Coagulation values in extreme premature infants.

We read with interest the recent publication by Christensen and colleagues. Standard coagulation test results vary with gestational age. Consequently, gestationspecific reference ranges for these variables are required. The paucity of published reference values for coagulation variables in extremely premature neonates represents a challenge for neonatologists managing infants at greatest risk of intraventricular hemorrhage and other bleeding complications. The data presented by Christensen and coworkers pertaining to infants of less than 28 weeks’ gestation are helpful, but were based on a small sample size of 24 infants. Moreover, tests were performed on cord blood. An equivalency comparison of samples obtained from cord and peripheral blood was limited to full blood count variables. Fetal plasma clotting times vary with maturation and do not necessarily correlate with published reference ranges for delivered premature infants at equivalent gestations. Physiologic changes and medical interventions (including vitamin K administration) after delivery appear to influence these variables. We have recently published data pertaining to the extremely premature neonatal population (n = 183) born at less than 27 weeks’ gestation. In this large patient cohort, we observed longer clotting times than those reported by Christensen and coworkers in a similar but smaller patient population. In our cohort, median (range 5th-95th percentile) prothrombin time (PT) and activated partial thromboplastin time (APTT) values were 20.2 (14.8-32.6) and 67.4 (43.3-130.2) seconds, respectively. In our population the coagulation values were obtained using combination of PT-S (20042008) and PT-Recombinant PlasTin 2G reagent (20082010) for PT values and APTT-SP reagent (2004-2008) and lupus-sensitive silica-based APTT reagent SytnASIL (20082010) for APTT values compared with Diagnostica Stago kits used by Christensen and colleagues. Of particular note, the blood source differed between the two studies. Christensen and colleagues report results obtained from plasma prepared from cord blood. In contrast, blood was obtained from peripheral catheters in our study, before instillation of heparinized saline. The degree to which coagulation variables derived from cord blood correlate with those obtained from peripheral blood in newly delivered extremely premature infants remains unclear. Coagulation values are dynamic and vary with gestational and postnatal age and other environmental stressors and therefore the different sources of blood may account for some of the differences in values observed. We aimed to pragmatically evaluate coagulation test results in a large cohort of infants born at less than 27 weeks in a retrospective study design. Until recently, interpretation of coagulation test results in this patient population was virtually impossible. We hope that the data described herein and future large prospective studies will further define reference ranges for coagulation variables in this high-risk patient cohort and ultimately guide clinical decision making in extremely premature infants.