Background: Echocardiographic criteria for left ventricular enlargement (LVE) used to classify subjects as affected in families with familial dilated cardiomyopathy (FDC) have been inconsistent. A recent report from a large Framingham echocardiographic study provides an opportunity to improve the assignment of LVE and FDC in kindreds, principally with a dilated phenotype. The objective of this study is to evaluate an alternative diagnostic criteria for FDC based only on LVE with no measure of fractional shortening (FS).
Methods and results: We compared our proposed criteria for LVE and FDC with previous approaches by applying them to 166 adults derived from three large FDC pedigrees. Our proposed FDC diagnostic criteria are a sex- and height-specific method based only on LVE, without regard for FS, set as a 97.5% upper limit for left ventricular end-diastolic dimension (LVEDD) from the Framingham study. Other methods used to assign LVE were (1) a 95% upper limit for LVEDD by the Framingham study; (2) the method of Henry et al. (1980) based on age and body surface area (BSA); and (3) the National Heart, Lung, and Blood Institute (NHLBI) method with a cut point of LVEDD greater than 2.7 cm/BSA. Three other commonly used diagnostic criteria for FDC were based on various LVE standards combined with an FS of 27% to 30%. For LVE, the Framingham-97.5% was the most stringent (21 of 134 subjects identified; 15.7%), the NHLBI standard the least stringent (57 of 161 subjects identified; 35.4%), and the Henry-112% method intermediate (44 of 161 subjects identified; 27.3%). More women were identified with the Framingham method (57.1%) versus the Henry-112% (40.9%). The Henry-112% and NHLBI methods identified 11.4% and 7.0% of subjects with body mass indices (BMIs) of 35 or greater, respectively. For FDC, our proposed FDC diagnostic criteria identified similar numbers of subjects (21 subjects) as the three other criteria (range, 22 to 27 subjects), but inconsistency was noted (54.2% to 66.7%), with kappa values from 0.49 to 0.55 resulting from different sensitivities to sex, LVE, FS, and BMI.
Conclusion: Our proposed FDC diagnostic criteria are stringent to assign FDC family members as affected compared with other commonly used criteria. The use of LVEDD alone may be preferable for FDC family screening, although further validation of this approach with phenotypic and genotypic data from other large FDC pedigrees is needed.