Objective: To determine retrospectively mean arterial pressure (MAP) for stable and unstable concordant and discordant very low birth weight (VLBW: 801 to 1500 g) twins during the first 24 hours of life. Background: Morbidity and mortality are much higher for extremely low birth weight (ELBW ≤ 800g) than for VLBW twins, Recently, we reported MAP trends and reference values in concordant and disconcordant ELBW twins. No comparable information is available for VLBW infants. Design: Retrospective cohort study. Methods: We studied 48 sets of concordant and 40 sets of discordant (birth weight difference ≥ 20%) consecutively born VLBW twins. Stable patients were defined as having umbilical cord hemoglobin ≥ 14 g/dl, nonacidotic blood gases, never treated for hypotension and survived at least 7 days. MAPs (Torr) were measured by oscillometry in 3163 and by transducer via umbilical artery in 2028 instances. Results: Concordant and discordant twins were similar in demographics, history of twin-twin transfusion (TTTX), antenatal steroids, chorioamnionitis, preeclampsia, cesarean delivery, cord hemoglobin, normal head ultrasounds or I to II intracranial hemorrhage (97 and 99%) and neonatal mortality (4 and 5%), but were different in incidence of preterm labor (83 and 58%), birth weight (1227 and 1509 g) and gestational age (GA) (30 and 32 weeks). In all, 66 (69%) concordant twins and 61 (96%) discordant twins were stable. Stable concordant twins, whether small or large, had comparable MAP on admission that increased to 24 hours. Twins of ≤ 32 weeks GA had lower MAP throughout than those of ≥ 33 weeks GA. Although their mean birth weights were similar (1262 and 1274 g), 23 stable concordant in had significantly higher MAP than 43 concordant females. Stable discordant twins were divided into 31 small (1241 g) and 30 large (1845 g); their MAPs were different (p<0.05): 35 and 39 (admission), 35 and 39 (1 hour), 36 and 46 (6 hours), 38 and 41 (12 hours), 40 and 41 (18 bours) and 42 and 42 (24 hours) Torr. In all, 88% of small discordant twins were IUGR and 91% of large discordant twins had normal growth. TTTX syndrome occured in 12 monochoroinic sets. Nine of 12 donors were IUGR while 10 of 12 recipients had normal growth. Four of 12 donors had grades III to IV intracranial hemorrhage, eight donors and all 12 recipients had normal ultrasounds. Although their cord hemoglobin levels were similar, donor and recipient MAPs were higher than in any other group and, opposite to concordant and discordant twins, their values decreased from birth to 24 hours. Conclusions: In stable concordant, stable discordant, and small and large discordant twins, MAP correlates with birth weight, GA and postnatal age, and increases during the first 24 hours. In recipient and donor twin-twin transfusion infants, MAP is higher throughout and declines over time.