The diagonal ear lobe crease (Frank's sign) is not associated with coronary artery disease or retinopathy in type 2 diabetes: the Fremantle Diabetes Study

Citation
Tme. Davis et al., The diagonal ear lobe crease (Frank's sign) is not associated with coronary artery disease or retinopathy in type 2 diabetes: the Fremantle Diabetes Study, AUST NZ J M, 30(5), 2000, pp. 573-577
Citations number
25
Categorie Soggetti
General & Internal Medicine","Medical Research General Topics
Journal title
AUSTRALIAN AND NEW ZEALAND JOURNAL OF MEDICINE
ISSN journal
00048291 → ACNP
Volume
30
Issue
5
Year of publication
2000
Pages
573 - 577
Database
ISI
SICI code
0004-8291(200010)30:5<573:TDELC(>2.0.ZU;2-L
Abstract
Background: The diagonal ear lobe crease (ELC) has been suggested as a simp le marker of vascular disease in the general population but there are few d ata from diabetic patients despite their increased risk of angiopathy. Aim: To determine whether the ELC is a clinically useful sign of coronary a rtery disease (CAD) or retinopathy in type 2 diabetes. Methods: One thousand and twenty-two patients from the multi-ethnic urban c atchment area of Fremantle Hospital in Western Australia were studied. This sample represents 79% of the type 2 diabetic subjects recruited to the ong oing Fremantle Diabetes Study and 49% of all 2072 patients with type 2 diab etes identified through active case detection in a postcode-defined region of 120,097 people. In addition to other comprehensive data relating to diab etes and its management, the presence of an ELC and evidence of both CAD an d retinopathy were ascertained for each patient. Results: The prevalence of ELC was 55%. Patients with an ELC were more like ly to have CAD than those without an ELC (p=0.019), but the proportions wit h retinopathy were not significantly different (p=0.085). The sensitivity a nd specificity of ELC for detecting CAD were 60% and 48%, and for retinopat hy 61% and 47% respectively. The patients with an ELC were significantly ol der, more likely to be male and had a higher systolic blood pressure than t hose without (p<0.02). After adjusting for known vascular risk factors, soc ioeconomic variables and ethnicity in a logistic regression model an ELC wa s neither a significant independent predictor of CAD (p=0.45) nor of retino pathy (p=0.14). Conclusions: The ELC is of little value as a sign of the presence of diabet ic vascular complications.