Open Access Open Badges Original investigation

Screening for left ventricular hypertrophy in patients with type 2 diabetes mellitus in the community

Jithendra B Somaratne, Gillian A Whalley, Katrina K Poppe, Mariska M ter Bals, Gina Wadams, Ann Pearl, Warwick Bagg and Rob N Doughty*

Author Affiliations

Cardiovascular Research Group, Department of Medicine, Faculty of Medical and Health Sciences, The University of Auckland, Private Bag 92019, Auckland Mail Centre, Auckland 1142, New Zealand

For all author emails, please log on.

Cardiovascular Diabetology 2011, 10:29  doi:10.1186/1475-2840-10-29

Published: 14 April 2011



Left ventricular hypertrophy (LVH) is a strong predictor of cardiovascular disease and is common among patients with type 2 diabetes. However, no systematic screening for LVH is currently recommended for patients with type 2 diabetes. The purpose of this study was to determine whether NT-proBNP was superior to 12-lead electrocardiography (ECG) for detection of LVH in patients with type 2 diabetes.


Prospective cross-sectional study comparing diagnostic accuracy of ECG and NT-proBNP for the detection of LVH among patients with type 2 diabetes. Inclusion criteria included having been diagnosed for > 5 years and/or on treatment for type 2 diabetes; patients with Stage 3/4 chronic kidney disease and known cardiovascular disease were excluded. ECG LVH was defined as either the Sokolow-Lyon or Cornell voltage criteria. NT-proBNP level was measured using the Roche Diagnostics Elecsys assay. Left ventricular mass was assessed from echocardiography. Receiver operating characteristic curve analysis was carried out and area under the curve (AUC) was calculated.


294 patients with type 2 diabetes were recruited, mean age 58 (SD 11) years, BP 134/81 ± 18/11 mmHg, HbA1c 7.3 ± 1.5%. LVH was present in 164 patients (56%). In a logistic regression model age, gender, BMI and a history of hypertension were important determinants of LVH (p < 0.05). Only 5 patients with LVH were detected by either ECG voltage criteria. The AUC for NT-proBNP in detecting LVH was 0.68.


LVH was highly prevalent in asymptomatic patients with type 2 diabetes. ECG was an inadequate test to identify LVH and while NT-proBNP was superior to ECG it remained unsuitable for detecting LVH. Thus, there remains a need for a screening tool to detect LVH in primary care patients with type 2 diabetes to enhance risk stratification and management.