Diagnostic accuracy of the Arabic composite autonomic symptom score 31 (COMPASS-31) for detecting autonomic neuropathy in SLE: a cross-sectional validation study

Abstract

Abstract Objective To evaluate the diagnostic accuracy of the Arabic Composite Autonomic Symptom Score-31 (COMPASS-31) for detecting objectively confirmed autonomic neuropathy in patients with SLE. Methods In this cross-sectional diagnostic accuracy study, 45 patients with SLE, 23 patients with neurologically confirmed autonomic neuropathy (positive-control group), and 45 healthy controls completed the Arabic COMPASS-31 before undergoing cardiovascular autonomic reflex testing (CART) and sympathetic skin response (SSR). Objective autonomic neuropathy was defined as abnormal CART and/or absent SSR. Diagnostic accuracy analyses (receiver operating characteristic (ROC) curve analysis, optimal cut-off determination and multivariable logistic regression) were performed within the SLE cohort. Results Objective autonomic neuropathy was present in 28 of 45 patients with SLE (62%). Total COMPASS-31 score discriminated autonomic neuropathy status with an area under the ROC curve (AUC) of 0.81 (95% CI 0.68 to 0.94) at a cut-off of ≥5 (sensitivity 89% and specificity 59%), increasing to 0.89 (95% CI 0.79 to 0.98) when restricted to definite/severe autonomic neuropathy. The orthostatic intolerance (AUC 0.78, 95% CI 0.64 to 0.91) and pupillomotor (AUC 0.77, 95% CI 0.64 to 0.90) domains performed similarly to the total score (DeLong’s test, both p>0.4), while the vasomotor, secretomotor, gastrointestinal and bladder domains performed significantly worse. The COMPASS-31 total score remained independently associated with autonomic neuropathy after adjustment for age, disease duration and disease activity (adjusted OR 1.29, 95% CI 1.06 to 1.57). Internal consistency of the Arabic COMPASS-31 was good (Cronbach’s α=0.83, McDonald’s ω=0.857). Conclusion The Arabic COMPASS-31 demonstrates good discrimination for objectively confirmed autonomic neuropathy in SLE, with the orthostatic intolerance and pupillomotor domains contributing most of its discriminative value. It can be used as a screening tool to identify patients who warrant confirmatory autonomic testing. Disease-specific validation should precede recommending COMPASS-31 as an autonomic neuropathy screening tool in other rheumatic diseases.