The impact of multimodal physiotherapy in an interdisciplinary setting for the management of women with persistent pelvic pain and pelvic floor tension myalgia

Abstract

Purpose

To determine the effectiveness of multimodal physiotherapy in an interdisciplinary setting for the treatment of women with persistent pelvic pain (PPP) who have concomitant pelvic floor tension myalgia (PFTM).

Materials and methods

Prospective cohort study with 2-year follow-up at an interdisciplinary tertiary referral centre for PPP and endometriosis. Thirty-four women with PPP of gynaecological origin >6 months duration and PFTM on examination underwent individualised physiotherapy. Outcome measures were pain, measured by Visual Analogue Scale (VAS) on a scale of 0–100, quality of life assessed by AQoL-6D score range 0.0–1.0 and sexual function measured by the Female Sexual Function Index (FSFI).

Results

Participants attended a median of 4 physiotherapy sessions (range 1–11) with the median treatment duration 7 weeks (range: 1–16 weeks). Results at 3 months demonstrated a significant decrease in VAS scores for dysmenorrhoea (72.5 vs. 52, p=.015), low back pain (55 vs. 47.5, p=.004), dyschezia (45 vs. 30.5, p=.031), abdominal pain (52 vs. 36.5, p=.049) and non-menstrual pain (53 vs. 41, p=.018) with a non-significant decrease in dyspareunia (57 vs. 50, p=.365). At 2 years, improvements in dysmenorrhoea persisted (72.5 vs. 60, p=.002) and dyspareunia was significantly improved (57 vs. 36, p=.033). For the 11/24 (46%) women who had no medical or surgical co-intervention throughout the study duration, quality of life score was improved at 2 years compared with baseline (0.72 vs. 0.84, p=.002).

Conclusions

Physiotherapy may improve certain pain outcomes in the short and long term. It offers an additional treatment option for women with PPP.

Keywords:

• Persistent pelvic pain
• physiotherapy
• pelvic floor tension myalgia
• chronic pelvic pain
• endometriosis

Introduction

Persistent pelvic pain (PPP) is defined as continuous or recurrent pain in the pelvis (below the umbilicus) for 6 months or more [1] that affects function and requires medical care. This broad definition and the issues obtaining large-scale epidemiological data mean that 4–43% of women describe PPP [2]. Gynaecologically, PPP accounts for 40–50% of laparoscopies and 12% of hysterectomies [2,3]. Although frequently recommended in clinical practice guidelines, laparoscopy creates no change or a worsening of pain for 20% of patients with endometriosis (the most common pathological cause of PPP) [4], and more than 60% have a repeat laparoscopy [5], indicating suboptimal outcomes for a significant number of patients [6]. PPP is a considerable individual and systemic healthcare burden, with $6 billion in direct costs per year related to medical and surgical treatments in Australia alone [7]. The individual cost per sufferer per year in Australia is estimated at USD $20,898, the majority due to lost productivity [8].

It is well established that there are a variety of biological, psychological and social contributors to PPP that have complex contributions to symptoms due to their confounding interrelationships. Identified biological factors in PPP include local pathology [9], repeated surgical procedures [10], neuromusculoskeletal contributions [11] and central and peripheral nervous system activation and sensitisation [12,13]. Central nervous system sensitisation is a substantive factor in persistent pain presentations and may lead to chronicity despite resolution of a primary pathology [2,10]. Psychological factors are reported to include high pain catastrophising; and past history of sexual assault; and social factors include a family history of chronic pain [11].

The biomedical approach aims to define underlying pathology in PPP and initiate specific medical or surgical treatment [14]. When there is no known aetiology, or continued PPP, symptomatic management via hormonal, centrally acting or analgesic medications; or interventions such as nerve blocks, or botulinum toxin injected to the pelvic floor may be offered [15–17]. A Cochrane review of 13 randomised controlled trials (RCTs) for non-surgical management of PPP found low evidence for any individual treatment, perhaps due to the multifactorial nature of PPP [18], and the need to treat the not only the biological, but the psychological and social components of pain.

Pelvic physiotherapists are well placed to provide effective treatment for PPP with World Physiotherapy’s education framework endorsing a client-centred biopsychosocial model for its 625,000 physiotherapists worldwide [19]. Despite anecdotal success, there is little high-quality research on pelvic physiotherapy for PPP and the need for further research into its effectiveness has been highlighted in the Australian clinical guidelines for the treatment of endometriosis [20]. A mainstay of pelvic physiotherapy treatment in PPP is to address the common co-existing condition of pelvic floor tension myalgia (PFTM), a state of pelvic muscle pain and increased pelvic muscle tone [21]. Previous observational and cross-sectional analytic studies suggest women with PPP are significantly more likely to present with increased pelvic muscle pain and tone compared to the general population (50–58% vs. 4–10%, p<.05) [22,23]. The aim of this study was to investigate the effectiveness of multimodal physiotherapy in improving PPP over 2 years in a cohort of women with identified PFTM.

Materials and methods

Design

The Bellberry Ltd. Ethics Committee approved this study (no.: 2015-03-165). All participants gave written informed consent before data collection began. Recruitment was undertaken at a private obstetric and gynaecological tertiary referral centre in Sydney, Australia. The centre specialises in the management of PPP and endometriosis through an interdisciplinary approach that includes pelvic floor physiotherapy, nutrition and naturopathy intervention and medical/surgical management including advanced laparoscopic excision surgery. Prospectively recruited participants underwent a complete history and examination by one of two consultant gynaecologists with expertise in managing PPP including minimally invasive surgery, medical management, and the identification of PFTM on physical examination [14–17,24]. The diagnosis of PFTM was confirmed by physiotherapists at the initial assessment. Follow-up data were collected at 3 months and 2 years from recruitment. It included changes to medications or surgery, and the AQol-6D and FSFI Questionnaires. Co-interventions were defined as any additional medical or surgical treatments during the study duration.

Participants

Participants assigned female at birth over the age of 18 with PPP and PFTM on examination were approached for inclusion in the study. The current terminology used in the assessment and diagnosis of pelvic muscle dysfunction is diverse and variably defined, and for this cohort we used PFTM to describe the disorder of increased pelvic floor muscle pain and tone as outlined in the 2021 ICS report on the terminology of pelvic floor muscle assessment [21]. Pregnancy, suspected or known malignancy, and patients scheduled for gynaecological surgery as a primary intervention at presentation were excluded. Patients already taking both hormonal and non-hormonal medications were not excluded, but the type, dosage and frequency of medications were recorded. Patients did not undertake any additional treatment in the interdisciplinary centre whilst undergoing their physiotherapy intervention. Both verbal and written information regarding the study and written consent prior to inclusion were mandatory.

Intervention

Participants underwent physiotherapy at the same private clinic with either of two physiotherapists experienced in treating PPP. Treatment involved one-on-one pain science education (PSE) within treatment sessions, a group of educational interventions to address the biology of pain through a biopsychosocial lens. The PSE delivered helped patients reconceptualise their pain as an output of the nervous system, not purely from tissue injury or damage, with a key message that their pain can be changed. All participants also received pelvic floor co-ordination and relaxation exercises. Some treatment programs also included transvaginal Thiele massage (17/34), electrical stimulation (7/34) and EMG biofeedback therapy (10/34). The techniques were adapted to each patient and evolved throughout the treatment sessions as they progressed. The number of visits was not predetermined and varied according to participant need.

Outcome measures

Primary outcome

Patient-reported severity of PPP rated on a Visual Analogue Scale (VAS) [25]. VAS scores 0–100 were recorded for dysmenorrhoea, dyspareunia, dyschezia, dysuria, low back pain and non-menstrual pelvic pain, where 0 is no pain and 100 is the worst pain imaginable.

Secondary outcomes

Assessment of Quality of Life (AQoL-6D). A multi-attribute utility health-related quality of life instrument, which displays appropriate levels of construct, concurrent and convergent validity [26]. The AQoL-6D provides a utility score which ranges from 1.00 (full health) to 0.00 (death-equivalent health states).

Female Sexual Function Index (FSFI) – A 19-item self-report inventory designed to assess female sexual function (score range 2–36) with a higher score indicating better function.

In addition, at the initial baselines assessment and all subsequent visits, the following pelvic examinations were undertaken:

• Modified Oxford scale of pelvic floor strength [27];

• Single digital assessment of pelvic floor tone which was described as either low, normal or high tone [21];

• Assessment of vaginal resting pressures using Peritron™ manometry [28];

• Single digital assessment of pelvic floor myalgia in the superficial and deep pelvic floor as well as the pelvic wall musculature [29].

Data analysis

Data were analysed using IBM SPSS Statistics 24 (IBM, Chicago, IL). Data were assessed for distribution using the Shapiro–Wilk test. VAS, AQoL-6D and FSFI differences between baseline and follow-up time points were assessed using Wilcoxon’s signed-rank tests or the Sign test reflecting its distribution. Sub-group analysis was conducted based on baseline reports of VAS pain, and co-interventions, using the Mann–Whitney U-test for between group comparisons. For calculation of VAS change in those with baseline reports of pain, not all participants completed both 3 month and 2 year follow-up. As such, pairwise deletions were utilised to mitigate loss of data.

Results

Between March and September 2015, 42 patients consented to the study and completed screening and baseline questionnaires. Thirty-nine proceeded to physiotherapy but five withdrew, citing no change in pain after a single physiotherapy session, leaving 34 who completed treatment. Four patients were completely lost to follow-up, with the remaining 30 completing at least one follow-up questionnaire.

The demographics of the consented and final evaluated population are below (Table 1). A sensitivity analysis reported no significant differences of demographic or functional baseline attributes for participants commencing the study and those completing the study.

Table 1. Participant demographics.

Demographics	Consented, n = 42	Included, n = 34	Consented vs. included, p valuesa
Median (range)
Age	30.5 (22–45)	29.5 (22–44)	.441
Body mass index	22.5 (17.6–42.6)	22.6 (17.6–42.6)	.208
Laparoscopies for endometriosis	1 (0–10)	1 (0–10)	.729
Total gynaecologic surgeries	2 (0–10)	2 (0–10)	.843
Number (%)
Country of birth			.424
Australia	31 (74)	26 (77)
Other	11 (26)	8 (24)
Relationship status			.836
Single	7 (17)	5 (15)
In a relationship	34 (81)	28 (82)
Other	1 (2)	1 (3)
Women who had vaginal deliveries	7	5	.474

^a Mann–Whitney’s U-test.

Participants had a median of 1 medical and 2 surgical interventions in the past for pelvic pain and are described in Table 2. 22/34 (65%) had had both medical and surgical interventions previously and 6/34 (18%) had used more than one hormonal medication. 13/34 (38%) had two or more surgeries. 27/34 (79%) patients were diagnosed with endometriosis, 4/34 (12%) had fibroids or polyps and 3/34 (9%) had no known medical comorbidities. 22/34 (65%) were taking medications at time of recruitment related to their pelvic pain. 7/34 (21%) had a combination of hormonal, anti-depressant and prescription analgesia whilst the remaining were limited to one medication type (Table 2).

Table 2. Previous medical and surgical interventions.

Intervention	Participants having tried, but ceasing medications prior to baseline	Participants on these medications at baseline
Medical
Hormonal
Oral contraceptive pill	19	8
LNG-IUDa	9	7
Etonogestrel depot	2	0
Oral progestins	3	2
Total	26	16
Centrally acting agents
Tricyclic	5	2
SNRIb	0	3
SSRIc	0	3
Pregabalin	2	2
Total	6	7
Analgesics
Opioids	0	2
OTC (NSAIDs/paracetamol)d	0	6
Total	0	7
Interventional
Laparoscopy for diagnosis/ excision of endometriosis		22
Hysterectomy		3
Injection of Botox™ to the pelvic floor		1
Total		22

^a Levonorgestrel intrauterine device.

^b Serotonin and norepinephrine reuptake inhibitors.

^c Selective serotonin reuptake inhibitors.

^d Over the counter drugs (nonsteroidal anti-inflammatory drugs/paracetamol).

Participants attended a median of 4 (range 1–11) physiotherapy sessions and the median treatment duration was 7 weeks (range 1–16 weeks). There were no significant changes in the Oxford pelvic floor strength grade from beginning to end of physical therapy (median 3 vs. 3, p=.655), or in participant resting tone. Resting pressure on vaginal manometry was similarly unchanged (55 vs. 53 cm H₂O, p=.099). There were no significant changes in the FSFI index throughout the study period (24.5 vs. 25.4 p=.568). The utility baseline score of the AQoL-6D showed a non-significant improvement at 3 months (0.65 vs. 0.67, p=.256) and reached significant improvement at 2 years (0.65 vs. 0.73, p=.011) for the 24/34 respondents. Figure 1 selects for those with baselines reports of pain that had at least one other completed follow-up and demonstrates a significant decrease in VAS scores at 3 months for dysmenorrhoea (72.5 vs. 52, p=.015), low back pain (55 vs. 47.5, p=.004), dyschezia (45 vs. 30.5, p=.031), abdominal pain (52 vs. 36.5, p=.049) and non-menstrual pain (53 vs. 41, p=.018) with a non-significant decrease in dyspareunia (57 vs. 50, p=.365). At 2 years, improvements in dysmenorrhoea persisted (72.5 vs. 60, (p=.002) and dyspareunia was significantly improved (57 vs. 36, p=.033).

Figure 1. VAS changes in women with reported baseline pain with at least one other completed follow-up.

A box and whisker plot graph outlining the mean VAS pain changes in participants with reported baseline pain in each pain parameter with at least one follow-up. The box and whisker plot graph describe the VAS pain scale changes from baseline to 3 month and 2 year follow-up. The X-axis shows the VAS score, and the Y-axis shows the types of pain which are dysmenorrhoea, dyspareunia, dyschezia, dysuria, low back pain, abdominal pain and non-menstrual pelvic pain. The graph demonstrates a significant reduction in dysmenorrhoea, dyschezia, low back pain and non-menstrual pelvic pain at 3 months, and a significant reduction in dyspareunia and dysmenorrhoea at 2 years.

Discussion

This prospective study provides evidence that physiotherapy as a treatment may improve pain symptoms for some women with PPP. The minimal clinically important difference (MCID) reported in VAS for endometriosis associated pelvic pain is 10 mm [30]. Based on these data, clinically significant improvements are reported at 3 months for dysmenorrhoea (18.5 mm), low back pain (14.5 mm), dyschezia (12.5 mm) and non-menstrual pain (10 mm). The improvement in dyspareunia (7 mm) at 3 months was neither statistically nor clinically significant. Significant clinical and statistical improvement in dysmenorrhoea (10.5 mm) was maintained at 2 years, with dyspareunia becoming significant on both clinical and statistical definitions (14 mm).

This study demonstrated a significant improvement in quality-of-life scores at 2 years at 0.08 above what is considered MCID of 0.06 [31]. However, this must be interpreted with some caution, since the decreased number of respondents due to low to follow-up may lead to selection bias, and this may not be clinically significant. It was determined that patients having co-interventions in this study had worse quality of life at study completion. Participants with a higher number of co-interventions reported the poorest response to physical therapy and the burden of failed physical therapy should be considered as a contributor to their worsening of quality-of-life scores. Conversely, those with identified PFTM and physical therapy with no additional co-intervention experienced better outcomes, potentially due to the neuromuscular symptoms being the predominant source of their pain.

In PPP, patients often experience multiple pain symptoms and the burden of prior failed interventions. In this study, patients had a median of 2 surgical interventions, and 21% of participants were already on a combination of three medications prior to attending physiotherapy. This reflects the clinical caseload in pelvic physiotherapy, where referrals are often made as a last line intervention for presentations of PPP, rather than as first line. It is unclear based on these data if medical or surgical interventions may be reduced by physiotherapy earlier in the clinical setting, or if PFTM occurs following these interventions and this is an area for future research.

During study protocol development, we considered vaginal manometry as an objective assessment of treatment response. However, with many participants having physiotherapy sessions close together, the resulting manometry measurements are often only one to two weeks apart, limiting discriminatory capacity. Given that manometry is not a participant reported outcome measure, and the limitations of this assessment for pain, we do not recommend this as a tool for routine assessment and measurement of response.

The entry criteria for participants in this study were the subjective determination of PFTM [21] which was made by two experienced gynaecologists with extensive previous research in the field of gynaecological pain and pelvic floor dysfunction [15–17,24]. Participants in this study underwent a multimodal treatment program addressing both the muscular and nervous system contributions to pain perception. Observational studies report women with PPP are significantly more likely to present with pelvic floor muscle tenderness [22,23], and pelvic muscle tenderness on digital palpation has confirmed convergent validity with central sensitisation index scores, suggesting central pain mechanisms [32]. In addition, the presence of myofascial symptoms (including increased muscle tone and pelvic muscle tenderness) is documented in PPP and found to be an independent factor associated with PPP severity [11]. This suggests a role for the pelvic neuromuscular structures in the pathogenesis or perpetuation of PPP.

Systematic reviews of individual physiotherapy interventions for women with PPP report that methodological flaws and small sample sizes make it difficult to draw conclusions of the effectiveness of these interventions [33,34]. In addition, studies included in the systematic reviews often have a biomedical ‘tissue based’ approach to treatment. The physiotherapy intervention in this study had a strong PSE focus. There is level 1 evidence that PSE alone can improve health parameters in patients with persistent musculoskeletal pain conditions [35] and it is a mainstay of modern pelvic physiotherapy intervention. This multimodal program also included pelvic muscle co-ordination and relaxation exercises as well as desensitisation treatments. Whilst standardising treatment to a single physiotherapy modality would improve internal validity and reproducibility, this does not reflect clinical practice. This pragmatic study utilised an individualised and multimodal approach to address the patient’s symptoms, and whilst an overall improvement in patient symptoms was demonstrated, the contributing treatment that led to that improvement is not clear.

Limitations

It is acknowledged that treatment compliance is a determinant of success, and 5/42 (12%) women withdrew from the study before 2 year follow-up, citing no change in pain with physical therapy as the predominant reason for withdrawal. Other factors for withdrawal may have been the out-of-pocket charge to patients in the private health-care setting. Four women were completely lost to follow-up during the study period, with the remaining 30 completing at least one follow-up questionnaire. 22/34 (62%) women completed follow-up at three months and 24/34 (71%) completed follow-up at 2 years. 19/34 (55%) women completed follow-up at both time points. This response rate is comparable to other prospective observational cohorts for PPP in women (response rates of 37.5–67.5%) [36].

The small sample size with loss of power when stratifying based on compliance or co-intervention was a limitation of this study and these small group sizes also limited power for secondary analysis. Further, due to the uncontrolled study design and the multimodality of the intervention including additional treatments of medications and prior surgery, all measured changes in the outcomes in the short- or long-term cannot be attributed to physiotherapy alone.

However, follow-up to 2 years provides detail to the possible outcomes following the addition of physiotherapy treatment in PPP and indicates that improvements from treatment may be maintained over the longer term, particularly with dysmenorrhoea and dyspareunia.

PPP is complex condition with multiple aetiologies and an interrelationship of biological, psychological and social factors. This study provides early evidence that physiotherapy may improve some pain outcomes for those with PPP. To be confirmed as an effective treatment in PPP, physiotherapy requires high-quality, appropriately powered trials to identify the effective individual treatments to be included in a multimodal program, and to determine the characteristics of patient presentation that are amendable to this intervention.

Acknowledgements

The authors wish to acknowledge and thank the participants for their time and to Alana Healthcare for their support throughout the study in providing space and infrastructure. No payment was made to participants and no financial support was received from Alana Healthcare for this study.

Disclosure statement

No author has any conflict of interest relating to this manuscript. This research did not receive any specific grant from agencies in the public, commercial or not for profit sectors.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.