FormalPara Key Summary Points

Why carry out this study?

Pain is largely reported by women during medical termination of pregnancy (MToP) in first trimester. Pain severity experienced by women appeared to be dose dependent for both antiprogestin and prostaglandin analogues, and an appropriate pain management is highly recommended by health organizations worldwide.

Pain management for MToP is heterogeneous, and analgesic strategy can vary between centres and should be adjusted according to risk factors associated with pain severity. Current data regarding this issue can be inconsistent, notably regarding the place of analgesic prophylaxis, factors associated with pain onset, and curative analgesic use.

The objective of this non-interventional prospective, longitudinal study was to assess in real-life settings the pain associated with MToP and the modalities of its management in France.

What was learned from the study?

Most of the women undergoing MToP were prescribed with an analgesic prophylaxis, and its intake was associated with a lower risk of curative analgesic intake. However, it was not related to a lower risk of more pain unlike history of at least one child.

Higher doses of prostaglandin analogue (i.e. misoprostol) were a significant risk for both pain and intake of curative treatment.

Introduction

The number of induced terminations of pregnancy (ToPs) in France was about 223,300 in 2021, with an increasing proportion of medical termination of pregnancy (MToP) representing 76% of all ToPs [1]. As per French Health Authority (HAS) guidelines available during the course of the study [2], MToPs procedures ≤ 7 weeks of amenorrhea (WA) included the administration of mifepristone 600 mg per os, followed 36–48 h later by the administration of misoprostol 400 µg orally or gemeprost 1 mg intra-vaginally. Alternatively, a combination of mifepristone 200 mg and gemeprost 1 mg could be used [3]. For MToPs between 7 and 9 WA, it was recommended to use 200 or 600 mg mifepristone followed by gemeprost 1 mg intra-vaginally [3]. Since March 2021, the French Health Authority recommended the use of misoprostol for women between 7 and 9 WA as an alternative to gemeprost, extending the armamentarium for healthcare providers [4]. Along this update, a compassionate use of misoprostol was instated recently in France to guarantee an appropriate women care [5]. Beyond 9 WA, surgical procedures are indicated.

Effectiveness of MToP has been largely reported. Nevertheless, women frequently report pain and the fear they may experience as important concerns regarding medical abortion [6]. Abdominopelvic pain during MToP is due to uterine contractions, which are an integral part of the MToP process [7]. Pain occurrence is thus far from marginal. Previous trials reported pain such as cramping in up to 96% of women receiving misoprostol [8]. Different rates of severe pain occurrence have been reported from 20% [9] to 80% [10]. Other have reported that 75% of women undergoing MToP during the first trimester use narcotic analgesics (mainly paracetamol and codeine) for managing pain [11].

Pain management for MToP remains heterogeneous, and narcotic analgesic use can vary greatly among centres. Ranges from 13% to 96% have been reported, suggesting different policies in offering analgesia [11]. Guidelines from the World Health Organization (WHO) recommend use of non-steroidal anti-inflammatory drugs (NSAIDs) during medical abortion for pain management. However, these recommendations were drawn from limited available evidence [12]. Clear and specific guidelines on pain management in first-trimester medical abortion were needed [13].

In this context, contemporary data on pain management in MToP remain inconsistent, notably regarding the place of analgesics prophylaxis and the pain level experienced by women, the factors related to pain onset and the use of curative analgesic treatment [14, 15]. Recently, we reported a high rate of success for MToP using mifepristone 600 mg, independent of the pregnancy term and the therapeutic protocol used [16]. We present here the pre-specified evaluation of pain resulting from the MToP procedure using a fixed dose of antiprogestin and the modality of pain management in real-life settings in France.

Methods

The study design has been previously described [16] and is summarized hereafter.

Study Design and Objectives

This was a non-interventional prospective, multicentre, longitudinal study performed between November 2015 and May 2016 in 23 active sites in France. The present report focuses on the secondary objective related to the evaluation of pain by the physician assessment and the description of analgesic prescription and use.

Study Population and Procedures

The study involved 893 women (≥18 years old) with confirmed intra-uterine pregnancy requesting MToP within the limit of 14 WA, for whom a drug regimen including mifepristone 600 mg (i.e. three tablets of 200 mg each; Mifegyne, Nordic Pharma) in a single oral dose followed by the administration of a prostaglandin analogue (PG) was prescribed. Women with known hypersensitivity to mifepristone or PG (misoprostol or gemeprost) or severe and progressive disease were not eligible.

At the inclusion visit, women were given mifepristone 600 mg and received PG to be either administered at the centre or self-administered at home according to the local procedure. Analgesic prophylaxis prescription (APP) requiring systematic analgesic treatment prior to any pain was at the investigator’s discretion. All women were given a curative analgesic prescription (CAP) to be used at their discretion. Data including demographic data, obstetrical history including history of previous ToPs, confirmation of pregnancy and MToP therapeutic protocol were collected during this visit [16].

The second visit was scheduled between 2 and 3 weeks after inclusion to assess MToP success (complete expulsion) as recommended in guidelines in place during the inclusion period [2]. Complementary data, such as intra-uterine surgical procedures, additional misoprostol intake and tolerance, were recorded during this visit [16].

The highest pain experienced by women prior to any curative analgesic intake (CAI) was assessed by the physicians using a five-point Likert scale (absence, mild, moderated, severe, extreme) during the follow-up visit. Analgesic prophylactic intake (API) and/or CAI including adjuvant (non-analgesic drug reducing discomfort, mainly gastrointestinal action) were recorded on the basis of women reporting.

Ethics

This study was performed in accordance with the Declaration of Helsinki. Informed consent to participate was obtained from women after having been informed orally by the physician and having been given the information sheet that included the electronic processing of their data. The study has been approved by the French Data Protection Authority (CNIL), the French Advisory Committee on Information Processing in Research in the Field of Health (CCTIRS, 10 September 2015) and the French Ethics Committee Comité de Protection des Personnes Sud Méditerranée I (RO – 2015/23).

Statistics

Results are expressed as mean ± standard deviation (m ± SD) or number with percentage (%) calculated on the number of observed data. Risk factors for pain (expressed as an ordinal variable) were investigated using a multiple ordinal logistic regression with stepwise selection with variables associated with p value < 0.2 on univariate analysis. The final model was run with the selected variables and gestational age (as a continuous variable). Analysis of risk factors for curative analgesic intake was performed with a multiple logistic regression with the same strategy. For statistical analysis, SAS 9.4 software (SAS Institute, Cary, NC, USA) was used.

Results

Disposition

A total of 893 women were included and analysed at baseline, and 750 women returned for follow-up visit with documented analgesic intake, including 740 whose pain was collected prior to any CAI.

Analgesic Prophylaxis

A total of 757 (73.7%) women received an APP, among whom 588 (89.5%) had at least one step I analgesic treatment prescription, notably NSAIDs (n=506, 77.0%), 272 (41.4%) received at least one prescription for a step II analgesic treatment, and 133 (20.2%) were to take at least one adjuvant (Table 1). Adjuvants included other drugs for functional gastrointestinal disorders such as phloroglucinol. From the 750 women with documented analgesic intake at the follow-up visit, 398 (53.1%) women had an API (that was prescribed or not) including 386 (73.7%) who had an APP and 12 (5.3%) who had no APP (Table 1). Demographic data and obstetrical history including history of previous ToPs were described in the previous publication [16] and were described for each group. Except for the MToP protocol, no particular differences in women characteristics were observed (Supplementary Tables S1 and S2).

Table 1 Prescription and intake of analgesic prophylaxis
Full size table

Pain Experienced and Risk Factors

The pain prior to any CAI was collected for 740 women at the follow-up visit. Few (n=94, 12.7%) declared no pain. Mild or moderate pain was reported by 558 (75.4%) women, while 88 (11.9%) experienced intense/extreme pain. This distribution was similar in women who had an API and in those who did not (Table 2).

Table 2 Experienced pain and curative treatment intake
Full size table

On univariate analysis, API was associated with a higher risk of more severe pain. Intake of gemeprost rather than misoprostol as well as higher doses of misoprostol (initial and total dose eventually taken) were also associated with a higher risk of more severe pain. Older age, history of previous pregnancies and having at least one child were significantly associated with lower risk of pain (Table 3). Descriptive statistics regarding women characteristics and MToP protocol according to pain severity are displayed in Supplementary Table S3. Causal relation with other baseline characteristics (history of at least one ToP, number of previous ToPs, history of at least one MToP, number of previous MToPs, history of at least one surgical ToP, number of previous surgical ToPs, number of children, duration between mifepristone intake and prostaglandin analogue intake) were investigated but showed no association with the severity of reported pain (data not shown).

Table 3 Factors associated with pain severity experienced during MToP (in four classes: no pain, mild pain, moderate pain, intense/extreme pain)
Full size table

On multivariate analysis adjusted on gestational age, increasing initial (OR 1.25, 95% CI 1.06–1.47) or total dose (OR 1.15, 95% CI 1.05–1.26) of misoprostol intake were independent factors associated with higher risk of more intense pain unlike having at least one child (OR 0.74, 95% CI 0.61–0.91). API was not retained in the final model, suggesting no significant association with the risk of increased pain when adjusting with other variables (Table 3).

CAI and Risk Factors

At least one CAI was taken by 349 (46.9%) women among whom 247 (70.8%) had a step I analgesic (NSAIDs 20.5%), 101 (28.3%) a step II analgesic, 11 (1.5%) a step III analgesic, and 123 (16.5%) an adjuvant. This distribution was observed regardless the API (Table 2).

On univariate analysis, effective API was not significantly associated with CAI (p=0.136). Higher doses of misoprostol (initial and total dose eventually taken) were associated with higher risk of CAI. Women characteristics such as history of previous pregnancies, number of previous pregnancies and parenthood were associated with lower risk of CAI. As expected, more severe pain was positively associated with CAI (Table 4). Descriptive statistics regarding women characteristics and MToP protocol according to CAI are displayed in Supplementary Table S4.

Table 4 Factors associated with intake of curative analgesic treatments
Full size table

On multivariate analysis adjusted on gestational age, increasing initial dose of misoprostol (OR 1.99, 95% CI 1.47–2.68) and increasing pain (OR 3.75, 95% CI 2.93–4.79) were significantly associated with higher risk of CAI while API (OR 0.52, 95% CI 0.36–0.75) was associated with a lower risk of CAI (Table 4).

Discussion

Analgesic prophylaxis was prescribed to nearly 75% of women undergoing MToP using mifepristone 600 mg and independently of their gestational age. Eventually, only half of them had an API, prescribed or not, which was not associated with a reduced risk of experiencing more intense pain. Increasing initial or total dose of misoprostol was associated with a higher risk of experiencing pain or CAI. History of having at least one child was a protective factor regarding the risk of more intense pain. No relation was observed between gestational age and risk of pain or CAI.

An expert panel recently advocated for systematic treatment for pain associated with first-trimester MToP, with access to additional stepwise pain treatment for women [17]. The experts completed the French National College of Gynaecologists and Obstetricians practical recommendations encompassing the necessity for pain management during MToP but highlighting the absence of evidence regarding systematic prophylaxis versus curative intake [18].

Waiting for the onset of painful cramping may be too late. Step I analgesics, notably NSAIDs, were mostly prescribed in our study. WHO does not recommend paracetamol to decrease pain during abortion [12]. Ibuprofen has been reported to be more efficient than paracetamol when given as curative treatment. In a randomized clinical trial of 120 women who underwent first-trimester termination of pregnancy and who received 600 mg mifepristone orally, followed by 400 μg of oral misoprostol 2 days later, the group that received ibuprofen 400 mg achieved greater reduction in pain, 4.8±1.5 points, compared with a reduction of only 2.7±1.4 points in the group that received paracetamoI 500 mg [10]. Nonetheless, analgesic prophylaxis alone may not be satisfactory as prophylactic ibuprofen alone does not necessarily address women’s pain with early medical abortion [19]. More recently, in 140 primigravid women who received ibuprofen 600 mg and paracetamol 1000 mg (first doses taken simultaneously with misoprostol and repeated, if needed, up to three times daily), 57.7% experienced severe pain (visual analogue scale ≥70 mm) during abortion care. Additional analgesics were taken by 93.5% of women in complement to prophylactic pain medication. The authors argued for more effective analgesics than ibuprofen and paracetamol to be prescribed, especially for women with history of dysmenorrhea [20]. Our results supported that prophylaxis alone did not preclude further pain experience. However, with about half of women without any API, there is a room for improvement in preventive measures for pain mitigation.

In our study, the increase in gestational age was not associated with higher risk of pain during MToP using mifepristone 600 mg. Others have reported that, in women treated with mifepristone 600 mg and oral misoprostol 400 µg, severe pain (defined as incapacitating, preventing usual activity) was more frequent (p<0.001) > 7 WA (53% between 7 and 8 WA and 54% between 8 and 9 WA) as compared with women ≤ 7 WA (43%) [21]. Another study reported an increase in pain score according to the gestational age in women who received mifepristone 200 mg with oral misoprostol 400 or 600 µg. Scores (from 0, indicating no pain, to 10, indicating worst pain) were respectively 5.0 and 5.4 before 6 WA, 6.4 and 6.5 between 6 and 7 WA, and 5.8 and 6.1 between 7 and 8 WA (p=0.007 and p=0.013) [22]. The variety of MToP regimens captured here compared with previous studies may explain, at least partly, these differences.

Women’s characteristics have been reported to be associated with different risk of pain during MToP. A survey assessing pain during MToP with home use misoprostol in 193 women reported that nulliparity (OR 4.10; 95% CI 2.04–8.22; p < 0.0001) was positively associated with severe pain (defined as pain score ≥ 6 on a 10-point scale) [23]. Consistently, we observed that having at least one child was significantly related to a lower risk of pain.

MToP regimen potentially affects further pain onset in women. In a study involving 207 women less than 8 WA treated with mifepristone 200 mg, abdominal pain/cramps were more frequent with sublingual misoprostol 400 µg (N = 46) than with oral misoprostol 400 µg (N=161) (84.8% versus 58.4%; p<0.001). Furthermore, fever/chills were also more frequent with sublingual route (39.1% versus 16.1%; p<0.001). The transmucosal route of administration was not associated with higher risk of pain versus oral route [24]. Here, these two routes of administration did not significantly differ in terms of risk of experiencing more important pain. Alternatively, the initial and total dose of misoprostol were both strong risk factors associated with increased pain. This contrasts with a previous study reporting no difference in pain level between women receiving 400, 600 or 800 µg of misoprostol [14]. The fact that MToP protocol with 200 mg of mifepristone was not included in the present study remains a notable difference.

Another objective of the study was to evaluate the CAI. Younger patient age and nulliparity were reported to be associated with a higher risk of analgesic use [11]. Westhoffa et al. reported also that intake of painkillers was less frequent in parous women compared with nulliparous women [25]. In addition, they found that its use was more frequent in women 8–9 WA compared with women ≤ 7 WA. It was also the case in women who were randomized to use misoprostol just 24 h after mifepristone compared with those who used misoprostol after a longer interval [25]. Here, gestational age was not associated with risk of CAI unlike history of pregnancies even after adjustment on the highest pain experienced. Increased initial dose of misoprostol was associated with a higher risk of CAI, whereas an inverse association was observed with API. Surprisingly, no significant relationship between API and risk of pain experience was observed but API was associated with a decreased risk of CAI. These results may indicate that API may influence women’s attitudes towards CAI more than pain experienced.

The study bears some limitations related to its observational nature. As recently stressed in a systematic review on pain management for medical abortion before 14 WA [15], current practice differs largely. When comparing the group of women who received APP with those who did not, we did not observe any collected women characteristics that were significantly associated with higher risk of APP. Only MToP protocol (i.e. type, route and dose of PG analogues) was a risk factor. One cannot rule out that investigators’ prescribing habits could have affected APP, but these data were not collected during the study. Other psychosocial factors may be intertwined to affect pain experience and use of analgesics. For instance, Cavet et al. showed that the lack of choice regarding the method of abortion (OR 2.32; 95% CI 1.13–4.78; p = 0.0218) and the lack of information about the level of pain associated with the procedure (OR 3.27; 95% CI 1.09–9.74; p = 0.0334) were significantly correlated with severe pain [23]. Women’s perspectives regarding MToP are important. The present study did not investigate these elements. Further studies specifically designed for this purpose are expected.

Conclusions

Our study showed that the majority of women received an APP, but a quarter of them did not take it. Most women experienced painful symptoms, and about half of them actually had a CAI. The use of analgesic prophylaxis was not associated with a lower risk of pain following MToP, whereas history of at least one child was. Experienced pain did not differ according to gestational age in this study. Increasing doses of misoprostol were strong risk factors for both pain and CAI. API was associated with a decreased risk of CAI, but no relationship was observed with pain experienced.