 75.5% of patients experienced fever subsequent to misoprostol usage for postpartum hemorrhage [5].

Misoprostol-induced fever has been reported to be dose dependent and route dependent [6]. In a meta-analysis of 33 randomized controlled trials (RCTs), the incidence of misoprostol-induced fever for postpartum hemorrhage was found to be 15%, 11.4%, and 4% with sublingual, oral, and rectal routes, respectively [7].

Additionally, a genetic predisposition has been described, where genetic variability in ABCC4 has been associated with an increased risk of this adverse effect [8].

It is worth noting that mifepristone is not available in Saudi Arabia, despite the preferred protocol for a medical abortion being a combination of mifepristone and misoprostol. This combination has been correlated with a high success rate, eliminating the need for surgical intervention [9]. Therefore, unavailability of mifepristone may have implications on the required doses of misoprostol, and subsequently, the incidence of misoprostol-induced fever.

Misoprostol-induced fever, although not a serious adverse effect, can pose challenges in management due to its association with possible intrauterine infection. The occurrence of fever, typically considered a sign of infection, may lead to inappropriate initiation of antibiotics, potentially contributing to the ongoing antimicrobial resistance epidemic. While previous studies have primarily examined the onset of fever in the postpartum setting [2,3,4,5], the incidence of misoprostol-induced fever specifically in the context of medical abortion has not been studied.

The objectives of our study were as follows. Firstly, to determine the incidence of fever post-misoprostol administration at a tertiary care hospital in Saudi Arabia. Secondly, to investigate the potential risk factors associated with the development of post-misoprostol fever within this particular clinical context. Lastly, to assess the extent of unnecessary antimicrobial utilization in relation to the occurrence of misoprostol-induced fever.

Methods

This retrospective cohort study was conducted at the Ministry of National Guard Healthcare Affairs (MNGHA) in Riyadh, Saudi Arabia. The study population consisted of all female patients who received misoprostol for first- or second-trimester terminations of pregnancy or management of missed or incomplete abortion. The study included patients treated between January 2017 and December 2019.

We included female patients undergoing first- or second-trimester termination of pregnancy or receiving treatment for missed or incomplete abortion using the MNGHA protocol with misoprostol; our protocol is in alignment with the International Federation of Gynecology and Obstetrics (FIGO) misoprostol-only recommended regimens 2017 chart, which varies by type of abortion and gestational age [10]. Patients with a history of fever or antibiotic therapy before initiating misoprostol, as well as those suspected to have an infection prior to induction, were excluded from the study.

The legality of terminations of pregnancy in Saudi Arabia is regulated by specific circumstances outlined in early 2011, which include [11]:

  1. 1.

    Abortion of a malformed fetus after 120 days of conception or 19 weeks of gestation following ensoulment is permissible if the pregnancy is certain to cause the death of the mother.

  2. 2.

    Abortion of a malformed fetus before 120 days or prior to ensoulment is permissible if it is certain that the fetus will die following birth or if it has severe incurable disabilities.

  3. 3.

    Abortion of the fetus at any stage of pregnancy is permissible if intrauterine death was confirmed.

Data collection for all patients receiving misoprostol (with and without fever) included the following:

Demographics: age, gestational age, comorbidities, indication for misoprostol, and reason for abortion.

Fever: number of patients with fever, time-to-onset of fever, and duration of fever. Fever was defined as an a.m. temperature exceeding 37.2 °C (> 98.9 °F) or a p.m. temperature exceeding 37.7 °C (> 99.9 °F). Fever is documented in patient’s electronic medical records on a vital signs sheet.

Misoprostol: total dose of misoprostol and route of administration.

Outcome of abortion: induction-to-expulsion time, successful abortion without surgical intervention, and requirement of epidural analgesia.

Indicators of shock: utilizing the systemic inflammatory response syndrome (SIRS) criteria, which is defined as having two of the following four:

  • Body temperature over 38 °C or under 36 °C.

  • Heart rate greater than 90 beats/min.

  • Respiratory rate greater than 20 breaths/min or partial pressure of CO2 less than 32 mmHg.

  • Leukocyte count greater than 12,000 or less than 4000/microliters or over 10% immature forms or bands.

Indicators of intrauterine infection: fever or chills persisting for more than 24 h after misoprostol administration, and increasing abdominal pain.

Infection parameters: C-reactive protein (CRP) and leukocyte count. The normal range for CRP was defined as < 10 mg/L, and the normal range for leukocytes in pregnancy as 5.6–13.8 × 109/L.

Administration of prophylactic antibiotics before the first misoprostol dose: type of antibiotic, initiation time, and duration of use.

Administration of antibiotics within 48 h after the last misoprostol: type of antibiotic, initiation time, and duration of therapy.

Blood, vaginal, urine, or other relevant cultures, if available.

Assessment of unnecessary antibiotic prescription: an independent review was conducted by a clinical pharmacist and an infectious disease physician for each patient who developed fever and received antibiotic therapy. Antibiotic therapy was deemed unnecessary in cases of non-infectious origin or failure to identify any source of infection. This included cases where cultures returned negative or no cultures were taken, where CRP and leukocytes levels were normal, and where fever or chills did not persist more than 24 h after the last administered misoprostol dose, and there was no increase in abdominal pain [12]. In case of any disagreements, thorough discussions were held until a consensus was reached.

Statistical Analysis

Statistical analysis was conducted using appropriate tests for the variables under investigation. Categorical variables were analyzed using the chi-squared test or Fisher’s exact test. Continuous variables were assessed using the Mann–Whitney U-test.

Gestational age was categorized into specific ranges (< 98 days, 98–112 days, 112–126 days, 126–140 days, 140–154 days, 154–168 days, and > 168 days), and the total administered doses of misoprostol were categorized as follows: < 600 mcg, 600–1199 mcg, 1200–1799 mcg, 1800–2399 mcg, 2400–2999 mcg, and ≥ 3000 mcg. To compare the differences in the mean highest temperature across these categorize variables, the Kruskal–Wallis test was employed.

To control for confounding factors such as comorbidities and epidural analgesia, a multiple logistic regression analysis was preformed. The presence of fever served as the dependent variable, while comorbidities, epidural analgesia, and the required dose of misoprostol were considered as determinants. The association between fever and the dosage of misoprostol was assessed using Pearson correlation.

The results are presented as odds ratios (OR) with corresponding 95% confidence intervals (CI). Statistical significance was set at p < 0.05. The analysis was conducted using SPSS software.

Additionally, differences between patients who received necessary or unnecessary antimicrobials were evaluated using the chi-squared test for nominal data and the Wilcoxon signed-rank test for continuous data. Statistical analysis was conducted using SAS 9.4 (SAS Institute Inc., Cary, NC, USA).

Compliance with Ethics Guidelines

The institutional review board (IRB) at King Abdullah International Medical Research Center (KAIMRC) (Protocol RC20/657/R) approved this study. No consent form was required by the Ministry of National Guard Health Affairs ethics committee. This study was performed in accordance with the 1964 Declaration of Helsinki and its later amendments.

Results

A total of 213 patients were included in the study, with a mean age of 33.9 ± 7.75 years. The median gestational age was 12.0 (interquartile range: 10.00, 15.00) weeks, where only two patients were in their third trimester. Among the patients, 24.9% (n = 53) had comorbidities. Of these, 4.2% had diabetes, 2.8% had hypertension, and 0.9% had cancer. None of the patients had renal or hepatic impairment. See Table 1.

Table 1 Patient characteristics, gestational age, and comorbidities
Full size table

The indications for receiving misoprostol were as follows: 51.6% of the cases were for missed miscarriage, 24.9% for incomplete miscarriage, 22.1% for termination of pregnancy, and 1.4% for incomplete abortion.

Regarding the setting of misoprostol administration, 46.9% of patients received the course in the emergency unit, 33.8% were admitted as inpatients, and 19.3% received it in an outpatient visit. Among the patients, 16% were nulliparous. Overall, abortion was unsuccessful in 19.7% of patients. Induction-to-expulsion time was recorded for only 13 patients at a median 15.0 (IQR 10.5–44.5) h.

Misoprostol Dosing and Route

The mean number of doses administered was 2.9 ± 2.68 (range 1–14), with a median of 2.0 (IQR 1.00–4.00) doses. The median dose was 1200 (IQR 600–1800) mcg. The most common route of administration for misoprostol was oral. However, approximately 32% of the patients received misoprostol via a multiple routes. See Table 2 for misoprostol dose and routes of administration.

Table 2 Misoprostol dose and routes of administration
Full size table

Incidence of Fever and Characteristics of Patients that Developed Fever

Among the patients included in the study, 8% (n = 17) developed fever within 24 h of receiving misoprostol. The mean age of these patients was 32.0 ± 6.00 (range 22–40) years, and the mean gestational age was 14.9 ± 4.18 (range 8–23) weeks. Among them, 76% were admitted as inpatients, and 29.4% were nulliparous. None of these patients had comorbidities. Among the 17 patients, 47% (n = 8) received misoprostol for termination of pregnancy, 47% (n = 8) for missed miscarriage, and 5.8% (n = 1) for incomplete miscarriage. The onset of fever from the time of misoprostol was between 1 h and 2 h in approximately half of the patients. Fever lasted for a mean duration of 0.9 ± 0.31 h (range 0.5–1.5), and the mean recorded temperature in these patients was 38.0 ± 0.59. The route and doses of misoprostol in these patients is in Table 3.

Table 3 Misoprostol dose, route and time from administration to the occurrence of fever in patients that developed fever
Full size table

Risk Factors

No significant associations were observed between any of the patient variables and the incidence of fever following misoprostol administration. See Table 4. However, it was found that there is a statistically significant correlation between the occurrence of fever and higher doses of misoprostol (p < 0.0032), as well as a higher gestational age (p < 0.0202). See Figs. 1 and 2.

Table 4 Association between any of the patient variables and the occurrence of fever post-misoprostol administration
Full size table
Fig. 1
figure 1

Association between misoprostol dose and fever. Total misoprostol dose administered in micrograms in patients who developed fever versus patients who did not

Full size image
Fig. 2
figure 2

Association between gestational age and fever. Association of highest recorded body temperature and gestational week

Full size image

Infection and Antibiotic Use

Out of the total number of patients, 17.8% (n = 38) had urine cultures ordered, 10.3% (n = 22) had vaginal cultures ordered, and 2.8% (n = 6) had blood cultures ordered. Among these cultures, only 5% (n = 11) yielded positive results.

Among the patients who did not develop a fever following misoprostol administration, 24.5% (n = 48) fulfilled the SIRS criteria. However, none of these patients met the criteria for a uterine infection on the basis of our definition. Antibiotics therapy was administered to 28.5% (n = 56) of these patients.

Furthermore, among the 17 patients who developed fever post-misoprostol administration, 29.41% (n = 5) met the SIRS criteria. However, signs consistent with intrauterine infection were observed in only one patient (n = 1). Out of these patients, 52.9% (n = 9) received antibiotic therapy. However, it was determined that antibiotic therapy was unnecessary in all of these patients.

Discussion

This is the first study to assess the incidence of misoprostol-induced fever within a large healthcare system in Saudi Arabia, specifically in patients receiving misoprostol for medical abortion. The motivation for conducting this study stemmed from an adverse drug reaction (ADR) reported to our institution’s well-established ADR reporting program [13].

The ADR report involved a renal transplant patient who underwent medical abortion with misoprostol due to her admission for acute kidney transplant rejection. Following the administration of misoprostol, the patient developed fever, prompting the medical team to initiate a course of unnecessary antibiotics. This incident raised concerns regarding the unnecessary exposure of the patient to antibiotic therapy and its potential adverse effects.

To address these concerns and gain a better understanding of the incidence of misoprostol-induced fever, this study was undertaken. Our findings revealed a lower incidence of misoprostol-induced fever compared with what has been reported in other studies. Notably, the Ecuadorian study reported the highest rate of misoprostol-induced fever, reaching 35% with dose of 800 µg [4]. Despite using higher doses of misoprostol in our cohort, the incidence of fever remained lower than what has been reported in clinical trials [2, 3].

However, we observed a concerning lower success rate of medical abortion in our study compared with reported success rates when misoprostol is used in combination with mifepristone. Our study’s overall success rate was approximately 80%, whereas success rates as high as 95% have been reported with the combination of misoprostol and mifepristone [14,15,16].

Furthermore, despite the low incidence of misoprostol-induced fever in our cohort, we observed an alarmingly high overall use of antibiotics, reaching nearly 30%, irrespective of the presence of fever. It is noteworthy that the reported incidence of infection with medication abortion has been reported to be as low as 0.9%. This incidence is lower than that observed after either surgical abortion procedures or childbirth. Despite regional variance in medication abortion regimens in the review reporting this low infection rate, which mainly included a combination of misoprostol and mifepristone, we believe a similar incidence of infections can be expected with any medication regimen that does not include uterine instrumentation [17].

Additionally, our study revealed a higher rate of antibiotic utilization among patients who developed misoprostol-induced fever. This finding further confirms our concerns that this averse effect can complicate the management of abortion and contribute to unnecessary antibiotic administration. Such unwarranted antibiotic use exposes patients to adverse effects and increases the risk of antimicrobial resistance.

In one of the largest systematic reviews, which included data from more than 46,000 patients undergoing medication abortion, regional differences in antibiotic prescription were observed. These variations were likely attributed to differences in medical practices and physician behavior [17]. It is worth noting that the American Society of Obstetricians and Gynecologists (ASOC) does not endorse the routine use of prophylactic antibiotics for medication abortion [16].

Considering this, it is essential to carefully monitor and audit the inclusion of routine use of antimicrobials in medication abortion setting as part of antimicrobial stewardship programs. While treating presumptive infections may align with standard of care practices, it is crucial to acknowledge that this approach can sometimes lead to inappropriate prescription. This situation may arise due to limitations of standardized diagnostic tools and reliance on vague symptoms for making decisions.

To strengthen antimicrobial stewardship programs, it is prudent to conduct research and identify various clinical circumstances that contribute to inappropriate prescription [18]. By gaining regional and institutional insights into these factors, healthcare providers can better support and implement effective stewardship strategies.

One of the limitations of our study, in addition to being retrospective, is that approximately 19% of the patients received misoprostol as an outpatient, therefore, it is possible that any fevers that developed at home might not have been recorded. This could potentially lead to an underestimation of the incidence of misoprostol-induced fever in our study. However, it is worth noting that excluding the patients who received misoprostol as outpatients did not significantly alter the overall incidence, which remains low.

Another limitation is the challenge in accurately clinically diagnosing infection, particularly with the low rates of cultures ordered.

While the occurrence of infectious complications following medical abortion are rare, it is vital that patients are counseled about the signs and symptoms of post-abortion infections such as prolonged bleeding, foul vaginal discharge, abdominal pain, and fever.

Conclusion

This study contributes to our understanding of the lower incidence of misoprostol-induced fever in Saudi Arabia among patients undergoing medical abortion, as compared with previous studies. However, despite the low incidence observed, we identified a high rate of unnecessary antibiotic utilization in this setting. This highlights the need for educational efforts to increase awareness among healthcare providers regarding the thermoregulatory effect of misoprostol and its ability to induce fever that may be mistakenly attributed to an infectious cause. Antimicrobial stewardship programs play a crucial role in promoting this awareness and can aid in reducing unnecessary antibiotic use in such cases.