Construct validity evaluation of the European Scleroderma Study Group activity index, and investigation of possible new disease activity markers in systemic sclerosis.

OBJECTIVES: To evaluate the construct validity of the European Scleroderma Study Group (EScSG) activity index and to propose modifications if necessary. METHODS: One hundred and thirty-one consecutive patients were investigated and re-evaluated 1 year later. Modified Rodnan skin score (MRSS), skin ulcers and joint contracture numbers, hand anatomic index (HAI), BMI, spirometry, carbon monoxide diffusing capacity (DL(CO)), left ventricular ejection fraction, pulmonary arterial hypertension, HAQ Disability Index (HAQ-DI), patient skin self-assessment questionnaire and several biomarkers were recorded, in addition to the data required for the EScSG activity index. Statistical analysis was performed by categorical principal component analysis (CATPCA). RESULTS: The EScSG activity index appeared in the same dimension as the HAQ-DI, ulcer score and joint contractures, MRSS, patient-reported skin score and HAI by CATPCA. Parameters of lung involvement appeared in another dimension. We constructed a 12-point activity index that was equally associated with both dimensions, by adding the forced vital capacity/DL(CO), change in DL(CO), change in the ulcer scores, HAQ-DI and patient-reported skin score. Biomarkers including vascular endothelial growth factor, soluble P-selectin glycoprotein ligand-1, CRP and albumin were related to both the EScSG and the 12-point index, though they did not improve the total variance of the model. CONCLUSION: The construct validity of the EScSG activity index is good, though the lung-related disease activity may not be sufficiently represented. Further validation steps may be required for both the EScSG and our 12-point activity index.

Establishment and partial validation of a patient skin self-assessment questionnaire in systemic sclerosis.

OBJECTIVES: To construct a patient self-assessment questionnaire testing whether patients can provide valid information about their skin thickness, tethering and thinness. METHODS: The questionnaire contained questions about the 17 body areas identical to those investigated in the Modified Rodnan Skin Score (MRSS). The baseline and 1 yr follow-up values of thickening, tethering and thinning were scored on a scale of 0-3, by 131 consecutive patients. RESULTS: The questionnaire was feasible. The test-retest reliability was good, with an intraclass correlation coefficient of 0.5-0.6. The MRSS correlated with the 17-area thickness score (rho = 0.435, P < 0.001). Both the patient-reported thickness and tethering scores and the expert-measured MRSS and tethering scores correlated highly with each other (rho = 0.523, P < 0.001 and rho = 0.637, P < 0.001, respectively). The patient-reported thinness scores did not show any correlation with the examiner-measured skin thickening/tethering. All thickness-related scores were sorted into the same factor by principal component analysis. Conversely, skin thinness scores were sorted into a distinct factor. Skin thickness scores reported by diffuse SSc patients were different from those of the limited SSc cases (P < 0.05). CONCLUSIONS: We validated the skin thickness domain of the questionnaire, except the responsiveness. Because of the low mean MRSS of patients in this investigation, the patient self-assessment validation procedure should be repeated with cases exhibiting more extensive skin involvement. Patient-reported and examiner-measured tethering may not be used as independent instruments of skin involvement. The skin thinness domain may contain valuable independent information, therefore it also merits further investigation.