Implementation of a neonatal platelet transfusion guideline to reduce non-indicated transfusions using a quality improvement framework

Lowering platelet-count threshold for transfusion in preterm neonates decreases the number of transfusions without increasing severe hemorrhage events

Thrombocytopenia is common in preterm neonates and can be associated with hemorrhage. Most platelet transfusions are prophylactic. Previously, higher platelet-count thresholds were recommended for neonates, but this recommendation has been questioned in recent studies. In the PlaNeT2 trial, mortality and serious bleeding were more frequent in neonates with the highest platelet-count threshold than in others. Following this trial, we changed our platelet transfusion practice by lowering the platelet-count threshold for prophylactic transfusion from 50,000 to 25,000/mm3. We conducted a before-after retrospective cohort study to quantify the frequency of platelet transfusions and assess the new protocol by analyzing death and serious hemorrhage events. This retrospective monocentric study included neonates born before 37 weeks of gestation with platelet count < 150,000/mm3 during the 2 years preceding the new platelet transfusion protocol (high prophylactic transfusion threshold, 50,000/mm3) and during the 2 years after the new platelet transfusion protocol (low prophylactic transfusion threshold, 25,000/mm3). The primary outcome was the proportion of neonates receiving at least one platelet transfusion in both groups. We also compared the proportion of deaths and severe hemorrhage events. A total of 707 neonates with thrombocytopenia were identified. In the high-threshold group, 99/360 (27.5%) received at least one platelet transfusion as compared with 56/347 (16.1%) in the low-threshold group (p < 0.001). The groups did not differ in proportion of deaths or severe hemorrhage events.

CONCLUSIONS

A reduced platelet-count threshold for transfusion allowed for a significant reduction in the number of platelet transfusions without increasing severe hemorrhage events.

WHAT IS KNOWN

• A recent randomized trial suggested that restrictive platelet-count thresholds for platelet transfusion could be beneficial for preterm neonates.

WHAT IS NEW

• On lowering the platelet-count threshold for transfusion from 50,000 to 25,000/mm3, the number of transfusions significantly decreased without increasing severe hemorrhage events in a neonatal intensive care unit.

Impact of restrictive platelet transfusion strategies on transfusion rates: A cohort study in very preterm infants

BACKGROUND

Evidence supports a restrictive platelet transfusion threshold in preterm neonates. We aimed to describe the effect of implementing this threshold on transfusion rates.

STUDY DESIGN AND METHODS

This retrospective observational cohort study included all very preterm infants (born <32 weeks' gestation) admitted to a neonatal intensive care unit between 2004 and 2022, divided into three epochs. Platelet transfusion thresholds changed from 30 × 109/L for stable neonates and 50 × 109/L for unstable neonates (January 2004 to December 2009) to 20 × 109/L for stable neonates and 50 × 109/L for unstable neonates (January 2010 to June 2019) to 25 × 109/L for non-bleeding neonates and 50 × 109/L for neonates with major bleeding (July 2019 to July 2022). The primary outcome was the percentage of transfused neonates in each epoch. Secondary outcomes included the median number of transfusions per neonate, the percentage of transfusions given above 25 or 50 × 109/L, and major bleeding and mortality rates.

RESULTS

The percentage of neonates transfused was 12.2% (115/939), 5.8% (96/1660), and 4.8% (25/525) in Epoch I, II, and III, respectively (p < .001), a relative reduction of 61%. The median number of transfusions per transfused neonate was 2.0 (interquartile range [IQR]: 1.0-3.0) in Epoch I, and 1.0 (IQR: 1.0-2.0) in subsequent Epochs (p = .04). The percentage of infants receiving at least one transfusion above 50 × 109/L in Epoch I, II, and III was 51.3% (59/115), 17.7% (17/96), and 20.0% (5/25; p < .001). Mortality and bleeding rates did not significantly differ between epochs.

DISCUSSION

Implementation of restrictive platelet guidelines led to reduction of the rate and number of platelet transfusions.

Implementing evidence-based restrictive neonatal intensive care unit platelet transfusion guidelines

Platelet transfusions are life-saving treatments for specific populations of neonates. However, recent evidence indicates that liberal prophylactic platelet transfusion practices cause harm to premature neonates. New efforts to better balance benefits and risks are leading to the adoption of more restrictive platelet transfusion guidelines in neonatal intensive care units (NICU). Although restrictive guidelines have the potential to improve outcomes, implementation barriers exist. We postulate that as neonatologists become more familiar with the data on the harm of liberal platelet transfusions, enthusiasm for restrictive guidelines will increase and barriers to implementation will decrease. Thus, we focused this educational review on; (1) the adverse effects of platelet transfusions to neonates, (2) awareness of platelet transfusion "refractoriness" in thrombocytopenic neonates and its association with poor outcomes, and (3) the impetus to find alternatives to transfusing platelets from adult donors to NICU patients.

Neonatal Intensive Care Unit Patients Receiving More Than 25 Platelet Transfusions

OBJECTIVE

A few patients in neonatal intensive care units (NICU) receive numerous platelet transfusions. These patients can become refractory, defined as transfusions of ≥10 mL/kg failing to increase the platelet count by at least 5,000/µL. Causes of, and best treatments for, platelet transfusion refractoriness in neonates have not been defined.

STUDY DESIGN

Multi-NICU multiyear retrospective analysis of neonates receiving >25 platelet transfusions.

RESULTS

Eight neonates received 29 to 52 platelet transfusions. All eight were blood group O. Five had sepsis, four were very small for gestational age, four had bowel resections, two Noonan syndrome, two had cytomegalovirus infection. All eight had some (19-73%) refractory transfusions. Many (2-69%) of the transfusions were ordered when the platelet count was >50,000/µL. Higher posttransfusion counts occurred after ABO-identical transfusions (p = 0.026). Three of the eight had late NICU deaths related to respiratory failure; all five survivors had severe bronchopulmonary dysplasia requiring tracheostomy for prolonged ventilator management.

CONCLUSION

Neonates who are high users of platelet transfusions appear to be at high risk for poor outcomes, especially respiratory failure. Future studies will examine whether group O neonates are more likely to develop refractoriness and whether certain neonates would have a higher magnitude of posttransfusion rise if they received ABO-identical donor platelets.

KEY POINTS

· Many of the platelet transfusions given in the NICU are given to a small subset of patients.. · Refractoriness to platelet transfusions is common among these very high recipients.. · Neonates who are high users of platelet transfusions appear to be at high risk for poor outcomes..

Patient Blood Management in Neonates

Patient blood management (PBM) is an evidence-based care package to improve patient outcomes by optimizing a patient's blood, minimizing blood loss, and the effective management and, when appropriate, the tolerance of anemia. It is relatively well-developed in adult medicine and remains in its infancy in neonatology. This review explores why evidence-based guidelines are insufficient, discusses the variations in neonatal transfusion practice and why this matters, and provides the key updates in neonatal transfusion practice. The authors give examples of a successful neonatal PBM program and single-center projects.

Prophylactic Platelet Transfusions: Why Less Is More

Preterm neonates are a highly transfused patient group, with platelet transfusions being the second most transfused cellular blood component. Historically, however, evidence to inform optimal platelet transfusion practice has been limited. In pediatrics, much of the evidence has been inferred from studies in adult patients, although neonatologists have generally applied more cautious and liberal platelet transfusion thresholds to mitigate the complications of intraventricular hemorrhage. A total of three randomized controlled trials have now been published comparing different platelet transfusion strategies in neonates.

The Usability of Platelet Mass Index Thresholds to Assess the Repeated Platelet Transfusion Requirements in Neonates

To evaluate the usability of platelet mass index (PMI) thresholds to assess the repeated platelet transfusion requirements in neonates who have received transfusion within the previous six days. This is a retrospective cross-sectional study conducted with neonates who received prophylactic platelet transfusion. The PMI was calculated as platelet count (× 1000/mm3) × mean platelet volume (MPV) (fL). Platelet transfusions were divided into two groups as first (Group 1) and repeated transfusions (Group 2). The increment and percentage of increment in platelet counts, MPV and PMI after transfusion were compared between the two groups. The amounts of changes were calculated as: (Post-transfusion) - (Pre-transfusion values). The percentages of changes were calculated as: ([Post-transfusion - Pre-transfusion values]/Pre-tansfusion values) × 100. Eighty three platelet transfusions were analyzed in 28 neonates. The median gestational age and birth weight were 34.5 (26-37) weeks, and 2225 (752.5-2937.5) grams, respectively. There were 20 (24.1%) transfusions in Group 1, and 63 (75.9%) transfusions in Group 2. There were no differences in the amounts of changes in platelet counts, MPV and PMI between the groups (p > 0.05). When the percentages of changes were analyzed, it was found that the platelet counts and PMI in Group 1 increased to a greater extent compared to Group 2 (p = 0.026, p = 0.039, respectively), but no significant difference was found in MPV between the groups (p = 0.081). The lower percentage of change in PMI in Group 2 was associated with the lower percentage of change in platelet counts. Being transfused with adult platelets did not affect platelet volume of the neonates. Therefore, PMI thresholds can be used in neonates with a history of platelet transfusion.

Platelet Transfusions in a Multi-NICU Healthcare Organization Before And After Publication of the PlaNeT-2 Clinical Trial

OBJECTIVES

To evaluate whether implementing more restrictive NICU platelet transfusion guidelines following the PlaNeT-2 randomized controlled trial (transfusion threshold changed from 50,000/μL to 25,000/μL for most neonates) was associated with fewer NICU patients receiving a platelet transfusion, without adversely affecting outcomes.

STUDY DESIGN

Multi-NICU retrospective analysis of platelet transfusions, patient characteristics, and outcomes during three years before vs. three years after revising system-wide guidelines.

RESULTS

During the first period, 130 neonates received one or more platelet transfusions; this fell to 106 during the second. The transfusion rate was 15.9/1000 NICU admissions in the first period vs. 12.9 in the second (p=0.106). During the second period, a smaller proportion of transfusions was administered when the platelet count was in the 50,000 - 100,000/μL range (p=0.017), and a larger proportion when it was <25,000/μL (p=0.083). We also saw a fall in the platelet counts that preceded the order for transfusion from 43,100/μL to 38,000/μL (p=0.044). The incidence of adverse outcomes did not change.

CONCLUSIONS

Changing platelet transfusion guidelines in a multi-NICU network to a more restrictive practice was not associated with a significant reduction in number of neonates receiving a platelet transfusion. The guideline implementation was associated with a reduction in the mean platelet count triggering a transfusion. We speculate that further reductions in platelet transfusions can safely occur with additional education and accountability tracking.

Platelet transfusion practice pattern before and after implementation of a local restrictive transfusion protocol in a neonatal intensive care unit

BACKGROUND

Following the results of the PlaNeT-2 randomized controlled trial showing decreased morbidity and mortality in neonates transfused at a threshold of 25 versus 50 × 10⁹ platelets/L, a protocol supporting restrictive platelet transfusions was established in 2019 at the Sainte-Justine Hospital neonatal intensive care unit (NICU). This study aimed to: (1) determine the impact of a local restrictive transfusion protocol on the number of platelet transfusions and donor exposure; (2) compare platelet-transfusion determinants and justifications before and after its implementation.

STUDY DESIGN AND METHODS

Prospective observational cohort chart-review study comparing all neonates consecutively admitted to the NICU during two 5-months periods: 2013 (before; N = 401) versus 2021 (after; N = 402). Possible determinants were assessed via logistic regressions and justifications via a questionnaire.

RESULTS

Mean (± standard deviation) gestational age and birth weight were 34.9 ± 4.2 weeks and 2.5 ± 1.0 kg, respectively. In 2021, 5.0% were platelet-transfused versus 9.2% in 2013 (p = .027). Platelet transfusions decreased from a mean of 2.6 ± 1.7 in 2013 to 1.4 ± 0.7 in 2021 (p = .045). Adherence to protocol thresholds was of 70%. After protocol implementation, no infant received ≥4 platelet transfusions nor was exposed to ≥4 donors, compared to 29.7% and 21.6%, respectively, in 2013. Platelet transfusion justifications and determinants remained similar, except for severe intraventricular hemorrhage being an additional determinant in 2021.

DISCUSSION

Restrictive local transfusion thresholds in a NICU decreased the proportion of platelet-transfused neonates by 46% and reduced donor exposure in transfused patients.

Neonatal platelet transfusions: New evidence and the challenges of translating evidence-based recommendations into clinical practice

Platelet transfusions are a common intervention for thrombocytopenia. Although the main reason for transfusing platelets is to improve hemostasis, platelets have many additional physiological roles, including interactions with immune pathways. Much of the evidence base for safe and effective transfusions has been informed by randomized trials in adult patients with hematological malignancies. Only three randomized trials have been conducted in sick neonates. These trials have indicated evidence of harm, including a significantly higher rate of death or major bleeding within 28 days after randomization for the largest trial, which enrolled 660 infants. The overall research indicates limited effectiveness of platelet transfusions to reduce bleeding risk. It is important that the results of trials are implemented into practice, but uptake of research findings into neonatal medicine remains inconsistent, as for many areas of health care. There is a need to establish which potential implementation strategies (cost-) efficiently enact change, such as audit and feedback, automated reminder systems for ordering transfusions, and use of opinion leaders. Research is exploring potential mechanisms underlying the lack of effectiveness of platelet transfusions and the increased bleeding and mortality observed in neonatal randomized trials. One potential mechanism concerns the roles of platelets to promote excessive angiogenic signals during a vulnerable period of brain development. A further hypothesis explores the effects of transfusing "adult" platelets into "neonatal" thrombocytopenic blood on primary hemostasis and immune responses.