Prediction of the postoperative left ventricular (LV) function in valvular heart disease that will cause LV volume overloading, such as chronic mitral regurgitation (MR) and aortic regurgitation (AR), remains elusive. We sought to test if 2-dimensional peak negative longitudinal strain (LS2D) was useful for prediction of the postoperative LV function in relation to this disease entity.

Newly developed speckle tracking imaging was performed preoperatively to measure the LS2D in 26 and 22 patients with MR and AR, respectively. A favorable response after the operation (FR) was defined according to the change in the LV ejection fraction (EF): 1) a LVEF ≥55% both pre- and post-operation, 2) a postoperative LVEF ≥55% with a pre-operative LVEF between <55 and ≥45%, or 3) an increase in the LVEF >10% with a pre-operative LVEF <45%.

Follow-up echocardiography was performed at an average of 6.7±2.3 months after the operation. FR was confirmed in 36 patients (LVEF from 55.4±10.1 to 58.3±5.0%, p=0.06), with the remaining 12 showing an unfavorable response (LVEF from 52.5±7.6 to 45.1±5.4%, p<0.01). There was no significant difference in the baseline characteristics, including underlying etiologies, operation techniques, and cardiopulmonary bypass time, and LV volumes and EF between the two groups. The only difference was the LS2D, which was significantly larger in the FR group (-19.8±3.9 versus -16.2±2.9%, p<0.01). An LS2D of -18.3% could predict an unfavorable response of the LVEF following an operation, with a sensitivity and specificity of 75 and 75%, respectively.

The preoperative LS2D is a useful predictor of the postoperative left ventricular function in AR and MR.