Abstract
Pulmonary hypertension (PH) is diagnosed by right heart catheterization (RHC), but non-invasive methods play an important role in screening and follow-up. ECG is routinely performed in the clinical work-up but its clinical value has not been defined. We hypothesized that a simple parameter, such as the presence of right axis deviation (RAD) in standard ECG might be useful in the diagnosis and follow up of PH.
We retrospectively analysed all patients who received a RHC and an ECG in our department between 2005 and 2010. The indication for RHC was heterogeneous, the pretest probability of PH was about 50%. We determined the ratio of the S and R waves in lead I in each patient, a value ≥1 (≥90°) was considered as RAD. The investigator was blinded to the RHC results. The sensitivity and specificity of RAD to predict PH (mean PAP≥25mmHg) was determined.
N=317 patients were included. RAD was present in n=71 patients. Within these, PH was detected in n=65 patients, and was missing in n=6 subjects (p<0,001). In n=246 patients without RAD, PH was present in n=87 patients. The specificity of RAD for PH was 96%, whereas the sensitivity was merely 43%. Accordingly, the positive predictive value of RAD to predict PH was 92%, while its negative predictive value to rule out PH was just 35%.
Our retrospective analysis on a large, heterogenous cohort of subjects including patients with and without PH suggests that the presence of RAD in the ECG has a high specificity but a low sensitivity for PH. This simple method may help to recognize PH patients within a population at risk for PH. In order to determine its appropriate role in a diagnostic algorithm, prospective, population based studies are needed.
- © 2011 ERS