Prolonged bedrest during pregnancy: does the risk of deep vein thrombosis warrant the use of routine heparin prophylaxis?

This is the first study to assess the risk of clinically apparent DVT in pregnant women placed in the hospital at prolonged bedrest. The outcome is discussed with reference to the risks associated with heparin. Information, including delivery data, length of hospital stay, and discharge diagnoses were extracted from a prospectively collected computerized data bank of all deliveries that occurred over a 5.5-year period at Long Beach Memorial Women's Hospital in Long Beach, California, and at St. Joseph's Hospital in Milwaukee, Wisconsin. One group consisted of all pregnant women who had been hospitalized at prolonged antepartum bedrest, as defined by 3 weeks or more. The other group consisted of the remaining population of women whose deliveries occurred during the same time period. There were 48,525 deliveries during the study period, and 266 (0.5%) women were hospitalized at prolonged antepartum bedrest. The mean number of days in the hospital for these women was 34.6 +/- 14 (range 21-82 days). Of these women, one received prophylactic heparin for a prior history of DVT. There were no cases of DVT in the 265 women who did not receive heparin, risk = 0.0 (CI = 0.00-0.99). Of these 265 women, 234 were hospitalized up to the day of delivery. Of these 234 women, 154 (65.8%) underwent cesarean section and no case of DVT occurred in the postoperative period, risk = 0.0 (CI = 0.0-1.7). Out of the remaining 48,259 women who were not hospitalized at prolonged bedrest, there were 18 cases of clinically apparent DVT, and the longest antepartum hospitalization was 4 days. A conservative risk of complications with prophylactic heparin therapy is 1.0% or greater. Although the risk of DVT in pregnant women hospitalized at prolonged bedrest is not zero, our study indicates that it is very low (< 1.0%). Whereas a risk of DVT of at least 1.0% could warrant heparin prophylaxis, even with 265 patients at prolonged bedrest and 48,525 controls, this risk could not be demonstrated. Using a power analysis with an alpha of 0.05 and a power of 80% to demonstrate this risk, one would need 247 cases and approximately 49,000 controls, which were clearly achieved in this study. In view of the risks associated with heparin, routine antenatal prophylaxis is not recommended unless other risk factors for DVT are present.