[Interobserver agreement on electrocardiographic diagnosis of left ventricular hypertrophy in hypertensive patients in Andalusia. PREHVIA study].

OBJECTIVE

To assess the agreement between Primary Care (PC) doctors and a cardiology specialist in diagnosing left ventricular hypertrophy in the electrocardiograph (LVH-ECG) in hypertensive patients.

DESIGN

Cross-sectional, multicentre study.

SETTING

Andalusian Primary Care Centres.

PARTICIPANTS

A total of 120 PC doctors who using a random sample selected patients of 35 years or more with AHT of at least 6 months of progression. PRIMARY VARIABLES: Demographic data, risk factors and cardiovascular diseases were recorded. The LVH-ECG was evaluated by applying Cornell voltage criteria, Cornell and Sokolow-Lyon product. The PC researchers read the ECG first and the cardiologist made a second reading blind.

RESULTS

A total of 570 patients (mean +/- SD of age, 65 +/- 11 years; 54.5% females); the LVH-ECG prevalence was 13.7% (95% CI, 10.8-16.6; 12.6% by Cornell and 1.6% by Sokolow-Lyon). The agreement in the diagnosis between the PC doctors and the cardiologist was 0.378 (95% CI, 0.272-0.486; disagreements in 15.5% of cases). The PC doctors slightly underestimated the LVH-ECG prevalence by Cornell and slightly overestimated it by the Sokolow-Lyon criteria. The agreement was also low for all of them (kappa = 0.367; 95% CI, 0.252-0.482, for Cornell, and kappa = 0.274; 95% CI: 0.093-0.454 for Sokolow-Lyon).

CONCLUSIONS

The agreement between the diagnosis by the PC doctors and the cardiologist was low. The implications of this study suggest the need to improve the reading of ECG among PC doctors. The use of computerised systems could be a good option.