Pelvic Floor Hypertonicity in Women With Pelvic Floor Disorders: A Case Control And Risk Prediction Study

Sabourin J C1, Cameron B2, Sanaee M S3, Koenig N A4, Lee T5, Geoffrion R4

Research Type

Clinical

Abstract Category

Pelvic Pain Syndromes / Sexual Dysfunction

Abstract 276
Female Incontinence
Scientific Podium Short Oral Session 16
Thursday 30th August 2018
09:05 - 09:12
Hall C
Pain, Pelvic/Perineal Pelvic Floor Prevention Retrospective Study Surgery
1. Providence, 2. Faculty of Medicine,, University of British Columbia, Vancouver, British Columbia, Canada, 3. De, 4. Department of Obstetrics and Gynecology, University of British Columbia, Vancouver, British Columbia, Canada, 5. Centre for Health Evaluation and Outcome Sciences, University of British Columbia, Vancouver, British Columbia, Canada
Presenter
J

Johanne C Sabourin

Links

Abstract

Hypothesis / aims of study
Myofascial pelvic pain is a chronic and debilitating condition, sometimes associated with pelvic floor disorder (PVD). Our aim was to identify risks factors in women with PVD and myofascial pain, compared to controls without pain.
Study design, materials and methods
This was a case control study (2009-2017) of patients with PFD (POP, UI, or both), and/or who underwent surgery for PVD, and had a diagnosis of hypercontracted pelvic floor. Cases were matched with patients who presented with the same PFD, and/or underwent surgery for the same PFD, but without pelvic floor hypertonicity.  Risk factors were compared between groups. 
The relationship between patient variables and the presence of hypercontracted pelvic floor in cases vs controls was examined via univariate logistic regression. Firth's penalized likelihood approach was employed to account for low outcome rate for some variable levels. Multivariate logistic regression was then used to examine the impact of these variables on myofascial pain adjusted for each other. Variables with p<0.05 in the univariate analysis and/or variables believed to be clinically relevant were included in the multivariate model. We further used stepwise regression with Akaike information criterion to retain variables that were most predictive of postoperative myofascial pain among patients underwent urogynecologic surgery. The estimated regression coefficients from the resulting model were then converted into a scoring algorithm to predict pain. Area under the receiver operating characteristic (ROC) curve was used to assess the accuracy of the prediction model.  Among women with pelvic floor hypercontraction, difference in demographic variables between those with urogynecologic surgery and those without was assessed using Chi-square test, Fisher's exact test or t-test as appropriate.
Results
Ninety-five cases with pelvic floor muscle hypertonicity were identified and matched. Sixty-five women suffered from both POP with or without UI and 30 women had UI only. Women sought care for a variety of pelvic floor symptoms. Among cases, 80/95 women (84.2%) presented with symptoms of pelvic pain. Forty-three percent of women with pain had pain alone whereas 57% had pain plus one or more pelvic floor symptoms (in decreasing order of frequency: urinary, bowel and/or prolapse complaints). Fifteen women (15.8%) had hypertonicity without a primary complaint of pelvic pain and instead presented with initial symptoms, in decreasing order of frequency, of urinary, bowel and/or prolapse complaints. 

Most women were post-menopausal. Case patients were younger than controls (54 versus 59, p=0.002). On pelvic examination, vaginal atrophy was noted in 42.1% of cases and 45.7% of controls (p=0.616). Among cases, 15 women (15.8%) had provoked vestibulodynia, 81 (85.3%) had levator muscle spasm, 73 (76.8%) had obturator muscle spasm, 40 (42.1%) had coccygeus muscle spasm, 6 (6.3%) had bladder base tenderness and no women had documented increased anal sphincter muscle tone (voluntary or involuntary). Two women (2.1%) had exposure from prior prolapse procedures with synthetic mesh. Eleven women (11.6%) had synthetic suburethral mesh exposure from prior midurethral sling for incontinence. 
Overall, univariate analysis showed being younger, having a history of depression, endometriosis, chronic constipation, dyspareunia, fibromyalgia, irritable bowel syndrome, musculoskeletal spine injury (from fall, fracture or motor vehicle accident), pelvic injury, chronic back pain, appendectomy, laparoscopy, hernia repair, back surgery and transobturator midurethral sling for UI to be significantly associated with hypercontracted pelvic floor in cases vs controls (Table 1). Multivariate analysis retained risk factors of decrease in age, history of depression, musculoskeletal spine injury (from fall, fracture or motor vehicle accident) and surgery for UI using a transobturator midurethral sling. Surgery for UI using a retropubic midurethral sling was protective against pelvic floor hypercontraction (Table 1). 
Seventy-one percent (n=67) had urogynecologic surgery as a likely trigger for hypercontracted pelvic floor. All presented with subjective complaint of pelvic pain on initial visit. In the postoperative urogynecology surgery cases, age was not significantly different between cases and controls. Univariate analysis showed a history of depression, chronic constipation, dyspareunia, fibromyalgia, irritable bowel syndrome, musculoskeletal spine injury (from fall, fracture or motor vehicle accident), laparoscopy, back surgery and transobturator midurethral sling for UI to be significantly associated with hypercontracted pelvic floor in cases vs controls (Table 1). Retropubic midurethral sling was protective (p<0.001). In contrast, only the transobturator type of midurethral sling for UI remained a significant risk factor in the multivariate analysis.

Within the group of cases with pelvic floor hypercontraction (n=95), women who did not have urogynecologic surgery as a trigger (n=28) had fewer vaginal deliveries (mean 1.8 vs 2.3, p=0.043), a higher prevalence of interstitial cystitis (14.3 vs 1.5%, p=0.025), more chronic back pain (28.6 vs 9%, p=0.014) and a lesser prevalence of cholecystectomy (7.1 vs 28.4%, p=0.023). 

The clinical predictive model for myofascial pain after urogynecologic surgery exhibited excellent predictive accuracy as reflected by the large area under the ROC curve (0.87; 95% CI: 0.80, 0.93).  The resulting scoring algorithm from the model has a base score of 4 and assigned probability scores to demographic variables of depression (+3), endometriosis (+7), irritable bowel syndrome (+5) and musculoskeletal spine injury (+5). UI surgery using transobturator midurethral sling was also assigned a probability score of +7. In contrast, UI surgery using retropubic midurethral sling was assigned a protective probability score of minus 3.  A total score of 7 or higher translated to an estimated probability of over 50% for persistent postoperative pelvic pain. The scoring algorithm showed good agreement between the observed and estimated probabilities (Table 2).
Interpretation of results
To our knowledge, our study is the first to evaluate risk factors for chronic pelvic pain after urogynecologic surgery, and to characterize a patient group with pain secondary to hypercontracted pelvic floor muscles in particular. This case control study of patients with pelvic floor disorder symptoms  showed younger patients, with a history of depression, musculoskeletal spine injury and surgery for urinary incontinence using a transobturator midurethral sling to be significantly more at risk for pelvic floor hypertonicity. For example, a patient with a history of depression, endometriosis, irritable bowel and a musculoskeletal back injury from a fall has over 90% probability of persistent postoperative pelvic floor hypercontracted muscles. If hypercontracted pelvic floor is detected at initial assessment, prior to PFD treatment, and the patient desires a midurethral sling, she may choose to delay surgery while undergoing treatments such as pain education, pelvic floor physiotherapy, medications or trigger point injections [1]. The clinicians might attempt to optimize intervention to avoid worsening of pre-existing conditions, or the development/exacerbation of central sensitization with local anesthetic agents and/or neuropathic medications [1]. 
 Weaknesses include the retrospective nature of our research.  A further limitation of our study is the variety of different operations undergone by our patients, different surgeons with variable expertise, tissue handling and surgical techniques and materials. Finally, our data collection did not include other factors thought to predispose women to persistent pain, such as genetic susceptibility, preceding experiences of pain in response to other noxious stimuli or detailed psychosocial factors [2].
Concluding message
A clinically useful prediction model was developed to direct gynecologic surgeons in patient counselling. Larger prospective study is needed to better characterize the cause and effect relationship between the risks factors and the onset of pelvic floor hypertonicity. For example, risk factors such as back injury, endometriosis, irritable bowel syndrome, spinal injury can contribute and/or are often associated with sensitization of the central nervous system which surgery fibrotic tissue might further exacerbate.
Figure 1
Figure 2
References
  1. Butrick, C. Persistent Postoperative Pain: Pathophysiology, Risk Factors, and Prevention. Female Pelvic Medicine & Reconstructive Surgery 22(5), September/October 2016, p 390-396
  2. Kehlet H, Jensen TS, Woolf CJ. Persistent postsurgical pain: Risk factors and prevention. Lancet 2006; 367:1618-25
Disclosures
<span class="text-strong">Funding</span> R Ge <span class="text-strong">Clinical Trial</span> Yes <span class="text-strong">Public Registry</span> No <span class="text-strong">RCT</span> No <span class="text-strong">Subjects</span> Human <span class="text-strong">Ethics Committee</span> Providence Health Care Research Ethics Board <span class="text-strong">Helsinki</span> Yes <span class="text-strong">Informed Consent</span> Yes