\Stopping Dysmenorrhea: A Systematic Review of Drugs Inhibiting Uterine Contractions

Published studies in English were included where the following PICOS (population, intervention, comparator, outcome, study type) criteria were met: Participants: non-pregnant females of reproductive age experiencing the symptoms of primary dysmenorrhea. Intervention: a drug inhibiting uterine contractions and menstrual pain. Comparator: participants prior to treatment or participants taking a placebo. Outcomes: changes in both uterine contractions and menstrual pain Study Type: published human clinical studies in English. Pregnant and other non-menstruating individuals were excluded as the objective was to examine individuals with dysmenorrhea. Additionally, studies that utilized rodents were excluded as rodent models of dysmenorrhea have multiple limitations for translation to human pain experience.[ 8 ] Studies that examined pieces of myometrium in vitro were also excluded as pain could not be analyzed in these paradigms. Studies were eligible only if they assessed both uterine contractions and menstrual pain; those that measured only one of these two outcomes were excluded. Databases used included PubMed and Embase This review was conducted in accordance with systematic review methodologies as per the Cochrane Handbook and has been reported in compliance with the PRISMA statement.[ 9 , 10 ] It is based on a prospectively registered protocol on PROSPERO, available at: https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=442828 , registration number CRD42023442828. The search strategy was developed based on search strategies employed by previous similar reviews.[ 11 ] MeSH terms on PubMed and Emtree terms on Embase included “dysmenorrhea,” “uterine contraction,” “drug therapy,” and “tocolytic agents.” Other indexing terms for “dysmenorrhea” included “menstrual pain” and “painful menstruation.” Alternative indexing terms for “uterine contraction” included “myometrial contraction.” Indexing terms for “drug therapy” and “tocolytic agents” included “pharmacotherapy,” “medication,” “tocolysis,” “tocolytic effects” and specific examples of tocolytic agents including: “nitrates,” “nifedipine,” “buscopan,” “prostaglandin antagonists,” “progesterone,” “calcium antagonists,” “calcium channel blockers,” “beta adrenergic agents,” “beta agonists,” “oxytocin antagonists,” “magnesium sulfate,” “spasmolytic agent,” “spasmolytic drug,” “betamimetics,” “terbutaline,” “indomethacin,” “ketorolac,” “nitroglycerine,” “oxytocin receptor blockers,” “atosiban,” “orciprenaline,” “ritodrine,” “salbutamol,” “fenoterol,” “nylidrin,” “isoxsuprine,” and “hexoprenaline.” No filters were used during search processes. This process was piloted on PubMed and then adapted to Embase. Further details of the search strategy can be found in Appendix A . Title and abstract screening was performed independently by two reviewers with Rayyan and any discrepancies were resolved by consensus between the two reviewers.[ 12 ] Full text screening for inclusion was conducted by two independent reviewers and discrepancies were resolved between the two reviewers. ( Figure 1 ) The search returned 352 studies, of which 291 did not fit the eligibility criteria. Sixty-one full text articles were screened, of which 36 were excluded for the following reasons: Measured dysmenorrhea only but not contractions, measured contractions only but not dysmenorrhea, used rodent models only, used myometrial strips only, or used the wrong study design. Twenty-five studies were included in the review. The data extraction process was adapted from the Cochrane Good Practice Data Extraction form and was piloted prior to use.[ 13 ] Data from included studies were extracted for participants (setting, participant type, sample size of treatment group, age range), intervention (type of intervention, method, timing/frequency of delivery, duration), comparator (type of comparator, sample size of comparator group, age range, design, timing, duration), and outcomes (pain scale used, timing of pain measurement, pain before, pain after, significant reduction in pain, uterine contraction measurement tool, timing of uterine contraction measurement, uterine contraction frequency before, uterine contraction frequency after, significant effect on uterine contractions, adverse effects). Data extraction was completed by two independent reviewers using the full text copy. Discrepancies were resolved by consensus between two reviewers. The Cochrane Rob 2 tool was used to evaluate the risk of bias for randomized controlled trials. The Cochrane Rob 2 tool assesses five domains, including risk of bias due to: the randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported result.[ 14 ] The ROBINS-I tool assesses biases based on confounding, selection of participants, missing data, and measurement outcomes.[ 15 ] Two reviewers independently conducted the bias assessment and reached consensus where discrepancies existed. Each study was analyzed to determine if there was a significant difference in amount of pain and/or uterine contractions due to the pharmaceutical agent administered. Some studies did not indicate whether or not findings were statistically significant, and this was noted where necessary. Many of the studies included in this review were conducted several decades ago (median publication year 1983; range 1970 to 2014) and did not report quantitative data regarding changes in pain intensity, uterine contractions, and statistical significance, rather providing qualitative descriptions as noted in Table 1 . The data that supports the findings of this study are available in the supplementary material of this article.

Searches were performed on July 24th, 2024, resulting in 352 returns once duplicates were removed. The PRISMA flow diagram ( Figure 1 ) details the flow of studies through the selection process. A total of 61 full text articles were examined, and 36 were then excluded at this stage. A list of studies excluded can be found in Appendix B . Twenty-five studies, published from 1970 to 2014, met the review inclusion criteria. This resulted in a total of 447 participants across all included studies. [ 16 - 40 ] Details of these studies are presented in Table 1 . The studies consisted of small- and medium-sized trials conducted primarily in outpatient settings or private hospitals. Participants were diagnosed with primary dysmenorrhea via clinical symptoms, except for two studies in which vasopressin was used to induce dysmenorrhea-like symptoms in non-dysmenorrheic patients [ 25 , 26 ] A range of pharmaceutical interventions were investigated, including: NSAIDs[ 16 , 17 , 19 - 21 , 23 , 28 , 30 , 36 , 37 , 39 , 40 ], prostaglandin synthesis inhibitors[ 24 ], vasopressin antagonists [ 18 , 25 , 26 ], beta-adrenergic agonists [ 27 , 35 ], oral contraceptives [ 29 , 32 , 33 , 35 ], calcium channel blockers [ 30 ], SERMs [ 38 ], and oxytocin antagonists [ 22 , 31 ] ( Table 2 ). Of the nine randomized controlled trials, seven studies were at low risk of bias[ 17 - 23 ], one had some concerns for bias[ 24 ], and one was at high risk[ 16 ]. In the high risk study, the method of randomization and the use of blinding was not indicated. Thus, reviewers were unable to determine whether the participants and outcome assessors were aware of the intervention assigned. The study that had some concerns for bias similarly did not indicate whether the outcome assessors knew which intervention was received by each participant. Several of the non-randomized studies were at moderate[ 25 , 28 , 30 , 32 , 34 - 40 ] or serious[ 27 , 29 , 33 ] risk of bias, with only two studies having a low risk of bias[ 26 , 31 ]. The main causes of biases were the likelihood of confounding and both participants and assessors being aware of the intervention received by participants, which could bias outcome reporting and measurement, respectively. In several of the randomized and non-randomized trials, there was a lack of information about how pain intensity was measured which made the determination of bias more difficult. In these cases, the reviewers erred on the conservative side and assumed more bias was present. The results of the risk of bias process are displayed in Figures 2 and 3 . Throughout this section, the value in parentheses (n) denotes the number of participants in the corresponding treatment arm of each study. Twelve of the 25 studies included in this review examined the role of NSAIDs in dysmenorrhea treatment. All 12 found pain levels and uterine contractions to be decreased with NSAID use. However, it must be noted that some of the studies were underpowered or did not report on statistical significance. Five studies reported statistically significant reductions of both menstrual pain and uterine contractions. These studies specifically examined the NSAIDs naproxen sodium (n=10, n=4), flurbiprofen (n=4), naproxen (n=12), nimesulide (n=15), and meclofenamate (n=18). [ 16 , 20 , 23 , 28 , 39 ] Four reported a statistically significant reduction in uterine contractions and also reported a decrease in pain but did not report on the significance of this decrease, studying the NSAIDs diflunisal (n=8), naproxen sodium (n=5), naproxen (n=6), indomethacin (n=9), and mefenamic acid (n=97).[ 17 , 21 , 30 , 37 ] On the other hand, one study, which had the largest sample size (n=69), reported a statistically significant reduction in pain using naproxen (n=31), naproxen sodium (n=38), and indomethacin (n=31), while finding that contractions were decreased using naproxen sodium, although it was not specified whether this reduction was significant, and contractions were not examined using naproxen and indomethacin.[ 19 ] Two studies reported that both pain and uterine contractions were decreased though they did not specify whether these results were statistically significant or not, studying the NSAIDs naproxen (n=6), naproxen sodium (n=15), and ibuprofen (n=15).[ 36 , 40 ] Overall, naproxen sodium was the most commonly used NSAID, being used in 72 participants across the studies with reported decreases in both pain and uterine contractions. Only one study examined prostaglandin synthesis inhibitors and found a statistically significant >50% reduction in pain and statistically significant reduction in uterine contractions. The study used PDC31, which inhibits the prostaglandin F2α receptor inhibitor (n=24).[ 24 ] Although dose-dependent effects were observed, this study lacked a placebo control. Three studies examined vasopressin antagonists, with all finding a decrease in both pain and uterine contractions, though sample size and statistical significance should be taken into account. Specifically, two studies found a reduction in pain though the reduction was not specified to be statistically significant or not, accompanied by a statistically significant reduction in uterine contractions. One of these studies used three analogues of posterior pituitary hormones - 1-deamino-2-D-Tyr(OEt)-4-Val-8-Orn-vasotocin(dE-VVT), 1-deamino-2-D-Tyr(OEt)-4-Thr-8-Orn-vasotocin(dE-TVT) and 1-deamino-2-D-Tyr(OEt)-oxytocin(dE-OXY) (n=11) - while the other study used a vasopressin antagonist IV infusion (n=16) with a within-subject placebo control. [ 25 , 26 ] The third study used a vasopressin antagonist infusion and found a statistically significant reduction in both pain and uterine contractions (n=8).[ 18 ] Two within-subject studies, albeit lacking placebo controls, examined beta agonists, finding reductions in both pain and uterine contractions. One of the studies using a β2 receptor stimulator found that pain and uterine contractions were decreased but did not report whether this change was statistically significant (n=11).[ 27 , 35 ] The other study used an infusion of salbutamol and found that pain was decreased although it did not specify if this change was statistically significant, meanwhile finding a statistically significant reduction in uterine contractions (n=5).[ 27 , 35 ] Four studies used oral contraceptives, with all reporting decreases in pain and uterine contractions. One study using chlormadinone acetate, norethindrone, and methyltestosterone found that pain and uterine contractions were decreased although it did not specify whether these changes were statistically significant (n=11).[ 29 ] Another study used levonorgestrel and ethinyl estradiol and found that pain was decreased although it did not specify if the change was statistically significant, meanwhile finding a statistically significant reduction in uterine contractions (n=14).[ 32 ] One study using combination estrogen progestin pills found a significant decrease in both pain and uterine contractions (n=21).[ 33 ] Finally, one study using ethinyl estradiol and lynestrenol found a nonsignificant decrease in pain and uterine contractions, though the sample size was five.[ 34 ] Only one within subject study lacking placebo-controls assessed calcium channel blockers, finding a decrease in pain and uterine contractions. Specifically, this study used nifedipine, finding that pain was decreased although they did not specify if this was statistically significant, and finding a statistically significant reduction in uterine contractions (n=8).[ 30 ] Only one within-subject study lacking placebo controls examined SERMs, finding a reduction in both pain and uterine contractions. This study used tamoxifen and found a statistically significant reduction in both pain and uterine contractions. Only four participants were included in the contraction part of the study, while 13 participants were included in the menstrual pain portion of the study.[ 38 ] Two studies examined oxytocin antagonists, with mixed results as to whether a true decrease in pain and uterine contractions occurred. One study used a bolus injection of oxytocin analogue, finding that pain was decreased though statistical significance was not reported, along with a statistically significant reduction of uterine contractions (n=10).[ 31 ] The other study used atosiban with placebo controls and found no statistically significant reduction in pain nor uterine contractions (n=16).[ 22 ] All of the above data is depicted in Table 3 .

The findings of this systematic review suggest that drugs that reduce menstrual pain often simultaneously reduce uterine contractions. While NSAIDs are the current best supported option for treating menstrual pain, other drug options have evidence suggesting they may reduce menstrual pain and uterine contractions (prostaglandin synthesis inhibitors, vasopressin antagonists, beta agonists, OCPs, calcium channel blockers, SERMs, and oxytocin antagonists). However, more research needs to be done to determine the relative efficacy of these alternative options, ideally using noninvasive methods to image uterine contractions, so that the effects of candidate medications on uterine contractions and menstrual pain simultaneously can be accurately assessed. Additionally, focusing on compounds that reduce uterine contractility remains a priority for new drug discovery. Examining other drugs that are not included in this study but are thought to impact uterine contractions and thus dysmenorrhea could also be helpful, such as muscle relaxants like magnesium, cyclobenzaprine, and tizanidine, vasodilators like sildenafil, or palmitoylethanolamide to reduce pain perception. The overall implications of this review are the reiteration that dysmenorrhea is understudied and undertreated, where many are left without relief of pain using NSAIDs alone, but also lack other options that are well-studied or well-powered. To improve the quality of life and reduce debilitating pain for numerous people nationwide, dysmenorrhea causes and treatments must be better explored.

Primary dysmenorrhea (PD) is characterized by painful abdominal cramping occurring before or after the onset of menses without any observable anatomical pathology. Between 45-95% of individuals of childbearing age report having PD, with 2-29% experiencing severe pain that interrupts their daily functioning.[ 1 ] This condition significantly impacts quality of life for many, resulting in disruptions to work and school.[ 2 ] The mechanisms behind PD are incompletely understood, but it has been hypothesized that uterine contractions are the source of painful menstrual cramping. Patients with dysmenorrhea have been shown to have uterine contractions with higher frequency or magnitude than their counterparts without dysmenorrhea.[ 3 ] Prior research has shown that breakdown of the endometrium in some individuals leads to higher presence of prostaglandins which in turn increase uterine contractility.[ 4 ] Increased contractility could cause pain via activation of mechanoreceptors or local ischemia.[ 5 ] Thus, uterine contractions may serve as a target for pharmacotherapeutics aimed at preventing dysmenorrhea. Nonsteroidal anti-inflammatory drugs (NSAIDs) are considered the first-line treatment for PD.[ 6 ] However, as many as 18% of individuals with PD have menstrual pain that does not respond to NSAIDs.[ 7 ] Oral contraceptives are also used in the treatment of PD, however, there is limited research demonstrating their efficacy. Other treatments, such as acupuncture, yoga, massage therapy, topical heat, and nutritional changes, have also been explored but similarly lack strong evidence of their efficacy.[ 6 ] The present review examines the effectiveness of a number of drugs in reducing uterine contractility and alleviating menstrual pain, to determine if other pharmaceutical options exist for those individuals who have not found adequate relief with existing first-line treatments.