Prosthetic Valve Thrombosis

Abstract & Commentary

By Michael H. Crawford, MD, Editor

Source: Ozkan M, et al. Comparison of different TEE-guided thrombolytic regimens for prosthetic valve thrombosis. JACC Cardiovasc Imaging 2013; 6:206-216.

The use of thrombolysis for prosthetic valve thrombosis (PVT) is controversial and various guidelines rate it differently. However, since PVT carries a significant mortality risk, more information on proposed therapies would be valuable. Thus, this single-center experience with thrombolysis for PVT is of value. Over a 16-year period at an academic hospital in Istanbul, Turkey, five different thrombolytic regimens were used sequentially. This experience was analyzed to determine the most effective regimen. Because of mortalities approaching 20% with surgery for PVT, thrombolysis was first-line treatment for almost all patients, with surgery reserved for those with major contraindications for thrombolytics or those in whom it failed. Their experience started in 1993 (3-hour infusion of streptokinase) and this protocol was changed in 1997 to a 24-hour infusion when the complication rate was noted to be the same as surgery. In 2001, tissue plasminogen activator (tPA) was used in three sequential protocols based on the results of the prior protocol. Because of concern for bleeding, heparin was not administered with thrombolysis, only afterward. In 182 consecutive patients, 220 episodes of PVT were treated. The overall success rate was 83% and was highest for the last protocol (85.5%), but was not significantly different across all five. The complication rate was significantly lower for the final protocol (10.5% vs 37.5%) as compared to the first protocol (P < 0.05), and there were no deaths during use of the final protocol vs six deaths (17%) using the four prior regimens. Intracranial hemorrhage occurred in seven patients (two died) and seven had ischemic strokes (four died). Thus, all six deaths (2.17%) were related to cerebral complications. By multivariate analysis, prior stroke/TIA was predictive of death (odds ratio, 3.47; 95% CI, 1.32-9.11; P = 0.01). The final protocol was a 6-hour infusion of 25 mg of tPA without a bolus repeated once 24 hours later (up to 6 times if necessary). The authors concluded that a low-dose, slow infusion of tPA without a bolus — and repeated if necessary — was an effective and safe thrombolytic regimen in patients with PVT.


PVT is an infrequent but urgent problem that doesn’t lend itself well to a randomized trial. Also, the guidelines of various organizations are not consistent. Although most agree that if surgery is very high risk or if the PVT is right sided where emboli can do less harm, thrombolysis is worth trying. Some believe you can treat small left-sided thrombi with thrombolysis and some believe that only obstructive thrombi should be considered for thrombolysis. In addition, there is no agreement on what thrombolytic regimen should be used. Thus, this large, systematic experience at one center is of interest.

In this study, about half the patients had nonobstructive thrombi by transesophageal echocardiography, which was done in all on entry and after therapy was given. Their contraindications for thrombolysis included asymptomatic nonobstructive PVT without evidence of systemic emboli and diameter < 10 mm. Also, they excluded anyone < 3 weeks from an ischemic stroke. Their criteria for successful thrombolysis were: Doppler echo resolution of the increased gradient, symptomatic improvement, and a decrease in thrombus diameter of > 75%. If all three were met, it was considered complete resolution and < 3 but > 0 was considered partial resolution; both were included in their 85% final success rate. Why 15% were resistant to therapy is unknown, but they point out that pannus is difficult to distinguish from thrombus by echo.

Their final protocol complication rate of 5% with no deaths is similar to the results of using thrombolytics in acute myocardial infarction. There was no surgical group reported but these morbidity/mortality rates are lower than those reported by others. However, surgical patients are probably a higher risk group in general. In comparing the various protocols, it becomes clear that the secret to their success is no bolus, low doses infused slowly, and no heparin until the thrombolytic is in. This approach probably enhances safety because it produces less emboli. The downside is having to repeat the infusion in a significant number of patients. Their final protocol required more than three infusions for success in 21 of the 124 patients treated (17%). Whether the same results or better ones could be obtained with some of the newer thrombolytics is unknown, but since tPA is still available, I would tend to go with their regimen until more data are forthcoming.