Abstract
This cohort study examines the use of closure time following stimulation with collagen and adenosine diphosphate vs the platelet count as a marker of bleeding in premature neonates with thrombocytopenia.
Platelet transfusions are frequently given to neonates with platelet counts (PCs) below an arbitrary trigger,1,2 but studies have shown a poor correlation between low PC and bleeding,2,3 highlighting the need for better tests to identify infants at risk.
Neonates have a unique primary hemostatic system, characterized by hypofunctional platelets counterbalanced by factors that enhance clotting (high hematocrit, mean corpuscular volume, and von Willebrand factor concentrations).4 Thus, we hypothesized that a global test of primary hemostasis, the Platelet Function Analyzer-100 (PFA-100), an in vitro equivalent to the bleeding time, would be a better marker of moderate to severe bleeding among preterm neonates with thrombocytopenia than the PC. The PFA-100 (Dade Behring) measures the time it takes for blood to occlude an aperture (closure time [CT]) following stimulation with collagen and epinephrine (CT-Epi) or collagen and adenosine diphosphate (CT-ADP). Based on a previous finding that the CT-ADP was better correlated with the PC than the CT-Epi in neonates,5 we focused on the CT-ADP.
Methods
This was a prospective, longitudinal study that enrolled patients with gestational age (GA) of 33 weeks or less and PC less than 100 × 103/μL (to convert to ×109/L, multiply by 1) in 2 neonatal intensive care units (Karolinska Institutet, Stockholm, Sweden; and Levine Children’s Hospital, Charlotte, North Carolina) between May 1, 2015, and September 30, 2017. Data analysis was conducted from October 16, 2017, to April 30, 2018. The study was approved by the Swedish Central Ethical Review Board and by the Carolinas Healthcare System Institutional Review Board, and signed written informed consent was obtained from parents/guardians. Study participants and their parents/guardians did not receive financial compensation.
The CT-ADPs and degree of bleeding were measured at the same time as PCs on 3 consecutive days. Bleeding was quantified using the validated Neonatal Bleeding Assessment Tool,6 which provides a bleeding score (BS) ranging from 0 (no bleeding) to 4 (major bleeding).
Repeated-measures regression analysis was performed to examine the association between CT-ADP and PC and simultaneous BS. A 2-sided P value <.05 was the critical level for statistical significance. Covariates included sex, GA, postnatal age, small for GA, severity of illness (SNAP-II [lowest level of severity, 0; highest level, 115]7), necrotizing enterocolitis, sepsis, hematocrit, transfusions, and medications (penicillin, cephalosporins, ibuprofen, indomethacin, acetaminophen). Across 3 categories of BS, Fisher exact test was used to compare fractions with maximum CT-ADP and Kruskal-Wallis test to compare the skewed distributions of CT-ADP and PC. SAS software, version 9.4 (SAS Institute), was used for analysis.
Results
Patient characteristics are displayed in Table 1. Repeated-measures regression analysis showed the CT-ADP being associated with BS (regression coefficient [SE], 0.16 [0.05] points/60 seconds, P = .003), whereas the PC was not (−0.05 [0.04] points/10 × 103/μL; P = .20). When both CT-ADP and PC were included in the model, the PC showed no association with BS (0.01 [0.04] points/10 × 103/μL; P = .73), but CT-ADP remained correlated (0.16 [0.05] points/60 seconds; P = .004). The CT-ADP also remained robustly associated with BS after controlling for clinical covariates. This association was strongest in infants with GA less than 27 weeks (0.28 [0.06] points/60 seconds; P < .001), and lost statistical significance in infants 27 weeks’ GA or older.
Table 1. Patient Characteristics.
| Variable | Mean (SD) |
|---|---|
| No. of patients | 76 |
| Gestational age at birth, wk | 26.1 (2.4) |
| Birth weight, g | 777 (310) |
| Small for gestational age, No. (%) | 27 (36) |
| Postnatal age at inclusion, median (IQR), d | 4 (3-14.5) |
| Postmenstrual age at inclusion, wk | 27.9 (3.2) |
| Sex, male, No. (%) | 43 (57) |
| Platelet count at inclusion, ×103/μL | 67 (27) |
| Hematocrit during the period, % | 41 (5) |
| CT-ADP during the study, s | 179 (88) |
| SNAP-II score during the study, median (IQR)a | 5 (5-16) |
| NEC during study, No. (%) | 25 (33) |
| Sepsis during study, No. (%) | 39 (51) |
| Received penicillin during study, No. (%) | 37 (49) |
Abbreviations: CT-ADP, closure time-adenosine diphosphate; IQR, interquartile range; NEC, necrotizing enterocolitis.
SI conversion factors: To convert hematocrit to proportion of 1.0, multiply by 0.01; platelet count to ×109/L, multiply by 1.
Lowest level of severity, 0; highest level, 115.
Subsequent analyses focused on 54 infants less than 27 weeks’ GA. Bleeding episodes (grades 1-4) occurred more frequently in those infants than in infants 27 weeks’ GA or older (P = .01), and more were grade 2 to 4: 34 of 99 (34.3%) vs 1 of 45 (2.2%) (P < .001). Infants younger than 27 weeks’ GA also had longer median CT-ADPs than infants 27 weeks’ GA or older (196 vs 136 seconds; P = .02), and a higher percentage of samples at the maximum 300 seconds (36% vs 20%; P = .05).
In neonates younger than 27 weeks’ GA with PC less than 100 × 103/μL, BSs were associated with CT-ADPs (+0.28 [0.06] points/60 seconds; P < .001) but not with PCs (−0.04 [0.04] points/10 × 103/μL; P = .33). As in the entire cohort, the association of BS with CT-ADP was not attenuated by adjustment for covariates. Breakdown of the BS into grades 0 to 1 (none or mild), grade 2 (moderate), and grades 3 to 4 (major to severe) also revealed a significant correlation between median CT-ADP and bleeding severity (median [range] 155 [70-300], 271 [69-300], 300 [183-300] points respectively; χ2 = 10.2, 2 df; P = .006), but not between the median PC and BS (67 [13-98], 56 [16-95], 56 [25-93] × 103/μL; χ2 = 3.7, 2 df; P = .15). Seven of the 8 infants (88%) with the most severe bleeding had CT-ADP values of 300 seconds (Table 2).
Table 2. Correlation Between CT-ADP or PC and Bleeding Scores.
| Bleeding Scorea | Fraction With Maximum CT-ADP, No./Total No. (%) | Median (Range) | |
|---|---|---|---|
| CT-ADP, s | PC, 103/μL | ||
| 0-1 | 18/65 (28) | 155 (70-300) | 67 (13-98) |
| 2 | 11/26 (42) | 271 (69-300) | 56 (16-95) |
| 3-4 | 7/8 (88) | 300 (183-300) | 56 (25-93) |
| P value | .003b | .006c | .15c |
Abbreviations: CT-ADP, closure time-adenosine diphosphate; PC, platelet count.
SI conversion factor: To convert PC to ×109/L, multiply by 1.
Score ranges from 0 (no bleeding) to 4 (major bleeding).
Fisher exact test.
Kruskal-Wallis test.
Discussion
This study suggests that the CT-ADP, but not the PC, is associated with the BS among thrombocytopenic neonates younger than 27 weeks’ GA. These extremely preterm neonates also had a higher frequency of grade 2 to 4 bleeding than more mature infants. Although the reasons for their high bleeding risk are multifactorial and likely include in vivo factors, 88% of infants with grades 3 to 4 bleeding had maximally prolonged CT-ADPs, suggesting that inadequate hemostasis might contribute to severe bleeding. We acknowledge that our study has limitations. First, we had a small number of infants with PC less than 50 × 103/μL, due to platelet transfusion practices in our NICUs. We also recognize that it would have been beneficial to obtain more CT-ADP measurements in each infant. However, the number was limited to 3 samples because of concerns that excessive blood loss or phlebotomies would lead to reduced enrollment and anemia. Finally, the CT-ADP does not incorporate critical in vivo determinants of a patient’s bleeding risk, such as vessel wall integrity, inflammation, and hemodynamic status. Our findings could lead to better approaches to the management of thrombocytopenia in preterm neonates than relying on PCs alone.
References
- 1.Cremer M, Sola-Visner M, Roll S, et al. . Platelet transfusions in neonates: practices in the United States vary significantly from those in Austria, Germany, and Switzerland. Transfusion. 2011;51(12):2634-2641. doi: 10.1111/j.1537-2995.2011.03208.x [DOI] [PubMed] [Google Scholar]
- 2.Stanworth SJ, Clarke P, Watts T, et al. ; Platelets and Neonatal Transfusion Study Group . Prospective, observational study of outcomes in neonates with severe thrombocytopenia. Pediatrics. 2009;124(5):e826-e834. doi: 10.1542/peds.2009-0332 [DOI] [PubMed] [Google Scholar]
- 3.von Lindern JS, van den Bruele T, Lopriore E, Walther FJ. Thrombocytopenia in neonates and the risk of intraventricular hemorrhage: a retrospective cohort study. BMC Pediatr. 2011;11:16. doi: 10.1186/1471-2431-11-16 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Ferrer-Marin F, Stanworth S, Josephson C, Sola-Visner M. Distinct differences in platelet production and function between neonates and adults: implications for platelet transfusion practice. Transfusion. 2013;53(11):2814-2821. doi: 10.1111/trf.12343 [DOI] [PubMed] [Google Scholar]
- 5.Deschmann E, Sola-Visner M, Saxonhouse MA. Primary hemostasis in neonates with thrombocytopenia. J Pediatr. 2014;164(1):167-172. doi: 10.1016/j.jpeds.2013.08.037 [DOI] [PubMed] [Google Scholar]
- 6.Venkatesh V, Curley A, Khan R, et al. . A novel approach to standardised recording of bleeding in a high risk neonatal population. Arch Dis Child Fetal Neonatal Ed. 2013;98(3):F260-F263. doi: 10.1136/archdischild-2012-302443 [DOI] [PubMed] [Google Scholar]
- 7.Richardson DK, Corcoran JD, Escobar GJ, Lee SK. SNAP-II and SNAPPE-II: Simplified newborn illness severity and mortality risk scores. J Pediatr. 2001;138(1):92-100. doi: 10.1067/mpd.2001.109608 [DOI] [PubMed] [Google Scholar]