Hypothesis / aims of study
Following gynaecological cancer treatment with curative intent, the sequelae and impact of pelvic floor dysfunction become more prominent as women seek to increase participation in a range of activities in their survivorship phase.1,2 Current clinical guidelines for gynaecological cancer survivors do not provide specific recommendations for conservative pelvic floor muscle therapies in this population. The aim of this systematic review was to identify, evaluate and synthesize the evidence that has investigated the effect of non-pharmacological and non-surgical pelvic floor interventions aimed at reducing pelvic floor symptoms or changing the structure or function of the pelvic floor muscles in women following treatment for gynaecological cancer.
Study design, materials and methods
PROSPERO registration CRD42018097290. Six electronic databases (Cochrane Library 2018, CINAHL 1982-2018, MEDLINE 1950-2018, EMBASE 1980-2018, PsycINFO 1806-2018 and EMCARE 1995-2018) were systematically searched in June 2018. Randomised controlled trials, cohort studies and case series were included if they investigated the effects of pelvic floor conservative interventions aimed at reducing pelvic floor symptoms or changing the structure or function of the pelvic floor muscles. These treatments could include pelvic floor muscle training, dilator training and adjunctive therapies such as electrical stimulation. Patient-reported outcome measures of pelvic floor symptoms or health-related quality of life (HRQoL), clinician-reported or device-measured outcomes of pelvic floor muscle structure or function, or clinician-recorded vaginal, bladder or bowel complications were included. Two reviewers independently assessed the risk of bias using the PEDro scale and Cochrane risk of bias domains for randomised controlled trials and the Newcastle-Ottawa scale for cohort studies. Levels of evidence were assessed using the GRADE approach.
Five randomised controlled trials, one pre-post comparison study and two retrospective cohort studies with a combined total of 905 participants were included. Five of these studies included pelvic floor muscle training, three studies included vaginal dilator training and one study applied a clitoral therapy device. On quality assessment, the randomised controlled studies scored between 3 and 6 points on the PEDro scale (out of a possible 10 points), and the cohort studies achieved between 4 and 6 stars on the Newcastle-Ottawa scale (out of a possible 9 stars). Due to heterogeneity of intervention parameters and heterogeneity of what was reported, only two of the studies of pelvic floor muscle training and two of the studies of vaginal dilator training were able to be combined in meta-analyses. The results of these meta-analyses provided moderate level of evidence that pelvic floor muscle training with counselling and yoga or core exercises was beneficial for sexual function (SMD -0.96, 95% CI -1.22 to -0.70, I2 = 0%) and HRQoL (SMD 0.63, 95% CI 0.38 to 0.88, I2 = 0%) in cervical cancer survivors (Figure 1), and very low-level of evidence that dilator therapy reduced vaginal complications in cervical and endometrial cancer survivors (OR 0.37, 95% CI 0.17 to 0.80, I2 = 54%) (Figure 2). There were insufficient data for meta-analysis of bladder or bowel function.
Interpretation of results
Based on the findings from this review, clinicians can conclude that pelvic floor muscle training may reduce sexual dysfunction and improve HRQoL in women after gynaecological cancer treatment. In the absence of reported adverse events, clinicians may also consider offering dilator therapy to women after gynaecological cancer treatment, as this may reduce vaginal complications, although the evidence for this is based on retrospective studies only. Evidence is not yet available to determine if pelvic floor therapies can improve bladder or bowel dysfunction in this population. Limitations of the findings of this study include the heterogeneity of the pelvic floor muscle training regimens, the lack of measurement of adherence to home treatment programs, lack of long-term follow-up and low-to-moderate methodological quality of the included studies.