Differential Sensory and Clinical Phenotypes of Patients With Chronic Widespread and Regional Musculoskeletal Pain

Marina M. Finnern; Dieter Kleinböhl; Herta Flor; Justus Benrath; Rupert Hölzl


Pain. 2020;162(1):56-70. 

In This Article


Sensory-clinical Phenotypes in Chronic Widespread and Localized Primary Musculoskeletal Pain

Patients with diagnoses of FMS and CBP were classified according to their sensory-clinical phenotypes of pressure and heat pain sensitivity combined with spatial spread and severity of the clinical pain. Sensory phenotypes were derived by stepwise data-reduction through descriptive, factor, and cluster analyses. We had hypothesized that the identified pain phenotypes would relate to several subgroups of widespread and regional musculoskeletal pain which differ in pain mechanisms, pain response, and certain not directly pain-related aspects such as comorbidity with somatic and mental disorders. The general hypothesis 1 was clearly supported in that 4 coherent clusters were identified which differed distinctly in pain sensitivity and degree of spatial spread of clinical pain as well as concomitant secondary pain responses.

Pain Sensitivity Across Disorders

Hypothesis 2 assumed that the overall number of sensitive body sites indicates a generalized pressure hypersensitivity representing a continuum from regionally confined pain (CBP) to highly pressure-sensitive patients with widespread pain across diagnoses. The hypothesis was supported by the increasing number of pressure-sensitive body sites from cluster 4 to the most sensitive cluster 1 with a parallel decrease of pressure pain thresholds.

However, the corollary assumption that ACR TPs would not qualitatively differ from CPs was rejected: Patients with FMS, in particular, reported higher pain intensity at tender than at control points and their provoked pain intensity increased systematically with sensitive TPs but not with CPs. Most but not all patients with CBP reported less than 11 sensitive ACR TPs, and pain intensity did not relate to the number of sensitive TPs or CPs. These results are concordant with previous studies on enhanced pressure sensitivity of patients with FMS in TPs, which is less pronounced in other body sites.[26,74] Our results emphasize the disputed clinical significance of TPs.[71]

Subgroup Characteristics

Our hypothesis 3 specified further that the 2 or more circumscribed subgroups could be differentiated in the continuum from localized to widespread pain by sensory and clinical pain characteristics without recourse to secondary indicators. This was confirmed by the isolation of 2 intermediary subgroups of patients between those in cluster 1 with widespread pain, many pressure-sensitive body sites, and low pressure pain thresholds and cluster 4 with regional pain, few pressure-sensitive body sites, and high pressure pain thresholds. The intermediary clusters 2 and 3 resembled the respective neighboring clusters in certain aspects (large vs small pain extent; many vs few sensitive body sites) but differed in other variables, thus representing 2 phenotypically discernible subgroups. Two-thirds of the patients in cluster 2 had an FMS diagnosis, and all but 1 patient had ≥11 TPs similar to the high-sensitive cluster 1, but they showed less enhanced pressure sensitivity and normal pressure pain thresholds. The latter was the only indicator which differentiated cluster 2 from cluster 1, suggesting an intermediary pain status with still developing pain enhancement. The few patients in the intermediary cluster 2 not meeting classical FMS criteria may be considered as having "widespread or FMS-like pain" and increased muscle and/or soft-tissue pain sensitivity, possibly on their way to aggravation. Cluster 3 was dominated by patients with CBP and was characterized by less pain loci and a spread of muscle and soft-tissue hypersensitivity to manual probing with quantitative thresholds still being normal. By contrast, patients in cluster 4 appeared qualitatively different from all others. They showed regionally confined pain with low to normal pressure sensitivity and consisted exclusively of patients with a CBP diagnosis. Their sensory characteristics deserve a closer look in a larger group to improve differential diagnostic assessment of "pure" back pain patients compared with those in cluster 3.

Our data are in support of other studies which suggested that already more than 6 TPs qualify patients for an FMS diagnosis.[37] Increasing numbers of hypersensitive body sites above that point could simply reflect a progressive spread with increasing pain severity rendering cut-off criteria problematic. We expect that thorough qualitative and quantitative sensory characterization of "tender" body regions would allow for better differentiation of subgroups. The number of body regions in pain was recently suggested to differentiate widespread from regional pain syndromes.[73] This variable was confirmed as distinct discriminator in this study; however, the number of pain regions was not sufficient to identify the intermediary sensitivity-based clusters without sensory data.

Contrary to hypothesis 5, our heat pain indicators did not reveal a separate phenotype of generalized hyperalgesia or enhanced windup which had been related to impaired descending inhibitory control in FMS.[32,36] Thus, the identified sensory phenotypes appear to be fairly specific to pressure pain. However, this interpretation is preliminary because of the limited thermonociceptive assessment, the special temporal summation procedure, as well as the marked sample attrition after combining pressure and heat pain sensitivity indicators in the analysis. The issue remains important and requires further mechanistic studies as already established for neuropathic pain.[3]

Psychosocial Characteristics and Psychopathology

This study set out to differentiate sensory-clinical pain phenotypes possibly related to different types of pain processing and to elucidate the role of mental and somatic comorbidity and psychosocial factors in defining these pain subtypes (hypothesis 4). This aim was only partially achieved in so far as the directly pain-related indicators differentiated between the phenotypes, whereas coping, interference by, and impact of the pain symptoms including global chronicity were not characteristic. This suggests that the phenotypic domains of primary sensory pain processing and the clinical picture should be distinguished from psychosocial and psychosomatic aspects. The conclusion is corroborated by the fact that psychopathology (depression and anxiety), sleep quality, and stress burden did not significantly differ between the phenotypically defined subgroups. This suggests that comorbid depression or depression equivalents such as sleep disturbance may not be cardinal criteria for FMS[29] as a pain disorder and is in line with the current controversy about further FMS revisions.[2,70] However, closer inspection revealed bimodal frequency distributions of depression and anxiety scores, especially within clusters 1 and 4. This might indicate subclusters with comorbid affective pathology to be identified by further dedicated designs.

The Role of Somatic Symptoms

In contrast to the fairly even distribution of comorbid psychopathology, the burden of functional somatic symptoms differed characteristically between the phenotypes. Self-reported body complaints and perceived somatic health scored highest in the pressure-sensitive clusters with widespread pain. Cardiovascular complaints, exhaustion, and gastrointestinal complaints were most prominent in these subgroups. The enhanced magnitude of medically unexplained somatic symptoms may reflect a general hypervigilance to somatic input with enhanced symptom perception.[5,11,47] Earlier work had already suggested that the high incidence of other somatic symptoms of patients with widespread pain might be due to an underlying common functional somatic syndrome.[68] The systematic trend of increasing functional somatic symptoms from the low to the high-sensitive clusters with increasing spatial spread of pain in our study would indeed argue for such a dimensional diagnostic view.[56] However, our results also suggest that sensory changes in afferent signaling may be an important contribution to the development of widespread pain with associated somatic complaints. The relative high incidence of gastrointestinal symptoms with the high-sensitive clusters concurs with the accumulating evidence of a discriminable subsyndrome of fibromyalgia with FGID.[4,10,62] This and the reported comorbidity with pelvic pain[35] might imply a relationship to sex-specific syndromes of enhanced sensitivity to soft-tissue and visceral pain.[58,63] Similar arguments apply to the comorbid chronic fatigue syndrome, sleep disorders, and other somatic syndromes of the depressive spectrum.[33,61] The high proportion of women in the FMS-like clusters of our data is suggestive in the same direction. Comprehensive sensory assessment should help to differentiate these hypersensitive subgroups early in the chronic development and to assign tailored interventions.