Chest X-ray and electrocardiogram in post-cardiac surgery follow-up clinics: should this be offered routinely or when...

Many centres in the UK carry out routine chest X-ray (CXR) and/or electrocardiogram (ECG) when patients attend follow-up clinic after cardiac surgery. Current evidence to support this practice is weak. This study investigated the appropriateness of carrying out these investigations in the absence of clinical indication.

All patients attending routine 6- to 8-week follow-up clinic after cardiac surgery in this hospital were prospectively reviewed over a 6-month period (October 2011-April 2012). Two groups were identified for comparison. Group A comprised patients who had CXR and/or ECG requested routinely, and those in Group B had the investigations only when clinically indicated. A proforma was designed to screen each patient for cardiac and respiratory symptoms, predischarge CXR abnormalities and the presence of atrial fibrillation/flutter postoperatively. Management alterations based on the findings from the investigations were noted. Patients who had thoracic, major aortic, or heart transplant surgery were excluded from the audit.

Three hundred and fifty patients were reviewed: 250 were in Group A and 100 in Group B. No patient had a significant management alteration in the absence of an indication for the tests. There were no differences in outcome between the two groups. In Group A, 111 (45%) patients had CXR and ECG done without indication and no abnormality was detected. In Group B, 52 patients had no indication for either tests and were thus not tested. None of these patients required readmission/intervention following discharge from clinic. Overall, 271 patients had CXR carried out, with only 83 being indicated. This led to a management alteration in 33 patients (12% overall, 40% if indicated). Two hundred and eighty-six patients had ECG carried out with 140 indicated. Management was altered in 122 patients (43% overall, 87% if indicated). The correlation between the clinical indication-based investigation and the resulting change in patient management was found to be significant (Goodman-Kruskal Gamma: 0.99, P = 0.000 for both investigations).

There is a strong correlation between clinical indication for CXR and/or ECG and management alterations. These investigations should be performed during the routine follow-up of adult cardiac surgical patients using a patient-centred approach based on signs and symptoms.