Clinical Evidence for Rebound Hypercoagulability After Discontinuing Oral Anticoagulants for Venous Thromboembolism

David Keith Cundiff, MD


Medscape J Med. 2008;10(11):258 

In This Article


Out of 228 matches from the MEDLINE search, 209 studies were excluded for the following reasons:

  1. No event data were reported for more than 2 months after discontinuation of OACs: 40;

  2. Timing of VTE recurrences was not reported: 17;

  3. A surrogate rather than a clinical endpoint was used (eg, positive Doppler leg scan in an asymptomatic patient) rather than a clinical recurrence (or extension): 4;

  4. The report duplicated an included study: 7;

  5. The study was not an RCT: 17;

  6. The RCT did not involve treatment of VTE: 100;

  7. OAC treatment was not a focus of the study: 33; and

  8. The report was in a language other than English: 1.

Table 1 shows the 37 arms of 20 studies that met the criteria of having data on VTE recurrences for more than 2 months subsequent to discontinuation of OACs and the timing of VTE recurrences.[9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28]

While on OACs, 0.49% of subjects per month had VTE recurrences and 0.38% per month had major hemorrhages ( Table 2 ).

VTE recurrences in these trials were 2.62 times as frequent in the 2 months following discontinuation of OACs compared with subsequently (Figure 1). The OAC treatment duration arms of ≤ 3 months had a greater tendency for rebound VTE recurrences than treatment arms of ≥ 6 months (OR = 2.98, 95% confidence interval [CI] = 2.27-3.90 vs OR = 2.36, 95% CI = 1.85-3.02). Heterogeneity in ≤ 3 months OAC treatment duration arms exceeded that in ≥ 6 months (42% vs 3%). Without the outlier Schulman 1995 RCT with 6 weeks of VKAs after the second deep venous thromboembolism episode,[25] the rate of VTE recurrences in the 2 months following discontinuation vs from > 2 months until the end of the trial in the ≤ 3 months OAC treatment arms was similar to that in the ≥ 6 months OAC treatment arms (OR = 2.42, 95% CI = 1.73-3.39 vs OR = 2.36, 95% CI = 1.85-3.02).

Venous thromboembolism recurrences within 2 months of stopping oral anticoagulants/# months vs recurrences after 2 months/# months.

In the 12 RCTs with both shorter- and longer-duration OAC treatment arms evaluable, overall rates of VTE recurrences were significantly less with longer-duration OACs vs shorter-duration OACs (Figure 2). The Schulman 1995[25] outlier RCT accounts for this difference (OR = 2.11, 95% CI = 1.45-3.07 vs OR = 1.06, 95% CI = 0.82-1.38 for the other 11 RCTs). When major bleeding events are added to VTE recurrences (ie, total adverse events), there is no longer a significant difference favoring longer anticoagulation (Figure 3).

Overall venous thromboembolism (VTE) recurrences. VTE recurrences/n: short-duration oral anticoagulant (OAC) arm vs long-duration OAC arm.

Adverse events overall. Venous thromboembolism recurrences plus major bleeds in short- vs long-duration oral anticoagulant arms.

Omitting trials in which all subjects did not have objective tests confirming the diagnoses of VTE[13,15,19,24] did not significantly change any of the findings.

The increased VTE recurrence rate following discontinuation of ximelagatran was almost identical to the increased incidence of recurrences after stopping VKAs (In Figure 1 Schulman 2003,[27] OR = 3.06 [95% CI = 1.83-5.11] vs OR = 2.62 [95% CI = 2.19-3.14] overall).

About 2% of subjects had a VTE recurrence possibly attributable to rebound hypercoagulability in the 2 months after stopping OACs on the basis of subtracting the rate after 2 months from the rate in the first 2 months after OAC cessation (Figure 1: 1.57% [176/11,196] - 0.56% [412/73,292] = 1.01% per month; for 2 months = 2.02%). The rate of VTE recurrences possibly attributable to rebound hypercoagulability for subjects receiving ≤ 3 months (2.70%: rate in the first 2 months: 2.08% [88/4235] - rate from > 2 months until the end of the trial: 0.63% [164/26,150] = 1.35%, for 2 months = 2.70%) was less than for those receiving ≥ 6 months of OACs (1.46%: rate in the first 2 months: 1.26% [88/6961] - rate from > 2 months until the end of the trial: 0.53% [248/47,142] = 0.73%; for 2 months = 1.46%).

Funnel plots did not show asymmetry, suggestive of publication bias.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.