The diagnosis of deep vein thrombosis (DVT) in pregnant women remains a challenge for physicians. The 'LEFt' clinical decision rule was recently derived, that might help in estimating the pretest probability of DVT during pregnancy. Our aim was to externally validate the LEFt clinical decision rule among pregnant women included in the OPTIMEV study. The OPTIMEV study is a diagnostic and epidemiologic study that included patients with suspected VTE between November 2004 and January 2006. All patients underwent standardized clinical data collection, and a bilateral whole-leg venous ultrasonography. A 3-month follow-up was performed in all patients with confirmed VTE, and in a randomly selected subsample of patients with negative diagnostic workup. Of the 8,256 included patients, 96 were pregnant women. A DVT was diagnosed at CUS in 9 women (9.4%). The LEFtscore was computed in all but 7 women with missing values: one point in case of left ('L') leg suspicion, one point for edema ('E') and one point if the suspicion occurred during the first trimester ('Ft') of pregnancy. Prevalence of confirmed DVT was as follows: 1/30 (3%) in women with no LEFt criteria, 3/35 (9%), 2/20 (10%), and 3/4 (75%) in women with 1, 2 and 3 points, respectively. Our results confirm the ability of the LEFt rule to estimate the pretest probability of DVT. Future studies are required to prospectively validate these findings and to define the role of the rule in a diagnostic algorithm for DVT during pregnancy.