• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

Andrews J, Yunker A, Reynolds WS, et al. Noncyclic Chronic Pelvic Pain Therapies for Women: Comparative Effectiveness [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2012 Jan. (Comparative Effectiveness Reviews, No. 41.)

Summary and Discussion

In this section, we summarize our findings about therapies for women with noncyclic chronic pelvic pain (CPP). We provide an overview of the state of the literature and outcomes for each Key Question (KQ), detail the strength of evidence for the impact of each major intervention on relevant outcomes, and describe major issues and gaps in the current body of evidence.

State of the Literature

Despite a high reported prevalence for noncyclic CPP,8 little research assessing therapies exists. While there are many publications regarding pelvic pain, there are relatively few addressing noncyclic CPP, and of those, few were high quality. Eighteen of 36 studies meeting our criteria were randomized controlled trials (RCTs); however, only 4 were placebo controlled.77,81,95,103 This lack of placebo comparison detracts from the active head-to-head trials because no initial validation of effect has been made.

Some surgical studies compared a surgical approach with diagnostic laparoscopy or compared surgical with nonsurgical management. In the nonsurgical literature, most studies compared active agents with active agents, and a number addressed hormonal therapies for endometriosis-associated CPP. We did not locate any studies of other potentially effective nonsurgical modalities (relaxation, yoga, cognitive-behavioral therapy, etc.).

The quality of those studies providing data about the prevalence of comorbidities varied by comorbidity, with the bulk of studies assessed as poor quality for a given condition (Appendix E). Among studies reporting data on the prevalence of comorbidities, the range of prevalence estimates tended to be more narrow in those studies that employed validated diagnostic criteria (e.g., Rome criteria for irritable bowel syndrome [IBS]), and studies using validated criteria were of higher quality.

Few intervention studies were rated as good77,93 or fair51,81,95quality. The higher quality evidence tended to demonstrate a lack of benefit: lysis of adhesions showed no benefit,93 a selective estrogen receptor modulator (SERM) had a negative effect on pain and the trial was stopped early.77,85 Some higher quality studies also suggested benefit of some approaches including depot leuprolide for endometriosis-associated CPP81 and botulinum toxin for myofascial-related CPP.95

The literature overall is muddled by a lack of standardized definitions for CPP and unclear diagnostic evaluation that make it difficult to determine whether studies are truly including women with CPP. Systematic reviews of effectiveness of interventions for a symptom or syndrome are fraught with difficulty; the lack of specific diagnostic criteria results in heterogeneity within and across studies. In order to effectively treat any chronic pain, one would assume that a thorough diagnostic investigation would first take place. For many conditions, this typically follows some pre-determined algorithm. However, for CPP, no such algorithm exists. Thus, in each study (and likely for each private practitioner), the patient is approached in a variable manner, and some possible diagnoses may or may not be ruled out before treatment begins. There is no assurance that the treated condition is the causative condition. Treating a symptom means that a study group will likely have a variety of etiologies—some may be amenable to the intervention under study, others may not. Compared with an intervention trial that follows established diagnostic criteria and targets an identified condition, dilution of potential benefits and harms may be occurring.

Summary of Outcomes by Key Question

KQ1. Prevalence of comorbidities. We located 26 studies 10,13-15,70-90,92 comprising 23 unique study populations addressing the prevalence of our comorbidities of interest. The prevalence rates for the co-morbidities we examined showed significant variation. Frequently no operational definition or diagnostic criteria were provided. When definitions or criteria were available, they were rarely consistent across studies. Methods of making diagnoses varied and included patient report of symptoms, patient report that they were given the diagnosis by a health care provider, evaluation by a health care provider, and objective diagnostic criteria. Undoubtedly these inconsistencies in defining the comorbidities and making the diagnoses contributed to the variation.

KQ2. Effectiveness of surgical approaches. Seven unique studies addressed surgical interventions for CPP in women.51,54,70,75,80,93,94 Given that many women with CPP are treated with invasive surgical procedures, remarkably little evidence exists that supports a surgical approach to the treatment of CPP. We identified and reviewed two studies comparing nonspecific surgical approach to nonsurgical approach for CPP therapy,70,75 one study addressing hysterectomy specifically,54 one study evaluating laparoscopic adhesiolysis at the time of diagnostic laparoscopy,93 two studies evaluating laparoscopic utero-sacral nerve ablation (LUNA) comparing to diagnostic laparoscopy,51,80 and one paper directly comparing LUNA and utero-sacral ligament resection.94

In none of the studies with comparison data was surgery in general or any specific surgical technique better than either nonsurgical intervention or the comparator technique in improving pain status in patients. Given the limited number of studies addressing heterogeneous surgical interventions and so few of good or fair quality, it is impossible to summarize the evidence for the effect for surgical interventions on any of the outcomes proposed. The most common surgical technique used as control in comparison trials was diagnostic laparoscopy; however, the direct therapeutic benefit of this procedure is poorly understood, and its role as a standalone surgical intervention for the treatment of CPP remains to be determined. The evidence is insufficient to conclude that surgical intervention is either effective or ineffective for the treatment of CPP.

KQ3. Evidence for differences in treatment outcomes by etiology. We did not locate studies addressing whether surgical outcomes differ if an etiology for CPP is identified after surgery.

KQ4. Effectiveness of nonsurgical approaches. We located 18 articles54,70,71,75,77,81-83,85,95-103 comprising 17 unique studies addressing nonsurgical approaches for CPP. Overall, little good quality data was available. This was usually due to a lack of reporting or significant bias, either for lack of blinding or failure to analyze by intent-to-treat.

Within our included studies, there was significant variation in study design that did not allow us to synthesize results or perform a meta-analysis. This significantly detracts from the ability to apply these study results to a broader population or provide concrete estimates for clinical effect. We saw this variation in definition of pelvic pain, patient populations, outcome measures, interventions, timing of outcome measures and participant followup, and comparators.

Only 4 of the 17 studies utilized a placebo arm for comparison. All of the other studies employed active treatments as comparators. As noted earlier, this lack of placebo comparison detracts from the active head-to-head trials because no initial validation of effect has been made. It could easily be assumed that each active intervention works simply by placebo effect, and this could explain why each hormone-based treatment seems equally effective in our review.

Nine of 17 studies evaluated hormone-based therapies for the treatment of pelvic pain. This is likely a result of most participant populations being recruited from endometriosis-based pain centers. However, it is well known that endometriosis is only one of many etiologies for pelvic pain, and hormone therapies one of many pain treatments.

The majority of the available literature focuses on hormone-based medical treatments, thus indirectly assuming that the majority of pelvic pain is gynecologic and/or endometriosis-driven. This bias toward gynecologic-based pain diagnoses is reflected in the absence of a true diagnostic work-up in most of these studies.

KQ5. Evidence for choosing intervention approaches. We did not identify any comparative studies of interventions for CPP in participants who had a prior intervention that failed to achieve the target outcome(s).

Strength of the Evidence for Effectiveness of Therapies


We assessed the literature by considering both the observed effectiveness of interventions and the confidence that we have in the stability of those effects in the face of future research. The degree of confidence that the observed effect of an intervention is unlikely to change is presented as strength of evidence and can be insufficient, low, moderate or high. Strength of evidence describes the adequacy of the current research, both quantity and quality, and whether the entire body of current research provides a consistent and precise estimate of effect. Interventions that have shown significant benefit in a small number of studies but have not yet been replicated using rigorous study designs will have insufficient or low strength of evidence, despite potentially offering clinically important benefits. Future research may find that the intervention is either effective or ineffective.

Methods for applying strength of evidence assessments are established in the Evidence-based Practice Centers' (EPCs) Methods Guide for Effectiveness and Comparative Effectiveness Reviews69 and are based on consideration of four domains: risk of bias, consistency in direction of the effect, directness in measuring intended outcomes, and precision of effect. We determined the strength of evidence for effectiveness outcomes (KQ 25) and assessed the body of literature deriving from studies that included comparison groups.

Tables 16 and 18 provide summaries of results, including strength of evidence, for each category of intervention addressing each KQ. Tables 17 and 19 document the strength of evidence for each domain of the major intervention-outcome combinations.

Table Icon

Table 16

Summary of results and strength of evidence of studies assessing surgical interventions.

Table Icon

Table 18

Summary of results and strength of evidence for studies assessing nonsurgical treatments for CPP.

Table Icon

Table 17

Strength of evidence domains for studies assessing surgical approaches to CPP.

Table Icon

Table 19

Strength of evidence domains for studies assessing nonsurgical approaches to CPP.

Strength of the Evidence by Key Question

KQ1. Prevalence of comorbidities. We did not assess the strength of evidence for studies addressing this KQ as we were interested in data regarding the prevalence of comorbid conditions and not effectiveness of interventions.

KQ2. Effectiveness of surgical approaches. We rated the strength of evidence for the effect of surgery on improving pain status as low or insufficient overall (Table 16). The strength of evidence was insufficient specifically for the comparison between a general surgical approach to CPP treatment as compared with nonsurgical approach. There were two studies in this area, one poor quality RCT75 and one poor quality prospective cohort,70 which were at high risk of bias, but did demonstrate consistent estimates of effects. With 2 RCTs, one of fair51 and one of poor quality,80 we considered the strength of evidence to be low for the lack of efficacy of LUNA to improve pain status over diagnostic laparoscopy alone. Both studies were consistent in the findings, with direct comparisons and precise measures, but were also subjected to potentially high levels of bias. The evidence was insufficient due lack of comparison studies to measure the effect of hysterectomy and utero-sacral ligament resection on improvement of pain status. The strength of evidence for the effect of adhesiolysis on pain status was low based on one good quality RCT.93

The evidence was insufficient to evaluate the effects of surgical treatment on any of the other prescribed outcome measures, i.e., improving functional status, satisfaction with care or quality of life, because lack of comparison studies, multiple studies and studies with significant risk of bias.

Table 17 summarizes assessments in each domain comprising strength of evidence for studies of surgical approaches.

KQ3. Evidence for differences in treatment outcomes by etiology. We did not locate studies addressing whether surgical outcomes differ if an etiology for CPP is identified after surgery.

KQ4. Effectiveness of nonsurgical approaches. We found few studies addressing nonsurgical therapies. In two of the five categories (antineuropathic agents and neuromuscular blocking agents), we found only one study, so evidence in this area is insufficient to evaluate effects on outcomes assessed as there are no comparison studies. Among hormonal therapies and nonpharmacologic interventions, none of the 11 studies used the same intervention or comparator (Table 18). Thus we have insufficient evidence to evaluate effects for outcomes for most of these as well, with the exception of low strength of evidence for the effects of raloxifene and depot leuprolide on pain stauts, assessed in one good and one fair quality placebo-controlled RCT, respectively.

For the primary outcomes and for secondary outcomes of patient satisfaction, quality of life, and functional status, nonsurgical interventions were evaluated to have insufficient strength of evidence, often because of single study. Studies generally had high risk of bias and frequent imprecision, and the studies examined diverse interventions and outcome measures (Table 18).

Table 19 summarizes domain scores for each of the domains comprising strength of effectiveness (risk of bias, consistency, directness, and precision).

KQ5: Evidence for choosing intervention approaches. We did not identify any comparative studies of interventions for CPP in participants who had a prior intervention that failed to achieve the target outcome(s).


We set inclusion criteria intended to identify studies with applicability to women with noncyclic or mixed CPP. Studies differed considerably in terms of study populations, interventions, and outcome measures (Table 20). Many of the studies were noncomparative.

Table Icon

Table 20

Summary of outcome measures.

Lack of direct comparisons of treatment options further hinders our ability to know what findings will best extend to a specific patient or to decisions about care protocols within clinics or health systems. Overall the data that are available have fair to good applicability to women with noncyclic/mixed CPP populations in specialist or secondary care settings within the United States. Table 21 summarizes applicability of the evidence across the studies included in this review; Appendix F contains applicability tables for individual KQs.

Table Icon

Table 21

Applicability of noncyclic CPP evidence reviewed.

Applicability of the Evidence for Surgical Approaches

The study populations represented by the reviewed studies consist of either highly selected women with CPP presenting to high-level referral centers specializing in CPP care or broader groups of women recruited from general gynecologic practices without CPP specialization. Some of the findings, therefore, may be generally applicable to larger populations, while some may not be applicable. Specific intervention comparator studies generally were single institution studies, which may not be applicable to general populations.

Several interventions are represented in the reviewed studies, some of which may not be applicable to general population of women with CPP. Laparoscopic lysis of abdominal adhesions, for example, is only applicable to a subset of women with CPP who undergo diagnostic laparoscopy and have visible adhesions.

Most studies compared surgical (nonspecific) interventions with nonsurgical interventions or specific surgical techniques to diagnostic laparoscopy alone. One study compared two active interventions, LUNA to utero-sacral ligament resection. The outcomes assessed by the various studies were not uniform. All studies measured the effect of interventions of pain status, but the measurement techniques and definitions for improvement or comparison were generally not uniform. Few studies assessed other outcome variables, such as functional status, quality of life, or satisfaction with care.

The settings in which studies were conducted was highly diverse, which limits general applicability of any findings. When specified, all surgical procedures were performed at facilities with operating rooms, but the various levels of specialty and referral care were quite disparate. Comparisons across studies are also complicated by variability among surgeons' training and skills.

Applicability of the Evidence for Nonsurgical Approaches

Within studies addressing surgical approaches, participants were recruited from academic centers (17/17), and of those, 12/17 were European and one was conducted in Australia. The patients referred to or attracted to this type of care may be different than the types of patients who present to a primary care physician or local gynecologist for the treatment of pelvic pain. Considering this, the participants in the included studies may not represent the typical pelvic pain patient—they could easily have a more refractory illness or complicating factors that precipitate referral. Patients in primary care practices and general obstetrics and gynecology practices may experience more improvement with the same treatments than the study populations. Another factor which may contribute to this is the preponderance of endometriosis-based CPP within the included articles. Eight of 17 studies focused primarily on endometriosis-associated pain. Thus, for patients without endometriosis or with milder disease, these studies' findings cannot be applied.

The interventions employed are highly variable. There was a large focus on hormone therapies, which are a common first-line therapy in many gynecology practices. However, because many of these studies took place in Europe, the same drugs may not be available in the United States and other countries. Thus, the results may not directly translate. Also, being located in academic centers, often the intervention is more than just a medication but also includes a significant interaction with the treatment team and multidisciplinary therapies. For most primary care and gynecology practices, this type of treatment algorithm may be difficult to replicate; thus, similar results may not be possible.

These interventions were often compared with other active treatments, which is common in practice. Many patients often switch between therapies when one becomes ineffective or intolerable. The lack of placebos, while detrimental to overall study strength, improves applicability.

The outcomes measured mostly included a form of the visual analogue scale (VAS). While helpful to measure improvement in pain ratings, most patients clinically desire not only improvement in pain, but in day-to-day functionality and quality of life. These were not often assessed. Similarly, few studies looked at overall treatment satisfaction which, in practice, typically drives the selection of certain treatments over others.

Future Research

Gaps in Areas of Research

Research addressing therapies for CPP is largely composed of trials of active agents or approaches with little placebo-controlled research and little evidence of thorough identification of patient characteristics and potential etiologies for CPP. Notably, we did not locate any studies providing evidence that surgical outcomes differ if the etiology of CPP is identified after surgery (KQ3). We did not locate any studies providing evidence for choosing one intervention over another to treat persistent or recurrent CPP after an initial intervention failed to achieve the target outcome(s) (KQ5). The following sections outline gaps in the literature and future research needs.

Etiology. The causes of CPP are not well understood and may be associated with gynecologic (e.g. endometriosis) and nongynecologic (e.g., irritable bowel syndrome [IBS]) conditions. Diagnosis of an underlying cause is complicated because the pain is rarely associated with a single underlying disorder or contributing factor. Future research needs include:

  • Analysis of distribution of the underlying causes of CPP
  • Systematic assessment of the relationship between pain and disease, using approaches such as translational research (animal to clinical models)
  • Longitudinal studies to identify subgroups at risk of developing CPP. Once these populations are identified, preventive strategies can then be investigated
  • Developing understanding of the role of pelvic floor myofascial dysfunction in CPP
  • Developing understanding of the role of visceral hyperalgesia in CPP
  • Assessment of the effects of variations in sex steroid hormone levels on pain perception in both pain-free women and those with CPP.

Iatrogenic pain. Iatrogenic pain (pain resulting from a procedure or complication of a procedure performed by a clinician) is another understudied etiologic factor for CPP. Emerging causes of iatrogenic pain include use of permanent mesh (post-mesh pain syndrome), tubal ligation/occlusion (post-tubal syndrome), and endometrial ablation (post-ablative pain syndrome). Because these women are often not followed for long periods of time and no formal reporting system for complications exists, the prevalence of chronic pain in these women is likely underappreciated.

It is imperative that, in addition to studying chronic pain that emerges de novo, practitioners look inward at how the medical community may be contributing to the CPP problem. Prevention of iatrogenic pain may be the key to saving many CPP patients from years of treatment and decreased quality of life. Future research needs related to iatrogenic causes of CPP include:

  • Understanding benefits and harms of interventions for newer procedures to treat pelvic organ prolapse and uterine bleeding
  • Assessing chronic postoperative incisional pain as factor contributing to CPP
  • Understanding connections between surgical approaches and nervous system changes that may perpetuate pain (e.g., damage to pelvic nerves from thermal energy)
  • Understanding the role of repeat surgeries in the same location with declining benefit as a pain source
  • Raising awareness of the importance of identifying an etiology for pain prior to hysterectomy with or without castration in young women.

Impact and resource utilization. CPP accounts for an estimated 1 in 10 outpatient gynecology visits, and an estimated $1.2 billion per year is spent on outpatient management of CPP in the United States (adjusted for inflation from $880 million in 1996).10 To understand better how to manage CPP care, future research needs include:

  • Assessment of the impact of CPP on the use of health services
  • Economic analysis to determine most cost effective diagnosis and management strategies.

Standardized definitions and diagnostic criteria. As noted in this review and previous studies,105 definitions of CPP vary across the literature and may conflate noncyclic and cyclic pain; Table 22 outlines definitions of CPP used in the studies comprising this review. Employing standardized definitions of CPP is a critical need in future research to establish clearly the condition under study and the effects of specific therapies. The lack of a standardized conception of CPP likely leads to a dilution of treatment effects that may be present, and clarifying our understanding of patient populations can help to bring treatment outcomes into focus. Similarly, few studies reporting comorbidity data used validated tools to diagnose comorbidities, and may relied on patient self-report. Future research needs related to defining and diagnosing CPP and comorbid conditions include:

Table Icon

Table 22

Descriptions/definitions of CPP in studies assessed.

  • Widespread use of accepted definitions of CPP across studies
  • Standard use of validated tools in studies to inform our understanding of the true prevalence of conditions reported to co-occur with CPP.
  • Larger, prospective studies examining the extent to which comorbidities modify treatment approaches and outcomes in CPP.

Diagnostic approaches. Standardized, thorough diagnostic approaches are an important area for future study as the literature currently lacks clear delineation of patient populations. Moreover, standardized evaluations can be help to ensure that clinicians are treating the actual cause(s) of CPP versus pain symptoms. The International Pelvic Pain Society has published a clinical assessment document which could be utilized to standardize the initial evaluation of potential participants.48 Research needs in this area include:

  • Estimation of the accuracy of individual and combinations of diagnostic tests for CPP
  • Assessing the role of magnetic resonance imaging (MRI) and positron emission tomography (PET)-scan in narrowing the differential diagnosis of CPP
  • Development and validation of a new pain assessment tool to capture the multidimensional experience of pelvic pain.

Standardized outcome measures. Studies used numerous outcomes measures to assess pain, quality of life, and patient satisfaction. While studies typically used a VAS measure for pain, VAS scales varied widely (Table 20), making comparisons across studies difficult. Similarly, quality of life measures varied, and patient satisfaction was typically reported using unvalidated instruments. Future research needs should address:

  • Use of standardized outcome measures such as those recommended by the IMMPACT (Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials) consensus conference.68

Nonsurgical and nonpharmacologic management. The research meeting our criteria largely assessed surgical and pharmacologic, especially hormonal, management of CPP, despite research suggesting the need to consider psychological and sociodemographic factors in understanding and treating chronic pain.106-110 The few studies to address nonsurgical or nonpharmacologic approaches were generally of poor quality but reported some benefit from a pelvic physiotherapy technique,96 ultrasonography plus counseling,102 and an integrated treatment approach75 emphasizing psychological, dietary, environmental, and physiotherapeutic treatment factors. One study reported no benefit of photographic display of pelvic findings at laparoscopy.101

Given the importance of a holistic approach to CPP, a better understanding of the potential effects of allied health, integrative medicine, and psychological and behavioral approaches is imperative, and research needs include:

  • Studies of nonpharmacologic/nonsurgical interventions for CPP including acupuncture, transcutaneous electrical nerve stimulation (TENS), physiotherapy, cognitive behavioral therapy, advice and communication about pain, and education
  • Studies of health care settings and consultation styles and their effects on the impact of the treatments for CPP
  • High quality assessments of multidimensional treatment packages, including psychological therapies.

Pharmacologic approaches. Much of the literature addressing pharmacologic interventions for CPP investigated hormonal therapies in women with endometriosis-associated CPP, and few studies were placebo controlled. Future research needs include:

  • Studies of nonhormonal agents such as tanezumab (a monoclonal antibody against nerve growth factor)
  • Placebo-controlled studies.

Surgical compared with nonsurgical approaches. The literature also lacks studies comparing surgical and nonsurgical approaches, and future research in this area is important. Two poor quality cohort studies comparing surgical and nonsurgical approaches reported similar effects on pain status between modalities54,70 and greater patient satisfaction with hysterectomy compared with nonsurgical management in one study.54 One study comparing a standard laparoscopic approach to an integrated approach emphasizing psychological and other treatments reported no differences in pain scores between approaches and improvements in “general pain experience” in the integrated approach arm.75 Research needs include:

  • Prospective, comparative studies addressing commonly employed surgical and medical treatment approaches for CPP.

Benefits of surgical treatment. Another important area for research lies in establishing whether surgical approaches are of benefit for CPP treatment, and if so, which approaches are superior. On study comparing LUNA with diagnostic laparoscopy alone (sham LUNA) reported similar outcomes between approaches.51,80 One study reported no benefit of lysis of adhesions compared with laparoscopy alone,93 and one comparing active approaches (LUNA vs. utero-sacral ligament resection)94 reported no differences in pain outcomes between groups. Future research needs in this area include:

  • Larger, rigorously conducted studies to help determine surgical outcomes, which patients are likely to benefit from surgery, and the optimal timing of intervention
  • Study design that plans to use operative findings and histopathology to categorize patients after the surgical intervention, according to pathology identified (by type) or no pathology identified; then performing planned subgroup analysis to determine if efficacy of the surgical intervention varies according to identified diagnosis.
  • Research to classify the therapeutic benefit of diagnostic laparoscopy, often used as a standard control arm in surgical studies.
  • Multicenter studies using standardized approaches to enhance comparability across studies.

Study designs that enhance applicability of findings. Most women with CPP will undergo numerous interventions to try to diagnose and remedy their pain. Study designs typically employed in most CPP research do not provide a clear method for providing an alternate intervention if an initial intervention fails, which may diminish the applicability of findings to the larger population. We did not locate any studies assessing the trajectory of care when an initial intervention is not successful (KQ5). Research to address this issue could include:

  • Developing study design methodology that plans for sequential alternate interventions if an initial intervention fails
  • Standardizing patient history taking to ensure that relevant information about prior pain history and interventions attempted is adequately captured

Methodologic Issues

While 18 of the studies identified for this review were RCTs, few adhered to standard study design and reporting conventions as reflected in the generally poor quality of studies. In particular, few trials adequately concealed treatment assignments from participants, investigators, and outcome assessors, and just under half reported an intention-to-treat-analysis (7/18). Eight studies had dropout rates exceeding 10 percent, and in 7 out of 18 trials, more than 20 percent of participants were lost to followup. Most trials (n=13/18) did report an a priori primary outcome of interest and sample size calculation. Seven reported missing or incomplete outcome data adequately.

Among cohort studies (n=3), none employed blinded outcome assessors, and 1 provided an a priori sample size calculation. All studies had significant drop-out rates (> 10 percent). Attention to study design and conduct in future research should aid researchers in drawing conclusions about outcomes of CPP treatment.

As noted, definitions of CPP varied across studies. Among studies reporting definitions for CPP, most (n=25) reported a pain duration (typically 6 months), and 5 specified an anatomic location for pain that was more specific than “lower abdomen or pelvis.” Five studies included dysmenorrhea or dyspareunia along with noncyclic pain in their definition, and 8 of 36 studies reported no definition for CPP.

Outcome measures similarly varied across studies. While many studies used a VAS scale to assess pain, few studies used the same scale. Quality of life measures used were typically validated tools (e.g., SF 36) but varied among studies. Measures of patient satisfaction were typically not validated. Ideally, future research will better characterize CPP study populations.

Studies were typically of 6 to 12 months' duration with few providing long-term followup data after the cessation of an intervention. Future studies should extend the followup period to assess the degree to which outcomes are durable, especially as many women with CPP fail to adequate pain relief despite multiple interventions. RCTs in this literature also typically included fewer than 150 women with CPP, despite the high reported prevalence of CPP. Future research including larger sample sizes should yield greater confidence in treatment effects.

A thorough diagnostic investigation is necessary to treat any chronic pain effectively. For many conditions, this investigation typically follows a pre-determined algorithm, but no such algorithm exists for CPP. Thus, in each study of CPP interventions (and likely for each private practitioner), the patient is approached in a variable manner, and some possible diagnoses may or may not be ruled out before treatment begins. There is no assurance that the treated condition is the causative condition, which certainly influences overall treatment success. We recommend that future studies outline and report the diagnostic process for participants. Pelvic pain researchers would improve the overall quality of literature if an established diagnostic algorithm was developed and put forward for use. A standardized assessment of potential study participants and standardized inclusion criteria would permit systematic analysis of data from multiple trials.

Only four trials included here were placebo-controlled; the bulk of nonsurgical studies compared active agents, and no surgical studies used a placebo. A major source of both false positive and false negative results in trials of treatment for pain is the placebo effect, which in analgesic trials is often substantial and may have a duration of weeks or months.63 The frequency of placebo effects varies among analgesic studies from zero to 100 percent.63-65 In large double-blind randomized placebo-controlled medication trials, the total percentage of patients receiving placebo reporting a clinically significant effect from the placebo intervention for neuropathic pain or functional pain syndromes is 22 percent.65 Placebo responses have also been large across a number of clinical trials for treatment of women's sexual dysfunction.111

Placebo-controlled trials of any surgical interventions are exceedingly rare. A challenge in interpreting observation trials of surgery, or randomized trials of surgical versus nonsurgical therapy, is that patients could not reasonably be blinded to the intervention,65 which may be responsible for some overestimation of surgical benefits for pain relief,112 as surgery can be associated with important placebo effects.113 Operations that later proved to be useless, including gastric freezing for duodenal ulcers, internal mammary artery ligation for angina pectoris, and knee arthroscopy were initially reported to improve or eliminate the pain of 60 to100 percent of patients for a year after surgery.63,114 Based upon the very small number of placebo-controlled randomized trials, the magnitude of the placebo effect of surgery for pain is about 35 percent.114

To understand the true effects of therapies for CPP, future research of interventions for relief of CPP should be placebo-controlled, with the exception of small pilot studies to evaluate the potential of a new intervention to be utilized in a future placebo-controlled trial.


The literature addressing therapies for CPP in women is of largely poor quality and inconclusive. While half of the literature comprised RCTs, only two were good quality77,93 and three were fair.51,81,95 Studies providing cross-sectional data about the prevalence of comorbidities varied in quality but were largely poor. Nonetheless, some conclusions can be drawn.

Among studies reporting data on the prevalence of comorbidities, prevalence estimates tended to be more tightly clustered in those studies that employed validated diagnostic criteria (e.g., Rome criteria for IBS), and studies using validated criteria were of higher quality. Studies of nonsurgical approaches typically addressed hormonal management of endometriosis-related CPP and were not placebo-controlled, thus limiting our ability to understand whether hormonal therapies would be beneficial for women with CPP without endometriosis and whether pain relief reported is due simply to the placebo effect. Some studies reported benefits of other nonsurgical approaches, but nonhormonal and nonpharmacologic management remain understudied.

Across the literature, higher quality intervention studies tended to demonstrate a lack of benefit: lysis of adhesions showed no benefit,93 a SERM had a negative effect on pain.77,85 Some studies suggest benefit of some approaches including depot leuprolide for endometriosis-associated CPP.81

Aside from the lack of benefit reported for adhesiolysis,93 little evidence demonstrates the effectiveness of surgical approaches. Studies typically reported no differences in improvements in pain scores between groups in studies comparing surgical interventions with diagnostic laparoscopy alone or active surgical interventions. Studies comparing hysterectomy and nonsurgical management54,70 reported similar improvements in pain scores between groups and greater patient satisfaction among women undergoing hysterectomy in a sample of women electing hysterectomy.54 Despite numerous surgical techniques used extensively in treating CPP, few studies included more than 50 participants, and few were considered high quality. All of the studies with comparison data failed to demonstrate that surgery in general or any specific surgical technique was more efficacious than either nonsurgical intervention or the comparator technique in improving pain status in patients. No surgical technique was superior, and the evidence to conclude that surgical intervention is either effective or ineffective for the treatment of CPP is insufficient.

Indeed, the strength of evidence for effectiveness across interventions ranges from insufficient to low with few studies comparing the same intervention and variable patient populations. The literature lacks placebo-controlled studies, studies of nonhormonal interventions, studies of nonpharmacologic interventions, and studies comparing medical and surgical management. Studies establishing the benefit of surgery as a treatment option for CPP are also lacking.

Despite a prevalence for CPP rivaling that of widely studied conditions such as asthma,8 little research assessing therapies exists. While there are many publications regarding pelvic pain, there are relatively few addressing noncyclic CPP, and of those, we evaluated few as providing high quality evidence. In sum, the literature overall is muddled by a lack of standardized definitions for CPP and unclear diagnostic evaluation that make it difficult to determine whether studies are truly including women with CPP. Similarly, understanding comorbidity prevalence with CPP is difficult as conditions may be considered part of the differential diagnosis or a concomitant condition.

Improved characterization of the targeted condition, intervention, and population in CPP research is necessary to inform treatment choices for this commonly reported entity. A uniform definition of CPP and standardized evaluation of participants are lacking across the literature; study populations likely vary widely, and studies may be reporting effects from treating symptoms rather than a diagnosed condition. Thus our understanding of potential treatment effects is diluted.

Cover of Noncyclic Chronic Pelvic Pain Therapies for Women: Comparative Effectiveness
Noncyclic Chronic Pelvic Pain Therapies for Women: Comparative Effectiveness [Internet].
Comparative Effectiveness Reviews, No. 41.
Andrews J, Yunker A, Reynolds WS, et al.

PubMed Health Blog...

read all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...