PURPOSE: The purpose of this study was to report the possible role of thrombocytopenia in the pathogenesis of aggressive posterior retinopathy of prematurity (APROP). METHODS: The index case described in this study showed spontaneous resolution of APROP with plus disease within 3 days of correcting thrombocytopenia and did not require laser treatment. The retrospective cohort of nine consecutive Asian Indian infants with APROP with similar stage and plus disease as the index case was studied. The mean platelet count of these infants before laser treatment was compared with 21 age- and birth weight-matched control subjects. Correlation of platelet count with APROP is discussed. RESULTS: The mean birth weight of the 10 cases with APROP was 1,030 g (+/-178 g), and the mean period of gestation was 29.4 weeks (+/-2.0 weeks). The mean platelet count of the cases was 82,870/mm3 (+/-58,702/mm3) and that of the control subjects was 178,285 +/- 57,051/mm3 (P = 0.0002). Five of the 10 cases (50%) and 1 of the 21 control subjects (4.8%) had a platelet count of <100,000/mm3 (P = 0.007). In all, 19.4% of infants had culture-proven sepsis. There was no correlation between sepsis and thrombocytopenia in our cohort (P = 0.567). CONCLUSION: The role of low platelets in the etiopathogenesis of APROP has not been previously elucidated. Our study shows that a platelet count of <100,000 was associated with severe disease. Recently, platelets have been reported to play a key role in angiogenic regulatory protein delivery. It is possible that premature infants who develop retinopathy of prematurity in the setting of low platelet counts may lack the function of either delivering the optimal level or incompletely scavenging the excess of vascular endothelial growth factor A present in APROP. The spontaneous resolution of disease in our index case with platelet correction alone needs additional studies to correlate the timing and magnitude of correction that may play a role.