# Over‐Transfusion and Unnecessary Transfusion Following Post‐Partum Haemorrhage at Te Toka Tumai Auckland Hospital

## TL;DR

This study examines how often blood transfusions were overused or unnecessary in postpartum hemorrhage patients at a New Zealand hospital.

## Contribution

The study introduces a method to assess transfusion appropriateness by adjusting pre-discharge hemoglobin levels per unit of blood given.

## Key findings

- Approximately 46.8% of transfused patients were over- or unnecessarily transfused.
- Over-transfusion rates were high in both acute and non-acute settings.
- Transfusion appropriateness did not vary significantly by ethnicity.

## Abstract

Blood transfusion is an important treatment for obstetric haemorrhage. Transfusion also engenders significant short and long‐term risks. Ensuring blood products are only given when necessary is a priority in improving outcomes.

To describe the population transfused at a single unit in New Zealand and identify the proportion of patients over and unnecessarily transfused via adjustment of haemoglobin per unit of blood given. To assess whether the rate of inappropriate transfusion was modified by demographic and treatment characteristics.

A retrospective cohort study inclusive of all people who gave birth from 20 weeks between 2018 and 2021 at one hospital was assembled. People who were administered red blood cell‐containing products were identified. The pre‐discharge haemoglobin was adjusted per unit of blood given with patients being considered over or unnecessarily transfused at a pre‐discharge haemoglobin of ≥ 90 mg/dL.

The transfused population comprised 694/25 915 pregnancies (2.7% of the cohort). Appropriate transfusion (pre‐discharge haemoglobin < 90) occurred in 332/694 (47.8%) people. There were 325 (46.8%) patients who were over‐ or unnecessarily transfused. There was no difference in appropriateness of transfusion for any ethnicity compared to Māori, our referent group. Over‐transfusion rates did not differ and were high in both acute (53%) and non‐acute (45%) settings.

The rate of transfusion for obstetric haemorrhage was 2.7% in our study population. Approximately half of people receiving blood received either too many units or did not require a transfusion.