Nailfold capillaroscopy by smartphone-dermatoscope for connective tissue disease diagnosis in interstitial lung disease: a prospective observational study

Nailfold capillaroscopy (NFC) is a non-invasive tool validated for systemic sclerosis diagnosis. The role and interpretation of NFC in interstitial lung disease (ILD) patients for the diagnosis of connective tissue disease associated ILD (CTD-ILD) remains undefined. In a prospective study, quantitative and qualitative NFC by smartphone-dermatoscope (3M Dermlite-DL4ΤΜ attached to iPhone-6plusΤΜ) was performed in 96 patients with well-defined CTD-ILD (n=27) and non-CTD ILD (n=69; idiopathic interstitial pneumonia n=42, interstitial pneumonia with autoimmune features n=27) by ILD-multidisciplinary meeting. NFC scoring was performed by two independent, blinded specialist rheumatologists. Comprehensive baseline clinical, serological, physiological and radiological data were included. Multivariable models for CTD diagnosis in ILD, comprising nailfold characteristics at empirical thresholds determined by receiver operating characteristic curve analysis and clinical variables, were explored. In 94 patients with complete NFC data (total 687 images, median eight images per patient from eight digits), low capillary density (<6 capillaries/millimetre), increased giant capillaries (≥3), avascular areas (≥2) and microhaemorrhages all strongly enhanced the discrimination of CTD-ILD from non-CTD ILD (OR 5.00–7.47) independent of clinical covariates. In multivariable analysis, low capillary density and microhaemorrhages were independent predictors of CTD in ILD additional to the risk conferred by serology and radiology. Microhaemorrhages were also a strong predictor of CTD (adjusted OR 13.45, p=0.006) independent of clinical manifestations. All pre-specified qualitative NFC classification schemes identified CTD-ILD (OR range 3.27–8.47). NFC performed by smartphone-dermatoscope is an accessible, clinically feasible tool that may improve the identification of CTD further to routine clinical assessment of the ILD patient.