Surgical interventions for the management of chronic pelvic pain in women
Abstract
Background
Chronic pelvic pain (CPP) is a common gynaecological condition accounting for 20% of all gynaecological referrals. There are wide ranges of causes with overlapping symptomatology, therefore the management of the condition is a formidable challenge for clinicians. The aetiology of CPP is heterogeneous and in many cases, no clear diagnosis can be reached. It is in this scenario that the label of chronic pelvic pain syndrome (CPPS) can be applied. We defined women with CPPS as having a minimum duration of pain of at least 6 months, including with a diagnosis of pelvic congestion syndrome, but excluding pain caused by a condition such as endometriosis. Many surgical interventions have been tried in isolation or in conjunction with non‐surgical interventions in the management with variable results. Surgical interventions are invasive and carry operative risks. Surgical interventions must be evaluated for their effectiveness prior to their prevalent use in the management of women with CPPS.
Objectives
To review the effectiveness and safety of surgical interventions in the management of women with CPPS.
Search methods
We searched the Cochrane Gynaecology and Fertility Group (CGF) Specialised Register of Controlled Trials, CENTRAL, MEDLINE, Embase and PsycINFO, on 23 April 2021 for any randomised controlled trials (RCT) for surgical interventions in women with CPPS. We also searched the citation lists of relevant publications, two trial registries, relevant journals, abstracts, conference proceedings and several key grey literature sources.
Selection criteria
RCTs with women who had CPPS. The review authors were prepared to consider studies of any surgical intervention used for the management of CPPS. Outcome measures were pain rating scales, adverse events, psychological outcomes, quality of life (QoL) measures and requirement for analgesia.
Data collection and analysis
Two review authors independently evaluated studies for inclusion and extracted data using the forms designed according to Cochrane guidelines. For each included trial, we collected information regarding the method of randomisation, allocation concealment, blinding, data reporting and analyses. We reported pooled results as mean difference (MDs) or odds ratios (OR) and 95% confidence interval (CI) by the Mantel‐Haenszel method. If similar outcomes were reported on different scales, we calculated the standardised mean difference (SMD). We applied GRADE criteria to judge the overall certainty of the evidence.
Main results
Four studies met our inclusion criteria involving 216 women with CPP and no identifiable cause.
Adhesiolysis compared to no surgery or diagnostic laparoscopy
We are uncertain of the effect of adhesiolysis on pelvic pain scores postoperatively at three months (MD −7.3, 95% CI −29.9 to 15.3; 1 study, 43 participants; low‐certainty evidence), six months (MD −14.3, 95% CI −35.9 to 7.3; 1 study, 43 participants; low‐certainty evidence) and 12 months postsurgery (MD 0.00, 95% CI −4.60 to 4.60; 1 study, 43 participants; very low‐certainty evidence).
Adhesiolysis may improve both the emotional wellbeing (MD 24.90, 95% CI 7.92 to 41.88; 1 study, 43 participants; low‐certainty evidence) and social support (MD 23.90, 95% CI −1.77 to 49.57; 1 study, 43 participants; low‐certainty evidence) components of the Endometriosis Health Profile‐30, and both the emotional component (MD 32.30, 95% CI 13.16 to 51.44; 1 study, 43 participants; low‐certainty evidence) and the physical component of the 12‐item Short Form (MD 22.90, 95% CI 10.97 to 34.83; 1 study, 43 participants; low‐certainty evidence) when compared to diagnostic laparoscopy.
We are uncertain of the safety of adhesiolysis compared to comparator groups due to low‐certainty evidence and lack of structured adverse event reporting.
No studies reported on psychological outcomes or requirements for analgesia.
Laparoscopic uterosacral ligament ablation or resection compared to diagnostic laparoscopy/other treatment
We are uncertain of the effect of laparoscopic uterosacral ligament/nerve ablation (LUNA) or resection compared to other treatments postoperatively at three months (OR 1.26, 95% CI 0.40 to 3.93; 1 study, 51 participants; low‐certainty evidence) and six months (MD −2.10, 95% CI −4.38 to 0.18; 1 study, 74 participants; very low‐certainty evidence). At 12 months post‐surgery, we are uncertain of the effect of LUNA on the rate of successful treatment compared to diagnostic laparoscopy. One study of 56 participants found no difference in the effect of LUNA on non‐cyclical pain (P = 0.854) or dyspareunia (P = 0.41); however, there was a difference favouring LUNA on dysmenorrhea (P = 0.045) and dyschezia (P = 0.05). We are also uncertain of the effect of LUNA compared to vaginal uterosacral ligament resection on pelvic pain at 12 months (MD 2.00, 95% CI 0.47 to 3.53; 1 study, 74 participants; very low‐certainty evidence).
We are uncertain of the safety of LUNA or resection compared to comparator groups due to the lack of structured adverse event reporting.
Women undergoing LUNA may require more analgesia postoperatively than those undergoing other treatments (P < 0.001; 1 study, 74 participants).
No studies reported psychological outcomes or QoL.
Authors' conclusions
We are uncertain about the benefit of adhesiolysis or LUNA in management of pain in women with CPPS based on the current literature. There may be a QoL benefit to adhesiolysis in improving both emotional wellbeing and social support, as measured by the validated QoL tools. It was not possible to synthesis evidence on adverse events as these were only reported narratively in some studies, in which none were observed. With the inadequate objective assessment of adverse events, especially long‐term adverse events, associated with adhesiolysis or LUNA for CPPS, there is currently little to support these interventions for CPPS.
PICOs
Plain language summary
Surgical management of chronic pelvic pain in women
Why we did this Cochrane Review
We wanted to find out whether there are any effective and safe surgical treatments for women with chronic pelvic pain. We wanted to understand how effective these surgical treatments are compared to alternative treatments or no treatment at all.
Background
Chronic pelvic pain in women is a common and debilitating condition. Definitions vary, but generally it is defined as pelvic pain for a period of six months or greater. There are many causes of chronic pelvic pain, but these can sometimes be difficult to identify. Regardless of identifying any or the specific cause, treatment is aimed at reducing symptoms. Occasionally, a diagnostic surgery with insertion of laparoscope is completed (inserting a telescope into the belly to visualise pelvic structures). When identifiable causes of chronic pelvic pain are present, such as endometriosis (tissue similar to the lining of the womb that starts to grow in other places) or adenomyosis (tissue similar to the lining of the womb is found deep in the muscle of the womb), there may be different treatment strategies necessary than when there are no obvious problems. When no disease is identified at the time of a diagnostic surgery despite chronic pelvic pain, we may consider various surgical procedures to treat the chronic pelvic pain, including removing scar tissue originating from infection or previous operation (called adhesiolysis), or cauterising (heat treatment) or excising (removing) the nerves carrying the pain sensation from pelvis to brain (called uterosacral ligament ablation/resection). Despite it being unclear how effective these surgical treatments are, they are being offered and done.
What we found
We found four randomised controlled trials (a type of study that gives the most reliable evidence about the effects of treatment) involving 216 women with chronic pelvic pain and no identifiable cause. The main outcome measures were pain scores after surgery and quality of life. The evidence is current to 23 April 2021.
Key results
Adhesiolysis versus no surgery/diagnostic laparoscopy
We are uncertain of the effect of adhesiolysis compared with diagnostic laparoscopy on pain scores at three, six and 12 months after surgery. Pain was measured using a visual analogue score (VAS), which is a widely used rating scale where the person ranks pain from 0 (no pain) to 100 (worst pain). Adhesiolysis may improve health‐related quality of life at six months after surgery when compared to diagnostic laparoscopy.
Laparoscopic uterosacral ligament ablation versus other treatment
We are uncertain of the effect of laparoscopic uterosacral ligament ablation (LUNA) versus diagnostic laparoscopy or vaginal uterosacral ligament resection, on pain scores measured by VAS at three, six and 12 months. Women undergoing LUNA may require more pain relief after surgery than those undergoing alternative treatments.
No studies in either comparison reported on psychological outcomes.
Certainty of evidence
The certainty of the evidence ranged from very low to low. Limitations included poor reporting of study methods and imprecision (too few events, too few included studies) for some comparisons.
Authors' conclusions
Summary of findings
| Adhesiolysis compared to no surgery or diagnostic laparoscopy | ||||||
| Patient or population: health problem or population | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | № of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with Control | Risk with Adhesiolysis | |||||
| Effectiveness of treatment | The mean effectiveness of treatment was 27.3 | MD 7.3 lower | ‐ | 43 | ⊕⊕⊝⊝ | Due to the wide CIs, we are uncertain of the effect of adhesiolysis on pelvic pain scores at 3 months postsurgery. |
| Effectiveness of treatment | The mean effectiveness of treatment was 27.3 | MD 14.3 lower | ‐ | 43 | ⊕⊕⊝⊝ | Due to the wide CIs, we are uncertain of the effect of adhesiolysis on pelvic pain scores at 6 months postsurgery. |
| Effectiveness of treatment | The mean effectiveness of treatment was 24 | MD 0 | ‐ | 43 | ⊕⊝⊝⊝ | We are uncertain of the effect of adhesiolysis on pelvic pain scores at 12 months postsurgery. |
| Adverse events | Authors reported there were no complications related to surgery | 43 | ⊕⊕⊝⊝ | Due to both the low certainty of the study and the lack of structured adverse event reporting we are uncertain of the safety of adhesiolysis compared to comparator groups. | ||
| Psychological outcomes ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | No studies reported on this outcome. |
| Quality of life | The mean quality of life was 6.3 change from baseline | MD 22.9 change from baseline higher | ‐ | 43 | ⊕⊕⊝⊝ | Adhesiolysis may improve physical health‐related quality of life at 6 months postsurgery. |
| Quality of life | The mean quality of life was ‐3.7 change from baseline | MD 32.3 change from baseline higher | ‐ | 43 | ⊕⊕⊝⊝ | Adhesiolysis may improve emotional health‐related quality of life at 6 months postsurgery. |
| Requirement for analgesia ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | No studies reported this outcome. |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| See interactive version of this table: https://gdt.gradepro.org/presentations/#/isof/isof_question_revman_web_427465143401005369. | ||||||
| a Downgraded two levels due to very serious imprecision due to small study size (43 participants). | ||||||
| LUNA (laparoscopic uterosacral ligament nerve ablation) compared to diagnostic laparoscopy/other treatments | ||||||
| Patient or population: health problem or population | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | № of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with Control | Risk with LUNA (laparoscopic uterosacral nerve ablation) | |||||
| Effectiveness of treatment | 469 per 1000 | 526 per 1000 | OR 1.26 | 51 | ⊕⊕⊝⊝ | Due to the wide CIs we are uncertain about the effect of ablation or resection compared to other treatments on pelvic pain 3 months postsurgery. |
| Effectiveness of treatment | The mean effectiveness of treatment was 40.6 | MD 2.1 lower | ‐ | 74 | ⊕⊝⊝⊝ | We are uncertain about the effect of LUNA on pelvic pain scores at 6 months postsurgery. |
| Effectiveness of treatment | The mean effectiveness of treatment was 48.5 | MD 2 higher | ‐ | 74 | ⊕⊝⊝⊝ | We are uncertain about the effect of LUNA on pelvic pain scores at 12 months postsurgery. |
| Adverse events | 2 studies reported on this narratively. Both studies reported no intraoperative or postoperative events. | 121 | ‐ | Due to the narrative nature of the reporting, we are uncertain of the safety of ablation or resection compared to other treatments. | ||
| Psychological outcomes ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | No studies reported this outcome. |
| Quality of life ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | No studies reported this outcome. |
| Requirement for analgesia | 1 study reported women in the LUNA group required significantly (P < 0.001) more analgesia (measured by number of vials of tramadol used during hospital stay) with a median of 7 (range 5–9) vials vs to women undergoing VUSR who used a median of 4 (range 2–5) vials. | 74 | ‐ | Women undergoing LUNA may require more analgesia postoperatively than women undergoing other treatments. | ||
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| See interactive version of this table: https://gdt.gradepro.org/presentations/#/isof/isof_question_revman_web_424428033590495750. | ||||||
| a Downgraded two levels due to very serious imprecision: one very small study (51 participants). | ||||||
Background
Description of the condition
Chronic pelvic pain (CPP) is a common condition experienced by women. Definitions of CPP may vary, but a minimum duration of six months is considered necessary to define the pain as chronic (ACOG 2004; RCOG 2012; Jarrell 2018). Accordingly, the prevalence rates vary with the definition used to describe the condition, country and social‐economic factors. The prevalence around the world ranges from 5.7% to 26.6% (Ahangari 2014). The reported annual prevalence in primary care in the UK in women aged 15 to 73 years is 38/1000 women, almost comparable to chronic back pain (41/1000) in men and women (Zondervan 2001). There are a number of possible aetiological factors in the genesis of CPP, including changes to the central and peripheral nervous system, endometriosis, adenomyosis, adhesions, irritable bowel syndrome, interstitial cystitis, musculoskeletal sources and nerve entrapment (RCOG 2012); therefore, the condition is a challenge to diagnose and treat. It is important to note that not every woman who exhibits one or more aetiological factors of CPP actually exhibits CPP. The aetiology of CPP is heterogeneous and, in many cases, there is no clear diagnosis. It is in this scenario that the label of chronic pelvic pain syndrome (CPPS) can be applied (Baranowski 2012). Historically, CPP was the indication for 40% of diagnostic laparoscopies (Howard 1993), and an estimated 12% of hysterectomies in the US (Reiter 1990). In 2016, CPP was an indication for hysterectomy in 39.2% of women in one large study of 12,118 hysterectomies (Mowers 2016). It remains a common indication for diagnostic laparoscopy today (Mirowska‐Allen 2019). Though our understanding of CPP and our ability to better diagnose some common aetiologies of CPP has advanced, surgical intervention for CPP is still utilised for diagnosis and treatment. In addition, various pharmacological and non‐pharmacological interventions have been tried in isolation or in conjunction with each other in the management of CPP (Stones 2005).
Description of the intervention
Current and past surgical interventions commonly employed for management of CPP are diverse, as would be expected based on the variety of aetiological factors. Possible interventions include diagnostic laparoscopy, excision or ablation (or both) of endometriosis, conscious pain‐mapping laparoscopy, microlaparoscopy, adhesiolysis (surgical removal of adhesions), uterosacral nerve ablation (UNA) (laparoscopic or open surgery to disrupt the nerve plexus), presacral neurectomy, pelvic vein ligation (interventional radiology or surgical), total or subtotal hysterectomy, oophorectomy and ventrosuspension. Though clinical and non‐invasive imaging diagnoses are constantly improving, laparoscopy remains a cornerstone in establishing (or confirming) a diagnosis and treatment of identifiable pathology, but also providing reassurance to women in its absence. Occasionally, other surgical approaches (laparotomy or vaginal surgery) may be performed to achieve diagnoses or administer treatment, or both.
How the intervention might work
Broadly, surgical intervention may be targeted at the primary aetiological factor of the CPP or be used diagnostically, to guide surgical intervention, which is often implemented at the same time as the diagnostic procedure. The mechanism of the intervention depends on the aetiology. In the case of this systematic review and meta‐analysis, we will focus on the interventions that are targeted at CPPS. Therefore, any surgical interventions that may target confirmed infection or other obvious local pathology will not be considered.
Adhesions are commonly seen in women with CPP but the value of adhesiolysis in the treatment of CPP is still controversial (Cheong 2006). Adhesions may originate from an organic source (e.g. pelvic inflammatory disease (PID), diverticulitis or endometriosis) in a virgin abdomen or occur in an individual who has undergone surgery (i.e. iatrogenic adhesions). Adhesions have been shown to carry nerve fibres (Tulandi 1998), and hence may be a possible mechanical cause for pelvic pain. Adhesiolysis is aimed at releasing this mechanical restriction and therefore disrupt the perpetuation of pain. Adhesiolysis can be performed during laparoscopy or at laparotomy. It is also possible that adhesions may not be the source of CPP in the context of other pathology (deep endometriosis beneath areas of adhesions, changes in the central and peripheral nervous system following an episode of treated PID). These potential pathologies may have the potential to be recognised, while others may only be diagnosed after excluding all other possibilities.
CPP can occur as a result of an altered physical, psychological or functional state; it is likely that interaction of the neurophysiological and psychological processes is complex and dynamic (Vercellini 2009). In many cases, even when pathology is removed, there may be ongoing generation of pain in the absence of stimulus, implying there have been changes to the central nervous system (CNS) (Brawn 2014). Pain from the uterus, cervix and proximal tubes passes through nerve fibres that join at the base of the uterosacral ligaments (USL) to form the Lee‐Frankenhauser plexus. From here they exit to join the superior and inferior hypogastric plexus. The ovaries derive their nerve supply from the ovarian plexus that originate from the renal and aortic nerve plexuses and travel in proximity to the ovarian artery and veins. In addition, these fibres also network with the presacral neural network. Visceral pain originates in the intraperitoneal organs and is transmitted through the sympathetic fibres of the autonomic nervous system. In the spinal cord, there is diffuse dispersal of the pain stimuli and hence patients may perceive the sensation across several dermatomes. Sometimes, this is poorly localised and can provoke associated autonomic manifestations (Vercellini 2009).
Many surgical interventions are designed to disrupt the nerve plexus that may be involved with the perpetuation of the perception of pain. These operations include UNA, uterosacral ligament resection (USR), or presacral neurectomy (PSN). These procedures can be performed via laparoscopic route (laparoscopic uterosacral nerve ablation (LUNA), laparoscopic uterosacral ligament resection (LUSR), PSN), vaginal route (vaginal uterosacral ligament resection (VUSR)) and laparotomy (UNA, USR, PSN), and one or both of the USLs are removed/ablated (Johnson 2004; Palomba 2006; Daniels 2009). Resection of the USL usually involves the surgical removal of a small segment of the ligament usually about 1 cm in length, although there is variation in the surgical practice (Latthe 2004). The ablation of the USL can be performed using laser (Gürgan 1992), mono or bipolar diathermy to ablate approximately 1 cm to 2 cm distally to the attachment of the ligament to the cervix (Latthe 2004). Resection allows for the histological examination of the tissue to ensure the presence of nerve fibre but not when the ablative technique is being used. UNA aims to transect the USLs to sever the sensory pathways entering the Lee‐Frankenhauser plexus. Theoretically, this would only disrupt the uterine and cervical nerve pain fibres and only benefit women with pelvic pain originating from these structures. Conversely, PSN divides the hypogastric nerve plexus at the level of sacrum and potentially may be a more effective technique. However, neither procedure interrupts the sympathetic and parasympathetic nerve fibres arising in the S2‐S4 nerves, as they travel through the uterovaginal vascular plexus and pelvic sidewalls.
Besides these anatomically oriented concepts, some have explored the pelvic vascular system. The female pelvis has extensive anastomosis and collateral vascular circulation with different visceral components including the uterus, bladder and bowel. The pelvic veins are thin‐walled and poorly supported with relatively few venous valves. These features predispose women to pelvic congestion syndrome, which is defined as CPP resulting from 'incompetent' pelvic veins (Jurga‐Karwacka 2019). 'Incompetent' veins are described by dysfunctional dilation of ovarian and para‐uterine veins, slow blood flow (congestion), retrograde flow and reflux (Champaneria 2016). However, the diagnostic criteria for pelvic congestion have not been clearly defined or validated (Amin 2021), and thus the literature in this field remains inconsistent and difficult to interpret. Stopping the flow through an incompetent vein is thought to be an effective treatment strategy (Champaneria 2016). Surgical pelvic vein ligation is occasionally performed in women with pelvic congestion syndrome (Gargiulo 2003). An alternative, and now more commonly used, tool to surgical ligation is an interventional radiological procedure called ovarian vein embolisation (Maleux 2000). While there is evidence for the release of potentially pain‐producing agents from venous endothelium and perivascular nerve terminals (Stones 2000), the role of pelvic venous congestion in the pathogenesis remains unclear.
More radical, though more common, procedures such as hysterectomy or oophorectomy (or both) are also considered in women with CPP of unclear aetiology (Jarrell 2018). In the past, CPP has been listed as the main indication for 10% to 12% of hysterectomies in the US (Lee 1984; Hillis 1995). Recently, CPP was noted as an indication for hysterectomy in almost 40% of the patients (Mowers 2016). The implications on family planning and menopause are significant, so there may be hesitancy in utilising these procedures. In women with CPP of unclear aetiology, these procedures may be implemented in an attempt to 'do something'. In some cases, unrecognised pathology such as adenomyosis, pelvic congestion syndrome or chronic PID may be inadvertently treated resulting in improvement. However, posthysterectomy CPP is possible and reported in up to 22% of women (Stovall 1990). Hence, the management of women with CPP in the absence of identifiable pathology is still a significant challenge.
Notwithstanding the above anatomical considerations and clinical applications, there are fundamental neurophysiological barriers to the likely success of nerve‐cutting procedures for the relief of chronic pain in general and pelvic pain in particular. The pathophysiology of establishing a chronic pain state includes changes in the CNS including expression of pain mediators, physical sprouting of afferents, and complex reorganisation of central pain processing (Brawn 2014). As‐Sanie and colleagues found structural grey matter alterations in the brains of women with CPP, which has also been identified in patients with other types of chronic pain (As‐Sanie 2012). Anatomical, mechanical and neurophysiological factors contributing to CPP also need to be placed in a context with psychological propensities such as fear avoidance, catastrophising and depression, which all influence pain experience and render the impact of surgical interventions more difficult to discern (Stones 2007; ACOG 2020).
Why it is important to do this review
Gynaecologists use various surgical procedures for the treatment of CPP, including in the absence of obvious infectious or other pathology (CPPS). For many of these surgical procedures, there is no current consensus for best practice. For example, the most recent 2020 American College of Obstetricians and Gynecologists (ACOG) Practice Bulletin on CPP makes no mention of hysterectomy, despite its frequent use (ACOG 2020). Therefore, it is important to perform a systematic review and meta‐analysis on this area to inform practitioners of the value of the various surgical interventions in the management of women with CPPS.
Objectives
To review the effectiveness and safety of surgical interventions in the management of women with CPPS.
Methods
Criteria for considering studies for this review
Types of studies
We included randomised controlled trials (RCTs).
Types of participants
Women with CPPS (minimum pain duration six months) including those with a diagnosis of pelvic congestion syndrome, but excluding those with pain known to be caused by:
-
endometriosis;
-
adenomyosis;
-
primary dysmenorrhoea (period pain), a recurrent painful condition exclusively related to menstruation;
-
pain due to active chronic PID, that is chronic low‐grade sepsis in devitalised tubal tissue with acute exacerbations that has been incompletely treated by antibiotics;
-
irritable bowel syndrome;
-
interstitial cystitis/bladder pain syndrome;
-
urethral syndrome.
However, if a study population was overall ineligible but included an eligible subgroup of participants, we considered data pertaining to this subgroup.
Types of interventions
Any study that undertook any surgical intervention for management of CPPS, including but not limited to:
-
diagnostic laparoscopy;
-
conscious pain‐mapping laparoscopy;
-
microlaparoscopy;
-
abdomino‐pelvic adhesiolysis (by any method);
-
LUNA/open uterine nerve ablation;
-
USL resection;
-
laparoscopic/open PSN;
-
hysterectomy;
-
oophorectomy/salpingo‐oophorectomy;
-
pelvic vein ligation (surgical);
-
ventrosuspension.
Types of outcome measures
Primary outcomes
-
Effectiveness of treatment: pain: measured by validated pain scales, for example, visual analogue pain scale (VAS) scores, the McGill Pain Questionnaire (MPQ), a pain improvement rating scale, general pain experience and a gynaecological pain questionnaire. If studies used multiple pain measures, we preferred VAS, followed by the MPQ and then any other validated pain scale. Most studies were expected to have assessed these outcomes at three, six and beyond six months to exclude placebo effect. Outcomes that were reported out of this three, six, 12 months period were combined with the nearest quarter if appropriate. Where trials expressed their outcomes as recommended by Initiative on Methods, Measurement and Pain Assessment in Clinical Trials (IMMPACT), which are benchmarked in terms of pain intensity, physical function, emotional functioning and global rating of improvement (Dworkin 2008), these were to be analysed accordingly.
-
Adverse events (e.g. intraoperative and postoperative surgical complications).
Secondary outcomes
-
Psychological outcomes indicated by scores such as depression scores (Hamilton Depression Rating Scale (HAM‐D) score, Hospital Anxiety Depression Scale, which would be given priority if any other scales were presented) and mood scores.
-
Quality of life (QoL): indicated by, for example, the Medical Outcomes Study Short Form 36 (SF‐36), the Social Adjustment Survey (SAS‐WR), the Sickness Impact Profile (SIP), a general health questionnaire (GHQ), the revised Sabbatsberg Sexual Rating Scale (rSSRS) and EuroQOL‐5D (EQ‐5D). If multiple scales were used, we gave the SF‐36 priority, followed by SAS‐WR and then any other validated QoL questionnaire.
-
Requirement for analgesia.
Search methods for identification of studies
We searched for published and unpublished RCTs from each database's inception to April 2021 with no language restriction and in consultation with the Cochrane Gynaecology and Fertility Group (CFG) Information Specialist.
Electronic searches
We searched the following electronic databases:
-
The Cochrane Gynaecology and Fertility Specialised Register of Controlled Trials, ProCite platform, searched 23 April 2021 (Appendix 1);
-
CENTRAL, via the Cochrane Register of Studies Online (CRSO), Web platform, searched 23 April 2021 (Appendix 2);
-
MEDLINE, Ovid platform, searched from 1946 to 23 April 2021 (Appendix 3);
-
Embase, Ovid platform, searched from 1980 to 23 April 2021 (Appendix 4);
-
PsycINFO, Ovid platform, searched from 1806 to 23 April 2021 (Appendix 5).
Searching other resources
We handsearched reference lists of relevant trials and systematic reviews retrieved by the search and contact experts in the field to obtain any additional trials. We searched two trial registries and handsearched relevant journals and conference abstracts that are not covered in the CGF register, in liaison with the Information Specialist.
Data collection and analysis
Selection of studies
Two review authors (TG and AC) conducted an initial screen of titles and abstracts retrieved by the search and retrieved the full texts of all potentially eligible studies. Two review authors (TG and AC) independently examined these full‐text articles for compliance with the inclusion criteria and selected eligible studies. We corresponded with study investigators as required, to clarify study eligibility. We resolved disagreements by discussion. We documented the selection process with a PRISMA flow chart.
Data extraction and management
Two review authors (TG and AC) independently extracted data from eligible studies using a data extraction form designed and pilot‐tested by the authors. We resolved disagreements by discussion. Data extracted included study characteristics and outcome data. We corresponded with study investigators for further information: where results data required for inclusion in the meta‐analysis were not present either in the text or able to be extracted from graphs/figures; and to clarify methods (e.g blinding, randomization) if needed to determine risk of bias accurately.
Assessment of risk of bias in included studies
Two review authors (ML and MA) independently assessed the included studies for risk of bias using the Cochrane risk of bias assessment tool (Higgins 2011) to assess: selection (random sequence generation and allocation concealment); performance (blinding of participants and personnel); detection (blinding of outcome assessors); attrition (incomplete outcome data); reporting (selective reporting) and other bias. Judgements were assigned as recommended in the Cochrane Handbook of Systematic Reviews of Interventions Section 8.5 (Higgins 2011). We resolved disagreements by discussion.
Measures of treatment effect
For dichotomous data, we used the numbers of events in the control and intervention groups of each study to calculate Mantel‐Haenszel odds ratios (ORs) or (where events are very rare) Peto ORs. For continuous data, if all studies report exactly the same outcomes, we calculated mean difference (MDs) between treatment groups. If similar outcomes were reported on different scales, we calculated the standardised mean difference (SMD). We reversed the direction of effect of individual studies, if required, to ensure consistency across trials. We treated ordinal data (e.g. QoL scores) as continuous data. We presented 95% confidence intervals (CI) for all outcomes. Where data to calculate ORs or MDs were not available, we utilised the most detailed numerical data available that would facilitate similar analyses of included studies (e.g. test statistics, P values). We assessed whether the estimates calculated in the review for individual studies were compatible in each case with the estimates reported in the study publications.
Unit of analysis issues
The primary analysis was per woman randomised. We planned to briefly summarised data that did not allow valid analysis in an additional table and not a meta‐analysis. We sought statistical advice regarding the analysis of cross‐over trials, to facilitate the appropriate inclusion of cross‐over data in meta‐analyses.
Dealing with missing data
We analysed the data on an intention‐to‐treat basis as far as possible (i.e. including all randomised participants in analysis, in the groups to which they were randomised). We attempted to obtain missing data from the original trialists. In the event of unobtainable data, we planned imputation for the primary outcomes, and only where data were dichotomous.
If studies reported sufficient detail to calculate MDs, but there was no information on associated standard deviations (SD), we assumed the outcome to have an SD equal to the highest SD from other studies within the same analysis.
Assessment of heterogeneity
We considered whether the clinical and methodological characteristics of the included studies were sufficiently similar for meta‐analysis to provide a clinically meaningful summary. We assessed statistical heterogeneity using the I2 statistic. An I2 statistic greater than 50% was taken to indicate substantial heterogeneity (Higgins 2011).
Assessment of reporting biases
In view of the difficulty of detecting and correcting for publication bias and other reporting biases, we aimed to minimise their potential impact by ensuring a comprehensive search for eligible studies and by being alert for duplication of data. If there were 10 or more studies in an analysis, we used a funnel plot to explore the possibility of small‐study effects (a tendency for estimates of the intervention effect to be more beneficial in smaller studies).
Data synthesis
If the studies were sufficiently similar, we combined the data. We assumed that the underlying effect size was the same for all the trials in the analysis and therefore used a fixed‐effect model to obtain an overall summary in the following comparisons.
-
Adhesiolysis versus no surgery/diagnostic laparoscopy, subgrouped by severity of adhesions.
-
Surgery including the ablation or resection of USL versus diagnostic laparoscopy/other treatment.
-
Surgery including pelvic vein ligation versus diagnostic laparoscopy.
-
Surgery including pelvic vein ligation versus placebo/sham procedures.
We performed statistical analysis using Review Manager Web (Review Manager Web).
Subgroup analysis and investigation of heterogeneity
If there was significant heterogeneity and the data were available, we conducted subgroup analyses to determine the separate evidence within the following subgroups:
-
severity and extent of adhesions;
-
variation in methodology of surgery on USLs (e.g. unilateral versus bilateral ablation/resection of USLs);
-
open versus laparoscopic interventions;
-
surgical versus interventional radiological interventions.
Sensitivity analysis
We planned sensitivity analyses for the primary outcomes to determine whether the conclusions were robust to arbitrary decisions made regarding the eligibility and analysis. These analyses were to consider whether the review conclusions would have differed if:
-
eligibility had been restricted to studies at low risk of bias, defined as studies at low risk of selection bias;
-
a random‐effects model had been adopted.
Summary of findings and assessment of the certainty of the evidence
We generated summary of findings tables using GRADEpro GDT software (GRADEpro GDT). These tables evaluate the overall certainty of the body of evidence for the main review outcomes: pain (effectiveness), adverse events, psychological outcomes, QoL and requirement for analgesia using GRADE criteria.
We justified, documented and incorporated judgements about evidence certainty (high, moderate, low or very low) into reporting of results for each of these outcomes.
Results
Description of studies
Results of the search
The search revealed 24 studies that were potentially eligible and were retrieved in full text. Four studies met our inclusion criteria. One study was awaiting classification (Daniels 2009). Sixteen studies were excluded. Two study protocols were identified and added as secondary references to primary study references for included studies (Cheong 2014; Molegraaf 2016). One translation of an excluded study was identified and added as a secondary reference to the primary study reference (Swank 2003). The selection process has been documented with a PRISMA flow chart (Figure 1). See the Characteristics of included studies and Characteristics of excluded studies tables.
Included studies
Study design and setting
We included four RCTs (Peters 1992; Johnson 2004; Palomba 2006; Cheong 2014). All studies had a parallel design. One was undertaken in the Netherlands (Peters 1992), one in Italy (Palomba 2006), one in New Zealand (Johnson 2004), and one in the UK (Cheong 2014).
Participants
The trials enrolled 216 participants (overall: intervention 102, control 114; Cheong 2014: 43; Johnson 2004: 56; Palomba 2006: 74; Peters 1992: 43). All studies used power calculations to estimate the sample size (Peters 1992: intervention 21, control 22; Cheong 2014: intervention 23, control 20; Johnson 2004: intervention 22, control 34; Palomba 2006: intervention 36, control 38).
The age distribution of the participants in this review ranged from 21 to 60 years. All studies only recruited women. Palomba 2006 recruited postmenopausal women, while all other authors included either reproductive‐aged women or a combination of reproductive‐aged and postmenopausal women. Most participants were multiparous. Only Palomba 2006 reported the body mass index (BMI) of the participants, which was comparable in both the intervention and control groups. In Palomba 2006, the mean BMI in the intervention group was 27.2 (SD 2.1) and in the control group was 28.2 (SD 2.3). Most participants in both groups had symptoms for more than one year. In Cheong 2014 and Johnson 2004, more than half of the participants in the intervention and control group had some prior abdominal/pelvic operative procedure before the index operation, whereas Palomba 2006 and Peters 1992 did not specify the number and type of previous surgeries the participants had prior to the index operation. All trials rigorously excluded the presence of any significant medical and gynaecological conditions. Only Johnson 2004 did not exclude women with a mental health disorder; participants in Palomba 2006 underwent formal psychiatric and psychological assessment. In Cheong 2014, the intervention group had a statistically significantly higher baseline adhesion assessment score compared to the control group.
Interventions
All participants in the four included trials were randomised at the time of surgery.
-
One study examined laparoscopic adhesiolysis versus diagnostic laparoscopy without adhesiolysis (Cheong 2014).
-
One study compared adhesiolysis via laparotomy versus diagnostic laparoscopy without adhesiolysis (Peters 1992).
-
Two studies compared LUNA versus an alternative (diagnostic laparoscopy (Johnson 2004) or VUSR (Palomba 2006)).
No studies evaluated the following possible interventions: conscious pain‐mapping laparoscopy, microlaparoscopy, conscious pain‐mapping laparoscopy, PSN, hysterectomy, oophorectomy, salpingo‐oophorectomy, pelvic vein ligation or ventrosuspension. No studies involved a comparison of surgical versus interventional radiological interventions.
Outcome measures
-
Three studies reported the difference in post‐treatment VAS scores at three, six and 12 months (Peters 1992; Palomba 2006; Cheong 2014). One study reported the reduction of pain outcomes using the VAS score (significant reduction in pain defined as at least 50% reduction from baseline VAS pelvic pain) at six months (Johnson 2004). One study also reported the number of participants 'cured' ('cured' was defined as complete relief of those with CPP not requiring further medical treatment) (Palomba 2006).
-
One study reported QoL using a variety of questionnaire‐based QoL outcome measures (SF‐12, EQ‐5D (measured on scale of −0.59 to 1 based on responses to five questions about QoL) and EQ‐VAS measured on 0 to 100 scale, modified Endometriosis Health Profile (EHP)‐30) (Cheong 2014). Although the modified EHP‐30 scoring system has not yet been validated for measuring CPP (only for endometriosis), the data were included due to the paucity of studies reporting on this outcome, and as no other scale was used in the one study reporting QoL.
-
One study compared the costs of the intervention and alternative (Palomba 2006).
No subgroup analyses were performed since each would only have a single study comprising the data.
Excluded studies
We excluded 16 studies (Peters 1991; Saravelos 1995; Ouhilal 1999; Ventolini 1999; Aaltomaa 2001; Moon 2001; Garcia Leon 2003; Swank 2003; Fernandez 2004; Palomba 2005; Keltz 2006; El‐Din Shawki 2011; Andersen 2015; Molegraaf 2016; Muzamil Ahmed 2019; Bautrant 2020); see Characteristics of excluded studies table).
Keltz 2006 conducted a double‐blind RCT of right‐sided paracolic adhesiolysis in management of CPP. They identified and recruited 25 women for each group. The intervention group was compared with women who underwent diagnostic laparoscopy, but no paracolic adhesiolysis. The study included women who complained of CPP (duration of symptoms prior to recruitment not specified) in the absence of any other identifiable cause. Randomisation was at the time of laparoscopy. Follow‐up was four to eight weeks after the procedure. There were no missing data in terms of withdrawal after recruitment or loss to follow‐up. The trial reported significant reduction in site‐specific pain, but reported no difference in pain scores between the two groups. The trial was excluded as there were no details of duration of symptoms prior to recruitment, randomisation method and allocation concealment, and the postoperative duration of follow‐up was only up to eight weeks, which was an inappropriate study outcome.
Swank 2003, Daniels 2009, and El‐Din Shawki 2011 conducted RCTs assessing the efficacy of LUNA in the treatment of unexplained CPP. However, as all three studies included women with mild endometriosis (American Fertility Society score less than 5), these studies could not be included unless additional data were obtained, pertaining to only women without pathology identified at laparoscopy. Despite our best efforts, we were unable to obtain the data required in order to include these studies.
Studies awaiting classification
One study is awaiting classification (Daniels 2009).
Ongoing studies
We found no ongoing studies.
Risk of bias in included studies
See Figure 2 for risk of bias graph and Figure 3 for risk of bias summary.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Allocation
Sequence generation
Three studies were at low risk of selection bias related to sequence generation, as they used computer randomisation (Johnson 2004; Palomba 2006; Cheong 2014). One study was at an unclear risk of selection bias due to an unclear method of randomisation (Peters 1992).
Allocation concealment
Two studies described methods of allocation concealment and were at low risk of bias in this regard (Cheong 2014; Johnson 2004); two studies were at unclear risk of allocation bias due to unclear method of allocation (Palomba 2006; Peters 1992). In all cases, it appears allocation was concealed to the point of surgery.
Blinding
Two studies were at low risk of performance and detection bias as both the participants and the assessors (research nurses) of the trial were blinded (Johnson 2004; Cheong 2014).
Two studies were at high risk of performance bias as the participants were not blinded by sham or placebo procedures (Peters 1992; Palomba 2006).
As these are surgical trials, the surgeons were not blinded but the research nurses assessing the outcomes were blinded to the intervention that the participants received. We did not consider the lack of blinding of surgeons to cause bias because we did not expect their performance to differ due to the trial context.
Incomplete outcome data
Two studies had attrition rate greater than 10% and were, therefore, at high risk of attrition bias (Johnson 2004; Cheong 2014).
Selective reporting
One study was at high risk of selective reporting bias (Cheong 2014), as the trial was prospectively registered and the outcomes reported in the manuscript differ from the protocol. Specifically, one stated primary outcome, medication usage, was not reported. Moreover, different QoL validated tools were used than planned (SF‐12 and EHP‐30 used when only SF‐36 was planned). Three studies were at unclear risk of bias due to the fact there was no trial protocol published and were not registered in a trial registry (Johnson 2004; Palomba 2006; Peters 1992).
Other potential sources of bias
There were no risks of other bias identified.
Effects of interventions
See: Summary of findings 1 Summary of findings table ‐ Adhesiolysis compared to no surgery or diagnostic laparoscopy; Summary of findings 2 Summary of findings table ‐ LUNA (laparoscopic uterosacral ligament nerve ablation) compared to diagnostic laparoscopy/other treatments
1 Adhesiolysis versus no surgery/diagnostic laparoscopy
Primary outcomes
See summary of findings Table 1.
1.1 Effectiveness of treatment at three months
One study reported effectiveness of treatment at three months (Cheong 2014). Due to wide CIs, we are uncertain of the effect of adhesiolysis versus diagnostic laparoscopy on pain scores measured by VAS at three months after surgery (MD −7.30, 95% CI −29.87 to 15.27; 43 participants; low‐certainty evidence; Analysis 1.1).
Sensitivity analysis: due to only one study being included in this outcome, none of the preplanned sensitivity analyses could be performed.
1.2 Effectiveness of treatment at six months
One study reported effectiveness of treatment at six months (Cheong 2014). Due to wide CIs, we are uncertain of the effect of adhesiolysis versus diagnostic laparoscopy on pain scores measured by VAS at six months after surgery (MD −14.30, 95% CI −35.91 to 7.31; 43 participants; low‐certainty evidence; Analysis 1.2).
Sensitivity analysis: due to only one study being included in this outcome, none of the preplanned sensitivity analyses could be performed.
1.3 Effectiveness of treatment at 12 months
One study reported effectiveness of treatment at 12 months (Peters 1992). We are uncertain whether adhesiolysis improves pain scores at 12 months postsurgery (MD 0, 95% CI −4.60 to 4.60; 43 participants; very low‐certainty evidence; Analysis 1.3).
Sensitivity analysis: due to only one study being included in this outcome, none of the preplanned sensitivity analyses could be performed.
1.4 Adverse events
One study assessed adverse events (Cheong 2014). There was no detailed adverse event reporting postsurgery so a meta‐analysis was not possible. The authors stated there were "no complications of adverse effects associated with surgery".
Sensitivity analysis: due to only one study being included in this outcome, none of the preplanned sensitivity analyses could be performed.
Secondary outcomes
1.5 Psychological outcomes
No studies reported psychological outcomes.
1.6 Quality of life
One study reported QoL at six months postsurgery (Cheong 2014). Adhesiolysis may improve both emotional wellbeing and social support components of the EHP‐30 when compared to diagnostic laparoscopy (emotional wellbeing: MD 24.90, 95% CI 7.92 to 41.88; social support; MD 23.90, 95% CI −1.77 to 49.57; 43 participants; both low‐certainty evidence; Analysis 1.4). Adhesiolysis may improve both the emotional component and physical component of the SF‐12 when compared to diagnostic laparoscopy (emotional: MD 32.30, 95% CI 13.16 to 51.44; physical: MD 22.90, 95% CI 10.97 to 34.83; 43 participants; both low‐certainty evidence; Analysis 1.4).
1.7 Requirement for analgesia
No studies reported requirement for analgesia.
2 Laparoscopic uterosacral ligament ablation versus diagnostic laparoscopy/other treatment
Primary outcomes
2.1 Effectiveness of treatment at three months
One study reported effectiveness of treatment at three months (Johnson 2004). Due to the wide CIs, we are uncertain about the effect of ablation or resection compared to other treatments on pelvic pain at three months after surgery (OR 1.26, 95% CI 0.40 to 3.93; 51 participants; low‐certainty evidence; Analysis 2.1).
Sensitivity analysis: due to only one study being included in this outcome, none of the preplanned sensitivity analyses could be performed.
2.2 Effectiveness of treatment at six months
One study reported effectiveness of treatment at six months (Palomba 2006). We are uncertain about the effect of ablation or resection compared to other treatments on pelvic pain at six months after surgery (MD −2.10, 95% CI −4.38 to 0.18; 74 participants; very low‐certainty evidence; Analysis 2.2).
Sensitivity analysis: due to only one study being included in this outcome, none of the preplanned sensitivity analyses could be performed.
2.3 Effectiveness of treatment at 12 months
Two studies reported effectiveness of treatment at 12 months (Johnson 2004; Palomba 2006). Data were not suitable for combination in the meta‐analysis due to the non‐parametric data from Johnson 2004 requiring reporting as a dichotomous outcome, and this study is, therefore, reported narratively.
Johnson 2004 reported that at 12 months, when using an intention‐to‐treat analysis, those in the group receiving LUNA showed no difference in the rate of successful treatment for non‐cyclical pain (P = 0.854) or dyspareunia (P = 0.41) compared to those who did not receive LUNA. There was a difference in the rate of successful treatment for those receiving LUNA for dysmenorrhea (P = 0.045) and dyschezia (P = 0.05). Success was defined as a 50% or greater reduction in VAS from baseline.
We are uncertain about the effect of ablation or resection compared to other treatments on pelvic pain at 12 months after surgery (MD 2.00, 95% CI 0.47 to 3.53; 1 RCT, 74 participants; very low‐certainty evidence; Analysis 2.3).
Sensitivity analysis: due to only one study being included in the meta‐analysis of this outcome, none of the preplanned sensitivity analyses could be performed.
2.4 Adverse events
Two studies reported adverse events narratively (Johnson 2004; Palomba 2006). Both studies reported no intraoperative or postoperative complications in either group.
Sensitivity analysis: due to the narrative outcome, a random‐effects model could not be used.
Secondary outcomes
2.5 Psychological outcomes
No studies reported psychological outcomes.
2.6 Quality of life
No studies reported QoL.
2.7 Requirement for analgesia
One study reported requirement for analgesia, but the data were not suitable for meta‐analysis and are reported narratively (Palomba 2006). Participants in the LUNA group required significantly more analgesia (measured by the number of vials of tramadol used during hospital stay) with a median of seven (range five to nine) vials compared to participants undergoing VUSR who used a median of four (range two to five) vials (P < 0.001).
3 Surgery including pelvic vein ligation versus diagnostic laparoscopy
No studies compared surgery including pelvic vein ligation versus diagnostic laparoscopy.
4 Surgery including pelvic vein ligation versus placebo/sham procedures
No studies compared surgery including pelvic vein ligation versus placebo/sham procedures.
Discussion
Summary of main results
This systematic review included four RCTs, involving 216 participants and evaluating two primary interventions. The included trials evaluated adhesiolysis and LUNA for the management of CPPS.
We are uncertain about the benefit of adhesiolysis or LUNA in the management of pain in women with CPPS based on the current literature.
Only one study reported the secondary outcome of QoL (Cheong 2014). There may be a benefit to adhesiolysis (compared to diagnostic laparoscopy) in improving both emotional wellbeing and social support, as measured by the EHP‐30, and in the emotional and physical components of the SF‐12. The benefit of adhesiolysis should be interpreted with caution given that EHP‐30 is not a validated scale for measuring CPP, but for pain that is related to endometriosis. There was a demonstrated difference in the baseline adhesion score whereby women who underwent adhesiolysis had statistically more severe adhesions than controls. However, as there was an adequate randomisation and concealment process, this random imbalance will have been accounted for in the 95% CI. The study was terminated early due to protracted recruitment and cessation of study funding.
Pain is one element of a person's interpretation of their QoL. If other elements of QoL (e.g. bloating, bowel symptoms, fatigue) were improved with adhesiolysis despite a lack of improvement in pain, this may explain why there is a benefit in the QoL metric. It is not possible to attribute this effect to the placebo effect, which is appreciated in the literature on surgery for endometriosis (Abbott 2004), as the Cheong 2014 study was placebo‐controlled and included abdominal wall incisions.
There was substantial clinical heterogeneity among the studies included in this review. For example, with respect to adhesiolysis, Cheong 2014 performed laparoscopic adhesiolysis, while Peters 1992 examined the role of adhesiolysis via laparotomy. The control group in Cheong 2014 underwent diagnostic laparoscopy without adhesiolysis, but the Peters 1992 control group did not have a sham surgery, resulting in an inability to blind participants to their group allocation. We are left uncertain whether adhesiolysis is helpful, harmful or does nothing.
There was also variation in surgical techniques for USL transection employed by the studies included in this review. For example, with respect to LUNA, the point of transection of the ligament from the attachment to the uterus, whether the ligament was completely or partially transected and the laterality of the surgery differed with the trials. The technique used in Johnson 2004 involved application of electrocautery 0.5 cm from the ligamentous insertion to the cervix until the tissue was blanched, followed by complete transection using scissors and concluded by finally applying electrocautery to the base. Palomba 2006 coagulated about 2 cm from the ligamentous insertion to the cervix until the tissue was blanched, followed by excision of a small piece of tissue for the pathological confirmation using scissors and concluded by applying electrocautery to the base.
Overall completeness and applicability of evidence
In view of a low number of studies evaluating this specific population and significant clinical heterogeneity in the included studies, the evidence must be interpreted with caution. All included trials were performed in high‐income countries, so generalisability to low‐ or middle‐income countries is an important consideration.
A further conundrum faced by researchers in the evaluation of CPPS is in selecting effective outcome assessment tools. The IMMPACT recommendations have highlighted significant issues with utilising the current pain assessment tools in the evaluation of CPPS and have recommended studies in this area to use at least two or more recognised pain assessment tools to be able to identify clinically significant changes in outcome (Dworkin 2008). Similarly, instruments used to measure health‐related QoL issues need to have both clinical face validity (address issues that are of importance to patients and reflect their experiences and concerns) and adequate psychometric properties. Due to the absence of instruments that are disease‐specific to CPPS (in the absence of clear pathological aetiologies of CPP), there is concern that QoL data produced do not adequately address clinical face validity relating to this area of research (Neelakantan 2004). Our meta‐analysis can only utilise outcome measures provided by the studies.
Finally, CPPS is often multifactorial. Although we excluded studies that were designed to assess other sources of pain (e.g. irritable bowel syndrome, interstitial cystitis/bladder pain syndrome), these studies did not systematically assess whether these were present.
Quality of the evidence
The available data were extracted from four small trials including 216 women. Unfortunately, there were too few studies for most planned comparisons. A meta‐analysis could only be performed for the outcome of pain at 12 months following LUNA versus diagnostic laparoscopy.
Using GRADE methods of assessment, the certainty of the evidence for effectiveness outcomes was low or very low for all comparisons. The reason for downgrading the certainty of evidence included risk of bias (e.g. unclear method of randomisation and blinding), indirectness (e.g. the outcome was reported by the number in each treatment group where the pain was reported as 'improved'), and imprecision due to small study size. As a result, we are uncertain of the effectiveness of the interventions compared. Therefore, further research is likely to have a significant impact on our confidence in the estimate of effect and may change the estimate. Evidence on QoL was of low certainty due to imprecision from a small sample size. Evidence on adverse events was also of low certainty due to unclear and unstructured methods of adverse events reporting.
Potential biases in the review process
Numerous steps were taken during the process of this review to prevent bias. First, the CGF developed and ran the search with no limitations in language or date. Second, two review authors independently performed screening and extraction. Any conflicts that could not be resolved by the two review authors were discussed with a third review author. Despite efforts to minimise bias, there were multiple outcomes assessed using evidence from a small number of trials, with a small sample size, which may have introduced bias and resulted in difficulties extrapolating clinically relevant conclusions. Furthermore, three studies were excluded despite including relevant populations as it was not possible to distinguish between ineligible populations (e.g. trials including men and women); thus introducing the potential of publication bias.
Agreements and disagreements with other studies or reviews
The findings of this review agree with the authors' conclusions in the now withdrawn 2005 Cochrane systematic review and meta‐analysis (Stones 2005). No subgroup analysis based on the degree of severity of adhesions was performed in this version of the review. We excluded Swank 2003, though it was included in the previous review (Stones 2005). This study was determined to be ineligible due to its inclusion of men and people with chronic abdominal (not necessarily pelvic) pain.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Comparison 1: Adhesiolysis versus no surgery/diagnostic laparoscopy, Outcome 1: Effectiveness of treatment (3 months)
Comparison 1: Adhesiolysis versus no surgery/diagnostic laparoscopy, Outcome 2: Effectiveness of treatment (6 months)
Comparison 1: Adhesiolysis versus no surgery/diagnostic laparoscopy, Outcome 3: Effectiveness of treatment (12 months)
Comparison 1: Adhesiolysis versus no surgery/diagnostic laparoscopy, Outcome 4: Quality of life (6 months)
Comparison 2: Surgery including the ablation or resection of uterosacral ligaments versus other treatment (e.g. vaginal uterosacral ligament resection (VUSR), Outcome 1: Effectiveness of treatment (3 months)
Comparison 2: Surgery including the ablation or resection of uterosacral ligaments versus other treatment (e.g. vaginal uterosacral ligament resection (VUSR), Outcome 2: Effectiveness of treatment (6 months)
Comparison 2: Surgery including the ablation or resection of uterosacral ligaments versus other treatment (e.g. vaginal uterosacral ligament resection (VUSR), Outcome 3: Effectiveness of treatment (12 months)
| Adhesiolysis compared to no surgery or diagnostic laparoscopy | ||||||
| Patient or population: health problem or population | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | № of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with Control | Risk with Adhesiolysis | |||||
| Effectiveness of treatment | The mean effectiveness of treatment was 27.3 | MD 7.3 lower | ‐ | 43 | ⊕⊕⊝⊝ | Due to the wide CIs, we are uncertain of the effect of adhesiolysis on pelvic pain scores at 3 months postsurgery. |
| Effectiveness of treatment | The mean effectiveness of treatment was 27.3 | MD 14.3 lower | ‐ | 43 | ⊕⊕⊝⊝ | Due to the wide CIs, we are uncertain of the effect of adhesiolysis on pelvic pain scores at 6 months postsurgery. |
| Effectiveness of treatment | The mean effectiveness of treatment was 24 | MD 0 | ‐ | 43 | ⊕⊝⊝⊝ | We are uncertain of the effect of adhesiolysis on pelvic pain scores at 12 months postsurgery. |
| Adverse events | Authors reported there were no complications related to surgery | 43 | ⊕⊕⊝⊝ | Due to both the low certainty of the study and the lack of structured adverse event reporting we are uncertain of the safety of adhesiolysis compared to comparator groups. | ||
| Psychological outcomes ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | No studies reported on this outcome. |
| Quality of life | The mean quality of life was 6.3 change from baseline | MD 22.9 change from baseline higher | ‐ | 43 | ⊕⊕⊝⊝ | Adhesiolysis may improve physical health‐related quality of life at 6 months postsurgery. |
| Quality of life | The mean quality of life was ‐3.7 change from baseline | MD 32.3 change from baseline higher | ‐ | 43 | ⊕⊕⊝⊝ | Adhesiolysis may improve emotional health‐related quality of life at 6 months postsurgery. |
| Requirement for analgesia ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | No studies reported this outcome. |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| See interactive version of this table: https://gdt.gradepro.org/presentations/#/isof/isof_question_revman_web_427465143401005369. | ||||||
| a Downgraded two levels due to very serious imprecision due to small study size (43 participants). | ||||||
| LUNA (laparoscopic uterosacral ligament nerve ablation) compared to diagnostic laparoscopy/other treatments | ||||||
| Patient or population: health problem or population | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | № of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with Control | Risk with LUNA (laparoscopic uterosacral nerve ablation) | |||||
| Effectiveness of treatment | 469 per 1000 | 526 per 1000 | OR 1.26 | 51 | ⊕⊕⊝⊝ | Due to the wide CIs we are uncertain about the effect of ablation or resection compared to other treatments on pelvic pain 3 months postsurgery. |
| Effectiveness of treatment | The mean effectiveness of treatment was 40.6 | MD 2.1 lower | ‐ | 74 | ⊕⊝⊝⊝ | We are uncertain about the effect of LUNA on pelvic pain scores at 6 months postsurgery. |
| Effectiveness of treatment | The mean effectiveness of treatment was 48.5 | MD 2 higher | ‐ | 74 | ⊕⊝⊝⊝ | We are uncertain about the effect of LUNA on pelvic pain scores at 12 months postsurgery. |
| Adverse events | 2 studies reported on this narratively. Both studies reported no intraoperative or postoperative events. | 121 | ‐ | Due to the narrative nature of the reporting, we are uncertain of the safety of ablation or resection compared to other treatments. | ||
| Psychological outcomes ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | No studies reported this outcome. |
| Quality of life ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | No studies reported this outcome. |
| Requirement for analgesia | 1 study reported women in the LUNA group required significantly (P < 0.001) more analgesia (measured by number of vials of tramadol used during hospital stay) with a median of 7 (range 5–9) vials vs to women undergoing VUSR who used a median of 4 (range 2–5) vials. | 74 | ‐ | Women undergoing LUNA may require more analgesia postoperatively than women undergoing other treatments. | ||
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| See interactive version of this table: https://gdt.gradepro.org/presentations/#/isof/isof_question_revman_web_424428033590495750. | ||||||
| a Downgraded two levels due to very serious imprecision: one very small study (51 participants). | ||||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1.1 Effectiveness of treatment (3 months) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 1.2 Effectiveness of treatment (6 months) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 1.3 Effectiveness of treatment (12 months) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 1.4 Quality of life (6 months) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 1.4.1 Endometriosis Health Profile (EHP)‐30 ‐ Emotional Wellbeing | 1 | 43 | Mean Difference (IV, Fixed, 95% CI) | 24.90 [7.92, 41.88] |
| 1.4.2 EHP‐30 ‐ Social Support | 1 | 43 | Mean Difference (IV, Fixed, 95% CI) | 23.90 [‐1.77, 49.57] |
| 1.4.3 12‐item Short Form (SF‐12) Physical | 1 | 43 | Mean Difference (IV, Fixed, 95% CI) | 22.90 [10.97, 34.83] |
| 1.4.4 SF‐12 Emotional | 1 | 43 | Mean Difference (IV, Fixed, 95% CI) | 32.30 [13.16, 51.44] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 2.1 Effectiveness of treatment (3 months) Show forest plot | 1 | Odds Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 2.2 Effectiveness of treatment (6 months) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 2.3 Effectiveness of treatment (12 months) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
