Diagnostic criteria for congenital long QT syndrome in the era of molecular genetics: do we need a scoring system?

AIMS: Previously published diagnostic systems, based on ECG analysis and clinical parameters (Schwartz criteria and Keating criteria), have been used to estimate the probability of inherited long QT syndrome (LQTS). Nowadays, a certain diagnosis can often be made by DNA testing. We aimed to establish the predictive power of the Schwartz and Keating criteria, using DNA testing as a reference, and to determine the best diagnostic strategy. METHODS AND RESULTS: We studied 513 relatives (aged >10 years) of 77 consecutive LQTS probands with a known disease-causing mutation. The Schwartz criteria identified 'high probability of LQTS' (score >or=4) in 41 of 208 mutation carriers, yielding 19% sensitivity and 99% specificity. The Keating criteria had 36% sensitivity and 99% specificity. Alternatively, by analysing QTc duration alone, we found that 430 ms is the optimal cut-off value to distinguish carriers (>or=430 ms) from non-carriers (<430 ms), yielding 72% sensitivity and 86% specificity (area under the curve 0.788). CONCLUSION: The existing clinical criteria have good specificity in identifying mutation carriers. However, their sensitivity is too low for clinical use. Analysis of QTc duration alone is more useful to screen for LQTS carriership (QTc >or= 430 ms) as its sensitivity is far superior, although its specificity remains acceptable. In genotyped families, genetic testing is the preferred diagnostic test.