The protocol for this study was approved by the institutional review boards of Drexel University College of Medicine and Georgetown University. This study was not registered as a clinical trial because it did not meet the federal definition of a clinical trial. Women were recruited for participation in the study using flyers posted on college campuses and the National Vulvodynia Association website. Additionally, women with clinically diagnosed PLV by a vulvar specialist at the Pelvic and Sexual Health Institute in Philadelphia, Pennsylvania, and the Center for Vulvovaginal Diseases in Washington, DC, were approached to consider enrollment in the study. Inclusion criteria for study groups were age of 18 to 55 years and a clinically confirmed diagnosis of PLV (PLV group) or without a diagnosis of PLV and self-reported lack of vulvar pain or other chronic pain conditions (control group). Exclusion criteria were current vulvovaginal infections, current pregnancy, postmenopausal status, history of vaginal surgery for prolapse repair or vulvar surgery, vaginismus, or major medical or psychiatric illness. Women with PLV with comorbid widespread pain conditions (fibromyalgia and chronic regional pain syndrome) were excluded. 

After written informed consent, participants completed a semistandardized questionnaire (patient health questionnaire 9 [PHQ-9]) that included demographic questions, clinical characteristics, and visual analogue scales on a 100-mm ruler for genital pain with sexual intercourse and with daily activity. Participants underwent quantitative sensory testing (QST) after completion of the PHQ-9. For participants with regular menstrual cycles, QST was scheduled between days 7 and 12 of the follicular phase of the cycle. For participants taking oral contraceptives, QST was scheduled between day 7 and 12 after the first placebo pill. We selected a sample size similar to the previous study⁸ of vulvar cold pain thresholds for comparative purposes. All participants received $50.00 for their participation. 

Detection and pain thresholds to thermal stimuli (cold and heat) were assessed using a thermosensory analyzer equipped with a 5 × 5-mm² contact thermode. A temperature range of 0°C to 50°C was used for measurements of thermal detection and pain thresholds. The method of limits was used, in which the intensity of the applied thermal stimulus increased or decreased until the participant stopped the stimulus with a handheld mouse when the stimulus was first perceived (detection) or when the stimulus was perceived as painful (experimental pain threshold). The QST was performed at the left volar forearm as a measurement of somatosensory processing at a site peripheral to the vulva. The QST was then performed in the posterior vestibule (left and right sides). We selected the posterior vestibule as the measurement site, as it was demonstrated that the posterior vestibule was the most sensitive site in patients with PLV,⁸ in agreement with clinical reports of the posterior vestibule being the most common site of pain.¹⁵ Pain thresholds to mechanical stimuli were assessed using a commercially available algometer equipped with either a blunt tip for forearm measurements or with a cotton-tipped applicator for vestibular measurements. A pressure range of 0 to 4 kg and 0 to 500 g was used for measurements of forearm and vestibular pressure pain thresholds, respectively. 

The cutoff values (0°C for cold pain, 50°C for heat pain, and 500 g and 4 kg for vestibular and forearm pressure pain, respectively) were assigned as the pain threshold for participants who did not report pain within the range of applied stimulus. The starting temperature for thermal measurements was 32°C, with a rate of change of 1°C per second. Thermal stimuli were applied in the following fixed order: cold detection (3 repetitions), warmth detection (3 repetitions), cold pain (2 repetitions), and heat pain (2 repetitions). Participants were asked to rate their pain intensity on a scale of 0 (no pain at all) to 10 (worst pain imaginable) using a verbal numerical rating scale after each pain measurement. 

Data are presented as the means and SEs of the means for continuous variables and the frequencies and percentages of categorical variables. For continuous variables, differences between the averages were tested using t tests when normality assumptions were satisfied. Differences between the left and right vestibule were compared using paired t tests, and differences between the PLV group and the control group were compared using unpaired t tests. The nonparametric Wilcoxon rank sum test (unpaired comparisons) and Wilcoxon matched pair test (paired comparisons) were used to examine differences if normality assumptions were not satisfied. For categorical variables, χ² and Fisher exact tests were used to examine differences between 2 groups. 

Seventeen participants were included in the PLV group and 16 in the control group (N=33; Table 1). Age, age at menarche, body mass index, parity, hormonal contraceptive use, handedness, marital status, education, and smoking status were not significantly different between the PLV group and the control group. The majority of participants were white (16 of 17 [94%] in the PLV group and 12 of 16 [75%] in the control group). The PHQ-9 score was lower in the PLV group (P<.05), with 1 participant scoring greater than or equal to 15, which suggested the presence of major depression. Self-reports (visual analogue scale) of genital pain with sexual activity (P<.001) and daily activity (P<.05) were significantly higher in the PLV group compared with the control group. Five participants in the PLV group reported the presence of an additional chronic pain condition (Table 1). 

No differences in thermal detection thresholds were recorded between the right and left vestibule in the PLV and control group. Vestibular and forearm thermal detection thresholds were also similar between the PLV and control groups (data not shown). 

Although the forearm pressure pain threshold tended to be lower in the PLV group, this difference was not significant (Figure 1A). Left forearm cold (P<.01) and heat (P<.01) pain thresholds were significantly lower in the PLV group compared with the control group (Figure 1B and 1C). No significant differences were found between the PLV and control groups’ pain ratings (verbal numerical rating scale) regarding the forearm pain thresholds to any stimuli applied (Table 2). 

Thermal and mechanical pain thresholds and ratings of pain (verbal numerical rating scale) were similar between the right and left vestibule in the PLV and control groups. Pain thresholds to cold were significantly lower at the right (P<.01) and left (P<.05) posterior vestibule in the PLV group compared with the respective side in the control group (Figure 2C). Heat pain thresholds in the PLV group were significantly lower on the right (P<.05) compared with the respective side of the control group (Figure 2B). Pressure pain thresholds were approximately 2-fold lower on the right (P<.01) and left (P<.001) posterior vestibule of the PLV compared with the control group (Figure 2A). Ratings of evoked pain were similar between groups (Table 2). 

The major finding of this study was 2-fold lower vestibular cold pain thresholds (cold pain reported at higher temperatures) in participants with PLV compared with control participants. To date, little attention has been paid to cold pain thresholds in this patient population despite numerous reports of decreased cold pain thresholds in QST studies of other types of chronic pain.^16-21 A 2001 study reported qualitative differences in vestibular cold pain thresholds in patients with PLV compared with controls.⁸ Subsequent QST studies of patients with PLV did not include vestibular pain thresholds to cold stimuli in the QST protocol.^5-7,9,10 In the study by Bohm-Starke et al,⁸ the majority of participants did not report cold pain in the posterior vestibule within the range of testing temperatures (31°C-6°C), which limited the ability to identify differences in cold pain thresholds. We therefore used a testing range of 32°C to 0°C, and 2 control participants did not report pain at 0°C. Our finding in the current study corroborates the qualitative report of Bohm-Starke et al.⁸ Together, these 2 studies suggest that cold allodynia is a component of the somatosensory profile of some patients with PLV and may provide insight into distinct underlying pathologic mechanisms in this subset of patients. Studies^18,22,23 of other chronic pain conditions that used a QST protocol subgrouped patients according to cold pain thresholds to initiate a mechanistically based classification scheme emphasizing the importance of the current findings; however, much work remains to be done to reach the goal of treatment targeted to the pathophysiologic mechanisms underlying an individual patient's pain. Large-scale, comprehensive QST studies are needed to identify subgroups of patients with vulvodynia based on somatosensory profiles and potentially different mechanisms of pain pathogenesis—a necessary step for the refinement of standard of care algorithms for vulvodynia. 

Self-reported ratings of genital pain with sexual and daily activity were higher in patients with PLV compared with controls (Table 1), highlighting the effects of PLV on overall quality of life. Although pain ratings of pressure pain thresholds were not different between groups, the threshold at which pain was reported was lower in patients with PLV (Figure 2A) than in controls. In contrast to studies using homemade vulvar algesiometers,^6,7,10,12,24 we used a commercially available algometer with a customized attachment for a cotton-tipped applicator and replicated previous findings of lower pressure pain thresholds in patients with PLV. This novel device may be useful for multicenter site studies, which require standard operating procedures across institutions, although future studies of inter- and intrarater reliability are needed. These algometers may also be useful as a tool to assess response to treatment in a clinical setting. 

We reported lower vestibular heat pain thresholds in the right vestibule of patients with PLV compared with controls (Figure 2B). Other studies^6-8 have also reported lower vestibular heat pain thresholds in patients with PLV. 

Our report of normal vestibular thermal sensory detection thresholds combined with reduced thermal (cold and heat) and mechanical pain thresholds provides further evidence that central sensitization contributes to the chronic vulvar pain of some patients with PLV. Although we did not evaluate pelvic floor function in this study, it is important to note that pelvic floor hypertonicity may also contribute to PLV pain. Therefore, it is recommended that the muscles of the pelvic floor, with attention to the levator ani and coccygeal muscle, be evaluated in these patients and that OMT techniques, including muscle energy and myofascial release, be considered in patients with evidence of hypertonicity. Furthermore, it should be acknowledged that the relative roles of thermoreceptor (eg, transient receptor potential cation channel subfamily M member 8), peripheral nervous system, or central nervous system dysfunction in patients with chronic pain remain incompletely defined.²⁵ It is possible that restoration of autonomic tone may be beneficial in pelvic pain conditions such as vulvodynia, and somatic dysfunctions of the pubic bone and sacrum should be evaluated and managed. Specifically, sacral dysfunction may alter both parasympathetic tone through pelvic splanchnic intervention at S2-4, as well as sympathetic tone at these levels, which could affect external genitalia. 

Reduced mechanical^5,10,12,13 and heat²⁶ pain thresholds have been reported at peripheral body regions in patients with PLV. In the current study, we found lower thermal (cold and heat) but not mechanical pain thresholds at the forearm (Figure 1). In light of the evidence of increased peripheral pain sensitivity, tender point examination should be considered as part of the clinical assessment of patients with PLV. Decreased pain thresholds at peripheral sites and reports of high rates of coexistence of PLV with other chronic pain disorders have also been suggested as evidence of central sensitization of pain in patients with PLV. However, because it is not always clear whether comorbid chronic pain conditions are used as exclusion criteria in PLV studies,^10,12,13,26 QST studies that report altered peripheral pain thresholds in patients with PLV should be interpreted with caution. We excluded patients with widespread pain conditions that would directly affect peripheral measurements but not patients with comorbid visceral pain conditions. A 2009 study⁷ failed to detect differences in forearm heat and mechanical pain thresholds in patients with PLV when the presence of other chronic pain conditions were used as exclusion criteria. It is possible, however, that this lack of difference is due to conservative cutoff pressures for applied pressure, which underestimates pain thresholds.⁷ More research is required to determine the contribution of central sensitization to the pathogenesis of pain in patients with vulvodynia. Furthermore, studies are needed to determine whether somatic dysfunction contributes to the pain profile of patients with PLV and to evaluate the effectiveness of OMT techniques, such as strain-counterstrain, as treatment options for women with this chronic pain disorder. 

It has been documented that PLV is associated with depression, stress, anxiety, and relationship distress.⁴ Similarly, we report that the PLV group had more depressive symptoms (Table 1). A community-based study showed that antecedent depression increased the risk of PLV, and PLV also increased the risk of new psychological conditions developing.²⁷ The impact of PLV on a woman's physical, psychological, social, and relational well-being highlights the importance of incorporating the mind-body-spirit philosophy of osteopathic medicine into the care for these patients. 

As a pilot study, our group sizes were small, and we did not differentiate between primary and secondary PLV. There is evidence of psychophysical and psychosexual differences between patients with primary and secondary PLV,^14,28,29 suggesting that these groups have distinct mechanisms of pain. Additionally, we only performed 2 repetitions of pain thresholds to minimize patient burden of these procedures despite the improvement in reliability with increased repetitions. Because of our recruitment method, QST measurements were not blinded. However, as participants were in control of determining their pain thresholds, this method likely did not have a significant impact on recorded values. 

Although vestibular sensory detection thresholds were similar in PLV and control groups, mechanical and cold pain thresholds were lower. Additionally, peripheral pain thresholds were lower in the PLV group. Taken together, our results provide further evidence of central sensitization of pain in patients with PLV and emphasize the importance of treating the whole person, as PLV involves more than just localized pain. Studies that examine the effect of OMT approaches on vulvar pain, pelvic floor dysfunction, and systemic pain in patients with PLV would help further define the role of OMT in the management of PLV. 

We thank Eshetu Tefera, MS, in the Department of Biostatistics and Epidemiology at Medstar Health Research Institute in Washington, DC, for his assistance with statistical analysis.