Central pain processing in women with chronic pelvic pain syndrome

Chronic pelvic pain (CPP) is chronic or persistent pain perceived in structures related to the pelvis of either men or women. It is often associated with negative cognitive, behavioural, sexual, and emotional consequences as well as with symptoms suggestive of lower urinary tract, sexual, bowel, pelvic floor or gynaecological dysfunction. The UK annual prevalence of CPP is estimated at 3.8% and it is thought there are approximately one million women with CPP living in the UK.CPP is the most common cause for gynaecology referral accounting for 20% of all gynaecology consultations and 40% of diagnostic laparoscopies. In approximately 35% of cases there is no obvious pelvic pathology, leading to the diagnosis of Chronic Pelvic Pain Syndrome (CPPS). Despite CPP being associated with significant socio-economic cost, it still remains poorly understood, making clinical management challenging. The overarching objective of my MD thesis was to understand how CPPS is clinically managed in the UK, understand the mechanisms underpinning central pain processing and how this knowledge can influence future clinical practice. Specifically, I had the following aims: 1) Investigate the current clinical management trends for women with CPPS in the UK. 2) Determine if there are differences in central pain processing between women with CPPS and healthy female volunteers with no pelvic pain. 3) Investigate if existing clinical questionnaires can stratify women with CPPS into clinically meaningful groups. To achieve the first aim, I designed and distributed an online survey to UK clinicians managing chronic pelvic pain. A descriptive statistical analysis of the responses was carried out using an online survey tool. My online survey generated 148 responses. 45% of survey respondents indicated that they felt the management of women with CPP in the UK is 'poor' or 'very poor'. 51% of clinicians prioritised 'pain management' above other aspects of care. Pelvic ultrasound and diagnostic laparoscopy were the most frequently performed investigations (97% and 78% respectively). The majority of clinicians (66%) discharged patients back to primary care without any further management plan or follow-up after negative findings at diagnostic laparoscopy. Of those clinicians who instigate post-operative treatment, hormonal therapies were the most commonly used (combined oral contraceptive pill - 84%, GnRH analogues - 69%, progestogens - 52%). The survey results indicate that CPP is managed predominantly by obstetricians and gynaecologists with relatively low case load, suggesting that majority are not 'specialists' in the field. There is surprisingly high level of dissatisfaction amongst the clinicians about the care they provide. The majority of women are referred back to community care following 'negative' finding at diagnostic laparoscopy, often without treatment or follow-up plans. Survey participants reported lack of seamless referral pathways and lack of multi-disciplinary approach as main barriers in delivering quality care to women with pelvic pain. This highlights an urgent need for research, education and improved clinical guidance to advance the management of this common but challenging gynaecological problem. To address the second aim, I performed a case-controlled study comprising two groups of participants, women with CPPS and healthy female volunteers with no pelvic pain. Women with CPPS and healthy volunteers meeting the inclusion criteria completed validated measures of pain and psychological state, and provided a blood sample for serum estrogen and progesterone levels. Subsequently, all participants underwent a functional MRI (fMRI) scan of their brain. The scan session included an acquisition of structural images and event-related task fMRI with variable interval design, during which participants received punctate stimuli over the suprapubic area of the abdomen using 300-gramm Von Frey filament. The MRI data were analysed using FMRIB's Software Library (FSL). Standard data pre-processing (brain extraction, registration, B0 field unwarping and motion correction) was carried out. FSL FEAT was used for the first and higher-level analyses using mixed effects methods (Z score threshold = 3.1, cluster corrected regions considered to be significant at P<0.05). A sub-group of healthy volunteers (n=13) was matched with women with CPPS (n=13) for their age (+/- 5 years) and hormonal status (method of contraception or the stage of menstrual cycle). This approach allowed further analyses controlling for hormonal state, a well-known factor influencing pain modulation. I investigated the hypothesised differences in central pain processing between women with CPPS and healthy female volunteers in two ways: 1) Base analysis: 42 women with CPPS compared with 15 healthy volunteers. This approach maximised sample size and statistical power but did not control for hormonal state differences between the groups. 2) Matched analysis: 13 'hormonally matched' women with CPPS compared with 13 healthy volunteers. This approach controlled for the hormonal state, but smaller cohort had reduced power to detect fMRI differences in brain activity between the analysed groups. In the 'base analysis', there was no difference (P=0.46) in the age between healthy volunteers (median=27, range: 23-35) and women with CPPS (median=26, range: 19-50). As predicted, women with CPPS experienced significantly higher (P<0.0001) level of experimental pain (mean=4, range: 0-9 out of 10) than healthy volunteers (mean=1, range: 0-2 out of 10). All other measures of pain (background pain, current pain, Brief Pain Inventory score and PainDETECT neuropathic pain questionnaire) and psychological distress measures (General Health Questionnaire, Pain Catastrophising Questionnaire) were also significantly higher in women with CPPS. Likewise, in the 'matched analysis', there was no statistically significant difference (P=0.69) in the age between healthy volunteers (mean=28, range: 23-35) and matched women with CPPS (mean=27, range: 21-35). There was also no difference in hormonal usage between both groups (15% - COCP, 16% POP, 15% - Nexplanon, 31% - Mirena IUS, 16.7% - follicular phase of unsuppressed menstrual cycle). Similarly, as in the 'base analysis', women with CPPS experienced significantly higher (P=0.002) level of experimental pain (mean=4, range: 1-8 out of 10) than healthy volunteers (mean=1, range: 0-2 out of 10) as well as all other measures of pain and psychological distress. Examination of the fMRI data in 'base' and 'matched analyses' did not identify any significant differences in stimulus-related neuronal activity between the groups using whole-brain and restricted (regions known to be altered in other chronic pain conditions) approach. My study investigating the differences in central pain processing between women with CPPS and healthy volunteers confirms significantly higher level of background and experimental pain, higher levels of mental health distress, catastrophic thinking, and higher likelihood of neuropathic pain amongst women suffering from CPPS. This, however, does not translate to altered brain activity in response to painful stimuli as it is documented in other chronic pain conditions. A likely explanation is the heterogeneity of CPPS phenotypes and dynamic central nervous system 'pain matrix' consisting of various brain regions involved in pain amplification. Indeed, the current literature provides inconsistent results in relation to central pain processing. To address the third aim, I identified three key factors known to be associated with pain modulatory mechanisms: type of pain (nociceptive/neuropathic), level of psychological distress and presence of catastrophic thinking. These factors were measured using established clinical questionnaires: PainDETECT, General Health Questionnaire (GHQ-12) and Pain Catastrophising Questionnaire (PCQ), respectively. Women with CPPS were stratified into high and low scoring sub-groups in each of the questionnaires. Pain and psychological measures were analysed defining clinically relevant pain/psychological phenotypes. Analyses of the fMRI data from the previous experiment were carried out to determine if there are differences in central pain processing between high and low scoring participant phenotypes. Stratification of women with CPPS using questionnaires was able to identify participants with significantly higher level of pain severity and pain interference but none of them was able to identify difference in response to the experimental painful stimuli or relief from analgesics. GHQ-12 and PCQ questionnaires stratified women with CPPS in all three key factors known to be associated with pain modulatory mechanisms. In summary, the results presented in my thesis considerably add to the existing data on our understanding and management of women reporting CPP. My survey highlights the need for urgent work to improve understanding of the problem and improvements in clinical care in UK. Behavioural and fMRI data presented here explored the complexity of central pain processing in CPPS and its uniqueness from other chronic pain conditions. Lastly, exploration of the potential for the use of questionnaires in daily clinical practice could streamline and improve the management pathways for women with CPPS.