Introduction: Systemic sclerosis is a chronic multisystem disease. Many patients with limited scleroderma do not meet the criteria of the American College of Rheumatology (ACR), although they clearly have skin changes. There is no information regarding the sensitivity and specificity of manifestations of scleroderma. The aim of this study was to determine the sensitivity and specificity of the vascular and skin signs in patients with limited scleroderma.
Methods: Forty patients with limited scleroderma according to the ACR criteria, 40 patients with other collagen vascular diseases, and forty healthy persons were selected for this study. Ten parameters (acrocyanosis, acro-osteolysis, gangrene, Raynaud's phenomenon, pitting ulcer, more than 5 facial and palmar telangectasia, hyper- and hypopigmentation, vertical and horizontal mouth diameter, and the number of radial furrowing around the mouth) were evaluated in them.
Results: The mean age for both groups of scleroderma and healthy persons was the same (40 years). The patients with collagen vascular disease had a mean age of 36 years old. More than 95% of patients in each group were female. The mean duration of the disease at onset was 7.6 years in the scleroderma patients. Raynaud's phenomenon and pitting ulcers had the highest sensitivity (97.5% and 82.5%, respectively). Acrocyanosis, acro-osteolysis, gangrene, pitting ulcer, hypo and hyper pigmentation, more than 5 facial and palmar telangiectasia had 100% specificity. Receiver-operator curve (ROC) was used to determine the best cut-off point for vertical and horizontal mouth diameter and the number of radial furrowing around the mouth. The number of the radial furrowing around the mouth had the largest area under the curve with 80% sensitivity and 87.5% specificity for 5 folds or more.
Discussion: Raynaud's phenomenon is the best and the first sign to rule in or out the disease because it has the most positive likelihood ratio (39) and the least negative likelihood ratio (0.02). The 100% specificity of the skin signs may be due to the lack of dermatologic patients in the control group. It is therefore recommended to design studies with patients in their early onset of the disease and control groups consisting of patients with skin diseases.
Hakim Research Journal 2005 8(3) 25-30.