Between 2011 and 2014, the total abortion rate in the United States declined 12%, whereas early medication abortions increased from 24% to 31% of all nonhospital abortions.1 Women report experiencing moderate to severe pain during medication abortion with maximum pain scores ranging from 6 to 8 on an 11-point scale.2–5 Limited data exist regarding the most effective analgesic regimen. Ibuprofen has been found to be superior to acetaminophen,4 although it is not beneficial when used prophylactically.2 A review of pain management during medication abortion revealed that 75% of women also required a narcotic.6 With nationwide goals of reducing narcotic use, we sought to find a nonnarcotic option for analgesia during medication abortion.
Pregabalin, a γ-aminobutyric acid analog, is increasingly used as a preoperative medication to decrease acute pain. Maximum therapeutic effect is achieved in 1 hour with a half-life of 6 hours.7 The rapid onset of action could be ideal for coadministration with misoprostol, because drug levels would peak at the same time as the misoprostol dissolves. A meta-analysis of pregabalin in gynecologic procedures demonstrated decreased pain scores, analgesic and opioid use, and postoperative nausea and vomiting with no difference in adverse effects compared with placebo.8
Our primary objective was to evaluate whether 300 mg pregabalin, coadministered with misoprostol during a medication abortion, reduces maximum pain scores. We also hypothesized pregabalin would decrease the use of adjuvant narcotic pain medication.
MATERIALS AND METHODS
We conducted a randomized, double-blind, placebo-controlled trial at the University of Hawaii Women's Options Center in Honolulu, Hawaii. We obtained institutional review board approval from the University of Hawaii's Human Studies Program (#22753) and the trial was registered in ClinicalTrials.gov (NCT02782169). After consenting to a medication abortion, women were approached about study participation and provided informed written consent. Women received counseling regarding pain expectations with medication abortion before being approached about study participation. We did not standardize how health care providers counseled patients regarding these expectations. Patients took 200 mg mifepristone orally in the office and were given 800 micrograms misoprostol to be taken buccally at home 24–48 hours later at a time of their choosing. All women were scheduled for office follow-up with their physician, where no data were collected for this study.
We included women aged 18 years or older who had a pregnancy up to and including 70 days of gestation and were willing to receive cellular phone text messages and complete electronic surveys over a 72-hour study period. Women were excluded if they had a contraindication or allergy to ibuprofen, acetaminophen, oxycodone, or pregabalin; a desire for vaginal misoprostol administration; an inability to read English; or participation in this trial during a prior pregnancy. Different routes of misoprostol administration are reported to result in different pain scores and side effects; therefore, we limited enrollment to only those women who chose buccal misoprostol administration to standardize this aspect of the medication abortion.9
In choosing a dosage of pregabalin, the meta-analysis of gynecologic procedures used dosages of 100–900 mg and showed a positive net effect.8 A second meta-analysis showed no significant difference between single and repeated dosing, although increased sedation was seen with multiple dosing.10 In an induced acute pain model, the dose of pregabalin needed to decrease pain by 30% was found to be 252 mg, so we chose to study a single dosage of 300 mg.11
Participants were randomized to one capsule of 300 mg pregabalin or a matched placebo to be swallowed immediately before buccal placement of misoprostol. A researcher not involved in the conduct of the study used a computer-generated randomization scheme of varied block sizes and placed the allocated study capsule in sequentially numbered bags identified only by study identification number so as to maintain blinding of participants and researchers. A supply of 12 800-mg ibuprofen tablets and eight 5/325-mg oxycodone with acetaminophen tablets was dispensed to all participants for analgesia. Participants were advised to take ibuprofen (one tablet every 6 hours) first with oxycodone with acetaminophen (one tablet every 3–4 hours) as needed for breakthrough pain.
Baseline demographics were collected as well as an anticipated maximum pain score for the upcoming abortion on an 11-point numerical rating scale, described as a numbered scale from 0 to 10, in which 0 indicates no pain and 10 indicates the most pain possible. Data were then collected at six specified time points over the 72-hour study period by online survey, starting immediately after taking the study medication capsule and misoprostol. Participants were asked to rate their pain during the medication abortion on an 11-point numerical rating scale, to report the type and number of analgesic tablets taken, and to indicate side effects experienced. After completion of the first survey, a text message prompt was sent automatically after 2, 6, 12, 24, and 72 hours. Each prompt included a link to a new survey that contained the same questions as well as a question about maximal pain on the numerical rating scale since the previous survey. A 5-point Likert scale assessed satisfaction with the abortion process and analgesia regimen at 24 hours. Participants who did not provide survey responses were contacted to provide a retrospective report after 2 and 6 hours by text message, by phone at 24 hours, and email at 72 hours unless a response was received.
Our primary outcome was maximum pain score on the 11-point numerical rating scale. The trial was powered to detect a clinically significant reduction in pain of 1.3 on the 11-point numerical rating scale.12,13 The numerical rating scale was chosen to provide a pain scale that would look the same when viewed on an electronic device of any size when answering the surveys as opposed to the size-sensitive visual analog scale. Based on a trial using the same medication abortion and analgesia regimen, we anticipated the maximum pain score in the placebo group to be 7.3±2.2.2 Assuming a normal distribution, a sample size of 92 participants (46 in each arm) had 80% power to detect a difference in mean pain score of 1.3 with a significance level of .05 using a two-sided t test. To account for up to a 20% dropout rate, we enrolled 110 participants. If the data were not normally distributed, 42 participants in each arm would have the same power using the Mann-Whitney U test. Secondary outcomes included pain scores at specific time points, analgesic use (ibuprofen and oxycodone with acetaminophen), side effects experienced, and satisfaction with the abortion process and pain control.
We analyzed all participants who provided a response for maximum pain, either in real-time surveys or in retrospective report, whether or not they took the study capsule as directed. Participants who were completely lost to follow-up were excluded from analysis. Categorical variables were analyzed using χ2 and Fisher exact tests. Continuous variables were analyzed with t tests and Mann-Whitney U tests. Analysis was performed using SPSS 24.
From June 2015 to October 2016, 241 women presenting for medication abortion were screened for inclusion. We enrolled and randomized 110 participants (56 pregabalin and 54 placebo) (Fig. 1). Baseline demographic characteristics were similar between groups (Table 1). The mean age of participants was 27 years (range 18–41 years) with a mean gestational age of 54 days (range 41–70 days) by ultrasonography. More than 93% of participants completed all six surveys, with more than 80% of all responses received within 2 hours of the requested time. Three women were lost to follow-up (2.7%). Blinding was maintained with the same proportion of women in each group (59.3%) believing they received pregabalin.
Participants in both groups had a mean anticipated maximum pain score of 6.8±2.0 (P=.99). The primary outcome of experienced mean maximum pain score was 5.0±2.6 and 5.5±2.2 in the pregabalin and placebo groups, respectively (P=.32). The maximum pain score distribution was not statistically different between groups (P=.64) (Fig. 2). The secondary outcome of mean pain scores at specified time points is graphically represented in Figure 3. Immediately after the misoprostol, participants in the pregabalin group reported statistically higher pain levels, although both were less than 1 on the numerical rating scale (0.7±1.3 vs 0.3±0.9; P=.04). Between 2 and 6 hours, maximum pain scores were lower in the pregabalin group (3.6±2.5 vs 4.6±2.3; P=.04) with the same finding between 6 and 12 hours (1.9±2.2 vs 2.8±2.0; P=.04), although this difference may not be clinically significant, particularly given evaluation of so many time points and the possibility of random variation. Maximum pain scores at other time points did not differ between groups. From 12 hours on, 62–80% of participants who received pregabalin reported no pain at each time point. For the placebo group, 46–82% reported no pain in the same period.
Numbers of analgesic tablets were examined as a secondary outcome. Median ibuprofen use was one tablet in the pregabalin group (interquartile range 0–4, range 0–8) and two tablets in the placebo group (interquartile range 1–3, range 0–8; P=.34). Median oxycodone with acetaminophen use was 0 tablets in the pregabalin group (interquartile range 0–1, range 0–5) and 0.5 tablets in the placebo group (interquartile range 0–1, range 0–8; P=.11).
Ibuprofen was taken before the study capsule by 30.9% (17/55) of the pregabalin group and 19.2% (10/52) of the placebo group (P=.17). A narcotic was taken before the study capsule by 7.3% (4/55) of the pregabalin group and 3.8% (2/52) of the placebo group (P=.68). Although participants were instructed to begin their pain management with ibuprofen, seven participants reported taking oxycodone before taking any ibuprofen (2/55 in the pregabalin group, 5/52 in the placebo group, P=.21).
Pregabalin use was associated with not requiring any additional analgesics. Ibuprofen was needed by 72.7% (40/55) of the pregabalin group compared with 88.5% (46/52) of the placebo group (P=.04). Those taking pregabalin were also less likely to use oxycodone with acetaminophen, 30.9% (17/55) compared with 50% (26/52) (P=.04).
To further understand the potential analgesic-sparing effect of pregabalin, only analgesics taken after the study capsule were considered. Excluding medication taken before the study capsule, ibuprofen was used by 63.6% (35/55) of the pregabalin group compared with 86.5% (45/52) of the placebo group (P<.01). Narcotics were used by 29.1% (16/55) of the pregabalin group compared with 50% (26/52) of the placebo group (P<.03).
The list of experienced side effects in each group is shown in Table 2. Participants who received pregabalin reported significantly less constipation than placebo (P<.02), but significantly more dizziness (P<.001) during the abortion. When including only those side effects reported after taking the study capsule, dizziness was more significant in the pregabalin group (P<.001), as was sleepiness (P<.04) and blurred vision (P<.05); no difference remained for constipation. There was no difference between groups for nausea, vomiting, diarrhea, headache, or dry mouth.
Satisfaction scores with the abortion process and analgesia are shown in Table 3. In the pregabalin group, 40.7% were very satisfied with the abortion process compared with 21.6% in the placebo group (P=.03). Satisfaction with analgesia was also higher in the pregabalin group with 47.2% very satisfied with their pain control compared with 21.6% in the placebo group (P=.006).
We did not find a difference between groups in our primary outcome (mean maximum pain score), although pregabalin was associated with higher satisfaction and a greater likelihood of not requiring additional analgesics at the expense of more reported side effects. Less than one third of pregabalin users took a narcotic compared with half of the placebo group. This is in contrast to previous reports of narcotic use in up to 75% of patients undergoing medication abortion.6 Given the potential for substance use disorders and side effects of narcotics, decreasing its use could have substantial public health benefit. This trial also provides evidence that, if narcotics are prescribed to patients, a small number of pills should be prescribed, because the interquartile range of the placebo group was 0–1 tablets.
Maximum pain scores were lower than expected with a mean maximum of 5.5 in the placebo group compared with 6 to 8 in previous research.2–5 Pain duration was also shorter than expected, with more than half of participants in both groups being pain-free by 12 hours after misoprostol in contrast to a previous study using the same regimen in which 60% of participants reported pain lasting 5 or more days.2 One potential reason for lower reported pain scores is our use of a real-time and short interval data collection tool. This allowed for immediate recording of pain scores, decreasing errors associated with retrospective report and recall bias. Prior studies have measured pain through retrospective maximum pain score, daily diary cards, and scores at the time of analgesia use. Compared with those studies, our study has comparable prior abortion rates and gestational ages. Our population included a slightly lower proportion of primigravid women (30% vs 40–50%), although the proportion of nulliparous women was similar (42% vs 45–65%).2–5
Side effects were common in both groups, particularly gastrointestinal, which are commonly experienced in early pregnancy as well as with misoprostol. More than 80% of women in the placebo group reported nausea at least once, almost 60% vomited, and more than 50% had diarrhea. These rates were lower in the pregabalin group, but not significantly. Pregabalin was associated with higher rates of sleepiness, dizziness, and blurred vision, which are known side effects.7 Despite the increase in some side effects, the pregabalin group had higher satisfaction with the abortion and pain control. This difference cannot be attributed to inadequate blinding, because equal numbers of participants in each group believed they received pregabalin (59%).
Strengths of this study include the randomized, double-blind controlled trial design and high follow-up, with 93.6% of participants completing all six surveys and only 2.7% loss to follow-up. Our combination of text message prompt and online survey data collection tool may have contributed to this success. The prompt through text allowed participants immediate access to the survey while protecting privacy by redirecting them to a survey without identifiers. Only three women (1.2% of those assessed for eligibility) were excluded as a result of lack of a cellular phone.
One limitation is that we included only the buccal route of misoprostol. Other routes of misoprostol administration (oral, vaginal, sublingual) have shown differing side effect profiles and have different timing options for administration.9 This single regimen may limit generalizability to other modes of misoprostol administration, although the mechanism of action for pregabalin would not change. Another limitation is that all pain medications were dispensed, not prescribed, which may have influenced medication use. However, by eliminating the barrier of payment for prescriptions, it is likely that our finding is higher than real-world use.
Pain that occurs over a period of time, as with medication abortion, is difficult to quantify with a single pain scale and is subjective by nature. We chose to use multiple data points and real-time assessments to improve validity and mitigate recall bias. Although more than 80% of responses were submitted within 2 hours of the requested time, another limitation is the comparison of retrospectively recalled and real-time data, which may not be equivalent. We did find significant differences in analgesic use and satisfaction, raising the possibility that type II error affected our primary outcome.
When studying pain, it may be that a maximum pain score is not the most important outcome. A peak value at a moment in time may not truly represent a longitudinal experience or reflect patients' satisfaction with their care. Although peak pain scores were not decreased with pregabalin, its association with increased satisfaction and decreased need for nonsteroidal antiinflammatory drugs and narcotics along with its ease of administration make it a worthwhile adjunct in medication abortion care.
1. Jones RK, Jerman J. Abortion incidence and service availability in the United States, 2011. Perspect Sex Reprod Health 2014;46:3–14.
2. Raymond EG, Weaver MA, Louie KS, Dean G, Porsch L, Lichtenberg ES, et al. Prophylactic compared with therapeutic ibuprofen analgesia in first-trimester medical abortion: a randomized controlled trial. Obstet Gynecol 2013;122:558–64.
3. Hamoda H, Ashok PW, Flett GM, Templeton A. A randomised controlled trial of mifepristone in combination with misoprostol administered sublingually or vaginally for medical abortion up to 13 weeks of gestation. BJOG 2005;112:1102–8.
4. Livshits A, Machtinger R, David LB, Spira M, Moshe-Zahav A, Seidman DS. Ibuprofen and paracetamol for pain relief during medical abortion: a double-blind randomized controlled study. Fertil Steril 2009;91:1877–80.
5. Hamoda H, Ashok PW, Dow J, Flett GM, Templeton A. A pilot study of mifepristone in combination with sublingual or vaginal misoprostol for medical termination of pregnancy up to 63 days gestation. Contraception 2003;68:335–8.
6. Penney G. Treatment of pain during medical abortion. Contraception 2006;74:45–7.
7. Toth C. Pregabalin: latest safety evidence and clinical implications for the management of neuropathic pain. Ther Adv Drug Saf 2014;5:38–56.
8. Yao Z, Shen C, Zhong Y. Perioperative pregabalin for acute pain after gynecological surgery: a meta-analysis. Clin Ther 2015;37:1128–35.
9. Fiala C, Cameron S, Bombas T, Parachini M, Saya L, Gemzell-Danielsson K. Pain during medical abortion, the impact of the regimen: a neglected issue? A review. Eur J Contracept Reprod Health Care 2014;19:404–19.
10. Mishriky BM, Waldron NH, Habib AS. Impact of pregabalin on acute and persistent postoperative pain: a systematic review and meta-analysis. Br J Anaesth 2014;114:10–31.
11. Wong W, Wallace MS. Determination of the effective dose of pregabalin on human experimental pain using the sequential up-down method. J Pain 2014;15:25–31.
12. Todd KH, Funk KG, Funk JP, Bonacci R. Clinical significance of reported changes in pain severity. Ann Emerg Med 1996;27:485–9.
13. Bijur PE, Latimer CT, Gallagher EJ. Validation of a verbally administered numerical rating scale of acute pain for use in the emergency department. Acad Emerg Med 2003;10:390–2.