OBJECTIVE: To determine if absent end-diastolic flow (AEDF) in the umbilical artery (UA) has a longer latency in monochorionic (MC) twin fetuses compared to singleton or dichorionic twin (DC) fetuses. METHODS: One hundred and eight pregnancies with a fetus with AEDF were reviewed: 47 MC and 17 DC twin pregnancies and 44 singletons. Because twin-twin transfusion syndrome (TTTS) is a potential confounder when determining latency, subgroup analysis was also performed on the 21 MC affected pregnancies without TTTS. Latency of AEDF (in days) was defined as the difference between the gestational age at diagnosis of AEDF and gestational age at delivery or intrauterine death. RESULTS: Latency was similar in MC twins (median, 39 days) and DC twins (30 days) but longer compared to singletons (11 days; P = 0.0001). After excluding pregnancies with TTTS, latency in non-TTTS MC twins (54 days) was longer than in both singletons and DC twins. This was due to an earlier gestational age at AEDF in non-TTTS MC twins of 20 weeks compared to 27 weeks in both singleton and DC twins because median gestational age at delivery was similar in MC twins, DC twins and singletons. CONCLUSIONS: The latency period of UA AEDF is longer in MC twins than in singletons. Our data suggest that in MC twin fetuses without TTTS, AEDF begins earlier and lasts about twice as long as in DC twin fetuses, which is consistent with placental insufficiency not being the sole factor mediating abnormal UA waveforms in MC placentation. This observation is important in counseling and managing twin pregnancies discordant for AEDF.