Howard I. Glazer, Ph.D., Marek Jantos, M.A., Elizabeth Heaton Hartmann, P.T., and Charles Swencionis, Ph.D.

OBJECTIVE: To compare surface electromyographic (EMG) studies of the pelvic floor in women diagnosed with dysesthetic vulvodynia to those of women with no urologic or gynecologic symptoms.

STUDY DESIGN: Fifty women were chosen to participate and placed in one of two diagnostic categories, asymptomatic (no report of urogynecologic abnormalities, n=25) and those diagnosed with dysesthetic vulvodynia (n=25). Testing was completed utilizing electromyographic equipment; an inserted, single-user vaginal sensor; and the Glazer protocol. SEMG variables compared were pretest and posttest resting amplitudes, contractile amplitudes, contractile and resting stability, recruitment latency and recovery, and muscle contraction spectral analysis.

RESULTS: The most reliable predictors of symptomatic women were pelvic floor contractile amplitudes of the tonic, phasic and endurance contractions, though 9 of the 15 variables tested proved significant. The phasic (three second) contractions of the symptomatic group were 46% less than in the pain-free group. Tonic (12-second) contractions were 49% less, and endurance (60-second) contractions proved to be only 47% of those produced by those with no dysfunction.

CONCLUSION: The results of EMG studies of the pelvic floor in women diagnosed with dysesthetic vulvodynia proved significantly different from those of their urogynecologically asymptomatic cohorts. Physiology of the pelvic floor is an essential piece of knowledge needed to further study the etiology and causative factors in dysesthetic vulvodynia. Though the sample size used in this study was not sufficient to quantify normal pelvic floor function, the study certainly suggests sufficient significant differences between the two groups to merit further study. ( J Reprod Med 1998; 43:959-962)

Keywords: vulvar diseases, pelvic floor, electromyography, dysesthetic vulvodynia.

Introduction
Several studies have been published discussing the involvement of the pelvic floor muscles in patients diagnosed with dysesthetic vulvodynia. Those electromyographic (EMG) studies have defined abnormal pelvic floor function as an inability to fully relax at rest, an inability to fully contract on command and instability in both those states. This study compared pelvic floor surface (SEMG) readings in patients diagnosed with dysesthetic vulvodynia to those of matched, asymptomatic controls.

Proprioceptive retraining provides individuals with information on pelvic floor and visceral functions of which they are not normally aware.(1) In the asymptomatic population, when this coordination has not been disrupted, normal, pain-free function is achieved and maintained. Within the population diagnosed with dysesthetic vulvodynia, this coordination has been lost, resulting in abnormal function, often with reports of coinciding irritable bowel syndrome, interstitial cystitis, fibromyalgia and low back pain, among others.(2) As shown by Glazer et al, (3) abnormal function includes elevated resting baselines, instability at rest and with contraction, and decreased contractile amplitude when compared to the same sample following pelvic floor retraining and symptom decline. White et al, (4) suggested that asymptomatic values would be as follows: resting baseline < 2.0 uV, contractile potential > 17 uV, resting SD >.20, recruitment contractile recovery < 0.2 seconds and spectral frequency > 115 Hz. While these values are extrapolated from symptomatic SEMG values, the data gained from each group have not been verified through stringent statistical analysis.

The purpose of this study was to compare EMG findings in symptomatic and asymptomatic samples, utilizing protocols and analysis as described by Glazer. (2) Pelvic floor muscle variables compared were pretest and posttest resting amplitudes (uV); phasic, tonic and endurance contractile amplitudes (uV); contractile and resting stability (SD, uV); recruitment latency and recovery (seconds); and power spectral analysis of muscle contraction (Hz).

Method
Fifty women were chosen to participate and were placed in one of two diagnostic categories, those who were asymptomatic (no report of urogynecologic abnormalities, n = 25) and those diagnosed with dysesthetic vulvodynia (n = 25). The asymptomatic group was solicited via advertisement and selected as a result of a personal questionnaire review. The symptomatic group had been diagnosed with dysesthetic vulvodynia, with physicians utilizing diagnostic criteria delineated and, described in 1993. (5) All subjects were instructed equally on proper pelvic floor contractions. Testing was completed utilizing EMG equipment and the Glazer protocol (2) and an inserted, single-user vaginal sensor. The Glazer protocol, as used to assess pelvic floor function, utilizes a five-segment sequence of varied pelvic floor contractile states. It begins with a one-minute, pretesting baseline assessment, evaluating the resting amplitude of the pelvic floor prior to testing. The second phase includes five phasic (quick) contractions, each separated by a 10second rest period. Following are five tonic (10second) contractions, with 10-second rest periods separating them. The fourth testing cycle includes a single, maximum, endurance pelvic floor contraction of 60 seconds. The assessment protocol ends with another 60-second, posttest resting baseline, measuring resting amplitude following the test maneuver. 

Results
The data were collected, and MANOVA indicated overall significance (Hotelling's T(2), 137.6, P <.001), allowing paired t testing and discriminant function analysis. Nine of the 15 variables tested proved significant. Table I lists the variables that best discriminated the two groups, ranked in descending order of correlation with the predicting function, corresponding well with the degree of significance when compared to the t test results.

The most reliable predictors separating the symptomatic group from the asymptomatic group were the contractile amplitude of the tonic, phasic and

Table I Variables, Ordered by Size of Correlation Within Function
(as Correlated with t Test Results)

Table II Variables Best Discriminating Symptomatic (n=25) vs. Asymptomatic Women(n=25)

endurance contractions. The amplitudes of tonic contractions, when compared, were significant (t = 5.88, P <.001), showing the asymptomatic group able to produce much higher contractile amplitude (mean =25.8 vs. 13.1 uV) with a 12-second contraction as compared to the symptomatic sample. The quick, three-second phasic contraction of the pain-free group also showed significance over the pain group (t = 5.43, P <.001), with those with no reported dysfunction contracting to a significantly higher amplitude (mean= 26.8 vs. 14.4 uV). The ability to hold a 60-second endurance contraction was significant as well (t = 5.14, P<.001), with the asymptomatic group's ability to contract to a higher amplitude greater than that of the symptomatic cohort (mean = 23.7 vs. 12.6 uV). Other variables also were significant for differentiation of the two groups:

· Age (t = 3.09, P <.003), with the asymptomatic group older than the symptomatic group (mean = 34.9 vs. 27.3 years old).

· Pretest baseline (t=-2.92, P<.006), showing the asymptomatic group better able to relax with a lower resting amplitude prior to testing as compared to the symptomatic group (mean = 2.0 vs. 3.3 gv).

· Births for the asymptomatic sample, significantly higher than in the symptomatic group (t = 2.83, P <. 007), with those having no symptoms reporting a higher birth rate (mean= 1.2 vs. 0.3 births).

· Pretest baseline SD significantly predicting the asymptomatic vs. the symptomatic (t = - 2.47, P<.018), with those reporting no dysfunction showing a more stable resting baseline (pretest), as demonstrated by a smaller measure of SD of the amplitude (mean =.450 vs. .866).

· Spectral analysis of the muscular contraction, significant according to the t test (t = - 2.3, P <.026) and showing a lower measure in those with no symptoms (mean = 108.0 vs. 116.8 Hz).

· Stability of the pelvic floor muscle at rest (measured as SD) showing significance for the asymptomatic population over the symptomatic one (t = - 2.08, P <.044), with the asymptomatic group able to return to a more stable and relaxed state following a 12-second contraction (mean 1.025 vs. 1.380) (Table II).

The remaining variables assessed, including latency and recovery, showed decreasing significance according to t test results and are not reported on below.

Discussion

When reviewing EMG pelvic floor evaluations, contractile abilities (phasic, tonic and endurance contractions) are the best differentiators of patients

Table III Summary of Varying Performance and Demographics, Symptomatic vs. Asymptomatic

with dysesthetic vulvodynia as compared to matched controls. Phasic contractions (three seconds) for the symptomatic group were 46% lower than in the asymptomatic sample. Tonic contractions (12 seconds) in the pain group, when compared to those with no complaints, were measured at 49% less. Endurance contractions (60 seconds) in the dysesthetic vulvodynia group proved to be only 47% of those produced by women with no dysfunction (Table III). Those with dysesthetic vulvodynia proved to be statistically younger and of lower parity than the comparison group. These two variables are not what would have been predicted. Rather, it would have been expected that those who were either older or who had had multiple births would have shown lower contractile amplitude abilities over their younger and less parous cohorts. Both pretest resting baseline amplitude and pretest stability significantly differentiated the symptomatic from the asymptomatic, with the pain group 32% higher at rest than the pain-free group and 49% more unstable. In contrast to previous studies, spectral analysis of the pain group was actually 8% higher than the pain-free sample. The stability of the muscle following the tonic 12-second contraction was statistically greater in the asymptomatic group, whereas instability was 27% greater in the symptomatic cohort.

Conclusion
Pelvic floor performance in those diagnosed with dysesthetic vulvodynia is significantly lower in contractile and resting ability, contractile and resting stability, and efficiency of contraction when compared to those with no symptoms. Though the sample size used for this study was insufficient to suggest normal pelvic floor EMG values, the data indicate significant differences when comparing the two populations. Since a number of studies have shown success in treating women with dysesthetic vulvodynia using the protocol described, further, in-depth studies focusing on pelvic floor function in both populations are warranted in an attempt to better understand the significance of the role of abnormal pelvic floor function in vulvar pain syndromes.

References

1. Markwell S, Murname G: A Role for Physiotherapy in Perianal and Perineal Pain Syndromes. Philadelphia, WB Saunders, 1990

2. National Vulvodynia Association: Epidemiological study, 1996, unpublished

3. Glazer HI, Rodke G, Swencionis C, et al: Treatment of vulvar vestibulitis syndrome with electromyographic biofeedback of pelvic floor musculature. J Reprod Med 1995;4:283-290

4. White G, Jantos M, Glazer H: Establishing the diagnosis of vulvar vestibulitis. J Reprod Med 1997;3:157-160

5. McKay M: Dysesthetic ("essential") vulvodynia: Treatment with amitriptyline. J Reprod Med 1993;38:9-13