Background: Diastolic dysfunction is often used as a marker of allograft rejection in patients with cardiac transplantation. However, the strength of evidence to support this approach needs to be determined.
Methods: We sought to review and evaluate the quality of the published data about the use of diastolic indices (DI) to predict a rejection episode. A MEDLINE search was performed to identify English-language articles published from 1967 to 2005 that used DI to predict rejection episodes. The references of each retrieved article were scanned to identify additional relevant articles. Studies showing the correlation between echocardiography findings and biopsy results were included. Articles were abstracted using a standardized instrument designed by the authors. Quality was assessed using standardized criteria. Discrepancies were adjudicated by consensus.
Results: Nineteen studies met the inclusion criteria. Quality of studies varied widely. Thirteen studies explicitly stated having an independent blind comparison with a reference standard, 4 studies presented reproducibility information, no studies presented likelihood ratios, and only 11 of them provided sensitivity and specificity. Left ventricular diastolic indices evaluated included inflow Doppler early diastolic (E) wave pressure half-time (sensitivity 23%-87%, specificity 76%-98%) and isovolumetric relaxation time (sensitivity 28%-85%, specificity 80%-98%), as well as tissue Doppler early diastolic (E') wave (sensitivity 69%-76% and specificity 59%-88%) and late diastolic (A') wave (sensitivity 67%-82%, specificity 49%-53%).
Conclusion: Because of inconsistent quality of the studies and low sensitivities, the current literature does not support the use of DI by echocardiography as a screening test in the prediction of allograft rejection in the heart transplantation population. Larger controlled studies incorporating newer DI such as tissue Doppler are needed.