Primary care assessment of capillaroscopy abnormalities in patients with Raynaud's phenomenon

Clin Rheumatol. 2015 Dec;34(12):2135-40. doi: 10.1007/s10067-015-3062-3. Epub 2015 Sep 23.

Abstract

Raynaud's phenomenon is a clinical symptom that can commonly present to a primary care provider or generalist. Proper identification of an underlying connective tissue disease in a patient with Raynaud's could allow for the prevention of possible critical digital ischemia. Capillaroscopy is a tool which can identify abnormalities associated with connective tissue disease. Patients presenting with a complaint of Raynaud's phenomenon were assessed with capillaroscopy. In twenty consecutive Raynaud patients, 8 digits were assessed by a ×200 magnification dermatoscope and an image was obtained. Each image was assessed for the following abnormalities: drop-out (<9 capillaries in 1 mm); microhemorrhage; dilated loops; and neoangiogenesis. These 160 images were then shown to 20 primary care physicians, who assessed these same abnormalities. The interrater reliability, a measure of agreement, of individual primary care providers with the expert provider was assessed using kappa statistics. Three raters had slight agreement (in the range 0 to 0.20), one rater had fair agreement (0.21 to 0.40), 11 raters had moderate agreement (0.41 to 0.60), five raters had substantial agreement (0.61 to 0.80), and no rater had almost perfect agreement (0.81 to 1.00) (14). The total agreement from the 20 primary care providers (n = 3,156) was moderate (Κ = 0.50, 95 % CI 0.49, 0.55). For the four providers with the slight to fair interrater reliabilities, the most common disagreement was providing a positive diagnosis when the expert rater diagnosed the digit negative. Ten of the twenty primary care providers provided at least one additional diagnosis following an abnormal diagnosis (n = 35 digits or 35 % of the 1556 abnormal ratings by the primary care providers). The four providers with the poorest interrater reliabilities were not among the ten providers who participated in making these additional specific diagnoses. These providers achieved the moderate agreement with the expert provider for diagnoses of microhemorrhage (Κ = 0.64, 95 % CI 0.57, 0.70), but fair agreement with the expert provider for diagnoses of dilated (Κ = 0.27, 95 % CI 0.20, 0.34) and neoangiogenesis (Κ = 0.22, 95 %CI 0.13, 0.31). Capillaroscopy is a potentially contributive clinical exam skill that could assist primary care providers and generalists in identifying and qualifying changes associated with the common presentation of Raynaud's disease. However, formal training is needed to ensure accuracy and reproducibility. Furthermore, training and scoring systems should address time constraints of busy primary care practitioners.

Keywords: Diagnostic tests; Methodology; Raynaud’s syndrome; Rheumatic diseases; Scleroderma; Systemic sclerosis.

Publication types

  • Research Support, N.I.H., Extramural
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Aged
  • Capillaries / pathology*
  • Dilatation, Pathologic / pathology
  • Female
  • Hemorrhage / pathology*
  • Humans
  • Male
  • Microscopic Angioscopy*
  • Middle Aged
  • Neovascularization, Pathologic / pathology*
  • Observer Variation
  • Physicians, Primary Care*
  • Raynaud Disease / diagnosis*
  • Raynaud Disease / pathology
  • Reproducibility of Results
  • Sensitivity and Specificity