BACKGROUND: After heart valve replacement, patients are at risk of complications that are linked to irregular oral anticoagulation therapy. Depending on the type and position of the valve, the complication rate is 3% to 6% per patient-year. This study presents clinical results of a device enabling early detection of prosthetic dysfunction by frequency measurements. METHODS: Between November 2003 and March 2006, ThromboCheck devices were handed out to 483 patients after mechanical heart valve replacement from six heart centers. In each case, the devices were calibrated for individual patients early postoperatively. Taking control measurements, the device calculated deviations from calibrated frequency and amplitude ranges using a fast Fourier transform and sent a warning message to the display if any relevant changes occurred. This message was taken as cause to determine morphologic correlations of the disturbance using imaging techniques, for example, echocardiography or fluoroscopy (observation period, 550.7 patient-years; median application period, 15.1 months). RESULTS: Twenty-five true positive, 1 false positive, and no false negative measurements occurred. Eleven patients (44%) received thrombolytic therapy with streptokinase or recombinant tissue-type plasminogen activator, resulting in normalization of the digital frequency analysis. In 4 patients (16%), international normalized ratio levels were adjusted under temporary heparinization therapy. Normalization of the valve sounds were achieved within 3 to 7 days. Seven patients (28%) required reoperation despite lysis therapy. Sensitivity was 100% during the observation period; specificity was 99.8%. CONCLUSIONS: Digital frequency analysis represents a reasonable addition to the monitoring of patients after artificial heart valve replacement to detect prosthesis-related complications early on.
BACKGROUND: After heart valve replacement, patients are at risk of complications that are linked to irregular oral anticoagulation therapy. Depending on the type and position of the valve, the complication rate is 3% to 6% per patient-year. This study presents clinical results of a device enabling early detection of prosthetic dysfunction by frequency measurements. METHODS: Between November 2003 and March 2006, ThromboCheck devices were handed out to 483 patients after mechanical heart valve replacement from six heart centers. In each case, the devices were calibrated for individual patients early postoperatively. Taking control measurements, the device calculated deviations from calibrated frequency and amplitude ranges using a fast Fourier transform and sent a warning message to the display if any relevant changes occurred. This message was taken as cause to determine morphologic correlations of the disturbance using imaging techniques, for example, echocardiography or fluoroscopy (observation period, 550.7 patient-years; median application period, 15.1 months). RESULTS: Twenty-five true positive, 1 false positive, and no false negative measurements occurred. Eleven patients (44%) received thrombolytic therapy with streptokinase or recombinant tissue-type plasminogen activator, resulting in normalization of the digital frequency analysis. In 4 patients (16%), international normalized ratio levels were adjusted under temporary heparinization therapy. Normalization of the valve sounds were achieved within 3 to 7 days. Seven patients (28%) required reoperation despite lysis therapy. Sensitivity was 100% during the observation period; specificity was 99.8%. CONCLUSIONS: Digital frequency analysis represents a reasonable addition to the monitoring of patients after artificial heart valve replacement to detect prosthesis-related complications early on.