Hemoglobin equilibration after transfusion in stable postpartum patients

Abstract

Objective

Hemoglobin levels are typically checked 4 h after transfusion to assess response, but this delay is unnecessary in nonpregnant adult patients. It remains unclear if postpartum patients need time for hemoglobin levels to stabilize due to post‐delivery pregnancy physiology. This study aims to determine if immediate reassessment of hemoglobin following transfusion is sufficient for postpartum patients.

Methods

This is a single‐center, prospective study from February to July 2025 in which serial hemoglobin assessments (immediate, 4 h, 12 h after transfusion) were completed following the transfusion of packed red blood cells (PRBCs). Exclusion criteria were the following: hemodynamic instability, >72 h from delivery, massive transfusion protocol activation, more than four units PRBC transfused, or missed lab draw timing >15 min. Power analysis determined that 23 patients would be needed for 80% power to detect a difference of 0.6 g/dL using paired t‐tests with a two‐sided p value of <0.05.

Results

Twenty‐four participants met the criteria and were enrolled. When compared to 12‐h posttransfusion hemoglobin, there was no difference between hemoglobin concentration in g/dL between the immediate (0.08 g/dL; 95% confidence interval [CI], −0.44 to 0.26) and 4‐h posttransfusion (0.10 g/dL; 95% CI, −0.41 to 0.21) values. The mean immediate rise in hemoglobin for those with one unit transfused was 0.97 g/dL (95% CI, 0.48–1.45) and for those with two units transfused was 1.99 g/dL (95% CI, 1.26–2.71).

Conclusion

Our findings indicate that hemoglobin equilibration occurs rapidly following PRBC transfusion in stable postpartum patients, and that an arbitrary 4‐h delay after the completion of transfusion before laboratory reassessment is not necessary. Earlier evaluation of response to transfusion could prevent delays in anemia management.

1. INTRODUCTION

Hemorrhage that requires transfusion is the leading cause of severe maternal morbidity in the United States, which increased 26% over the last two decades and accounted for 18% of maternal deaths in 2023 [1, 2, 3].

In general, the consensus is that 12–24 h of equilibration time is required after a major bleeding episode to stabilize clinical and laboratory findings. However, significant hemodynamic changes driven by postpartum physiology occur in obstetric patients [4] and could reasonably be expected to have various effects on the establishment of hemodynamic equilibrium.

Studies in nonpregnant adults suggest that early assessment of hemoglobin after transfusion as early as 15 min after completion of transfusion is as accurate as waiting for 4–6 h [5, 6]. One study of 39 adult general medicine patients that excluded recent bleeding episodes found no difference in paired 15 min and 24‐h hemoglobin concentrations (p = 0.84) [6]. Another study by Elizalde et al. [5] of 32 patients with upper gastrointestinal varices/ulcer bleeding within 5 days that had resolved gave two units packed red blood cells (PRBCs) to all patients who met criteria over 2 h and then rechecked hemoglobin concentrations at 15 min, 30 min, 60 min, 120 min, and 24‐h time points. They found no difference in hemoglobin between the time points (p = 0.4), and 95% of their patients had less than a 2% observed difference between 15 min and 24‐h levels. Finally, a more recent study by Karndumri et al. in 2020 had three cohorts of 20 adult orthopedic surgery patients each with planned 1‐h, 4‐h, and 24‐h posttransfusion hemoglobin assessments [7]. They found no difference in the hemoglobin rise from baseline per unit transfused between the three cohorts after controlling for confounders. Their study was limited because they did not perform serial laboratory values on all patients but instead had three separate patient cohorts per time point.

In pregnancy and the postpartum period, there is no high‐quality evidence supporting the conventional recommendation to wait 4 h after PRBC transfusion before retesting hemoglobin. Retesting hemoglobin rise after transfusion is important to gauge response to transfusion, screen for ongoing occult bleeding, and appropriately manage transfusion volume [8]. Although there is a paucity of data for postpartum patients specifically, there are reports that the expected increase in hemoglobin for postpartum patients after transfusion is 1 g/dL per unit of PRBC, regardless of body mass index (BMI) [9]. The common practice of a four‐hour delay for checking hemoglobin labs has not been studied in postpartum patients after obstetric hemorrhage requiring transfusion. As such, clinical practice can be better informed by data from this specific population.

The objective of our study was to determine if immediate posttransfusion hemoglobin tests are as similar in performance to a 4‐h delay to accurately assess final posttransfusion hemoglobin concentration in stable postpartum patients.

2. METHODS

This prospective study was conducted at an academic metropolitan tertiary maternal care facility from February to July 2025. Usual practice for our institution prior to the study was to obtain a hemoglobin assessment, usually with a complete blood count (CBC), four hours after completion of PRBC transfusion in stable postpartum patients. Quantitative blood loss was calculated at the time of delivery by weighing pads and supplies for all deliveries. During the study period, all patients received an immediate (defined as within 15 min of transfusion completion), 4‐h, and 12‐h posttransfusion CBC drawn after a transfusion of PRBC was completed. Our institution defines postpartum hemorrhage as EBL > 1000 mL for vaginal delivery or cesarean, or signs and symptoms of hypovolemia [10]. All other clinical care and hemorrhage management were performed as per usual protocols [10]. Additional evaluations, including repeat laboratory studies, clinical examinations, or imaging, were performed at the discretion of the treating physician.

The study population comprised all hospitalized pregnant and postpartum patients who experienced obstetric bleeding and received blood transfusion during or after stabilization due to signs or symptoms of hypovolemia. Patients were excluded if there was suspicion for ongoing bleeding, hemodynamic instability, massive transfusion protocol activation, planned return to the operating room, more than 72 h from delivery had elapsed, or if a collection timing was missed within 15 min of the protocol. Consecutive deliveries that met the criteria were enrolled.

Data collection was abstracted by individual chart review. Race was self‐reported at the time of hospital registration. Anemia is defined as hemoglobin less than 10.5 g/dL on admission. Where multiple etiologies were suspected (e.g., laceration, uterine atony, etc.), they were each marked as present. “Starting hemoglobin” is the hemoglobin value on admission for labor, “pre‐transfusion hemoglobin” is the value after delivery stabilization but prior to transfusion. Posttransfusion hemoglobin values were collected at the three study protocol time points (immediate, 4‐h, and 12‐h posttransfusion).

The primary outcome was the average difference in hemoglobin concentration between the immediate posttransfusion CBC and the 12‐h CBC. The 12‐h hemoglobin concentration was selected as the “fully equilibrated” gold standard time point, as it is sufficiently distant from delivery in stable patients to reasonably assume that any necessary equilibration would have been completed by that interval [5, 6, 7, 11]. The secondary outcome was the rise in hemoglobin concentration at each time point per unit of PRBC transfused.

A pre‐study retrospective review of eight patients at our institution (unpublished data) found that the average difference in hemoglobin value between 4‐h and next‐day‐posttransfusion hemoglobin concentration was 0.125 g/dL with a standard deviation (SD) of 0.24 g/dL. Therefore, we determined that if the immediate posttransfusion hemoglobin had more than two SD discrepancies from the remote (12‐h) hemoglobin value then the immediate test would be considered insufficient for use in clinical management. Thus, we designed the study to detect a difference of 0.6 g/dL.

Statistical analysis consisted of a paired t‐test comparing 12‐h posttransfusion hemoglobin values with those obtained immediately and 4‐h posttransfusion. Results were considered significant if the p value was less than 0.05. Power analysis determined that 23 patients would be needed for 80% power to detect a difference of 0.6 g/dL using paired t‐tests with a p value of <0.05. Shapiro–Wilk test was used to verify normal distribution of linear data such as hemoglobin concentration that was reported using mean and SD. Because the analysis was comprised of paired values within the same patient, additional adjustments were not performed. All analyses were performed using R Statistical Software (v4.3; R Core Team 2023). The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for reporting observational studies were followed throughout the manuscript [12].

3. RESULTS

During the study period, there were 2634 deliveries with a total cesarean section rate of 46.5%. Thirty‐eight patients received non‐emergency postpartum PRBC transfusion due to acute blood loss at delivery, and 24 completed the study lab draw protocol. The mean age of participants was 28 years (SD 6.3), and the cohort was racially diverse, with 45.8% identifying as Black, 25.0% as Hispanic, 8.3% as White, and 20.8% as other races. The mean BMI was 33.4 (SD 6.6). Chronic hypertension was present in 8.3% of participants, anemia in 62.5%, diabetes (gestational or pre‐existing) in 8.3%, and preeclampsia with severe features in 8.3%. More than half of the patients (56%) delivered by cesarean (Table 1).

TABLE 1.

Baseline characteristics of study participants.

Characteristic	N (%) Total = 24
Mean age (SD)	28 (6.3)
Race
Black	11 (45.8)
Hispanic	6 (25)
White	2 (8.3)
Other	5 (20.8)
Mean body mass index (SD)	33.4 (6.6)
Chronic hypertension	2 (8.3)
Gestational hypertension	8 (33)
Preeclampsia with severe features	2 (8.3)
Diabetes (GDM or pre‐existing)	2 (8.3)
Anemia	15 (62.5)
Delivered by cesarean	14 (56)
Etiology of postpartum hemorrhage b
Surgical	14 (58.3)
Atony	11 (45.8)
Laceration	2 (8.3)
Abruption	1 (4.2)
Hemoglobin values (g/dL) a
Starting hemoglobin	9.5 (1.0)
Pretransfusion	7.2 (0.8)
Immediate posttransfusion	8.6 (1.1)
4‐h Posttransfusion	8.6 (0.9)
12–24 h Posttransfusion	8.6 (0.9)
Quantitative blood loss (mL)
Mean (SD)	990.3 (411.1)
Range	400–1725
QBL > 1000 mL	11 (45.8)
Packed red blood cells Given
One unit	14 (58.3)
Two units	10 (41.6)

^a Presented as mean (Standard Deviation).

^b Patients can have more than one attributed etiology.

The etiologies of postpartum hemorrhage included surgical causes (58.3%), uterine atony (50%), laceration (8.3%), and abruption (4.2%). The mean quantitative blood loss was 990.3 mL (SD 411.1), with a range from 400 to 1725 mL. A total of 14 (58.3%) patients received one unit of PRBC and 10 (41.6%) patients received two units. The mean starting hemoglobin was 9.5 g/dL (SD 1.0) with a pre‐transfusion value of 7.2 g/dL (SD 0.8). The mean hemoglobin concentration immediately after transfusion was 8.6 g/dL (SD 1.1), which remained stable at both 4‐h (8.6 g/dL, SD 0.9) and 12‐h posttransfusion (8.6 g/dL, SD 0.9; Table 1).

When compared to the 12‐h posttransfusion value, there was no significant difference in mean hemoglobin concentration in g/dL between the immediate (mean difference 0.08 g/dL; 95% confidence interval [CI], −0.44 to 0.26) and 4‐h posttransfusion values (mean difference 0.10 g/dL; 95% CI, −0.41 to 0.21; Figure 1). When participants who received one unit of PRBC were compared to those participants who received two units of PRBC, the difference between immediate, 4‐h, and 12‐h results were also not significantly different (Table 2). The mean rise in hemoglobin in our study for those with one unit transfused was 0.97–1.1 g/dL and for those with two units transfused was 1.89–2.01 g/dL, depending on the time point measured (Table 2).

FIGURE 1.

Box and whisker plot of hemoglobin levels in g/dL at each time point after transfusion for stable symptomatic anemia due to perinatal blood loss postpartum. Immediately refers to laboratory assessment within 15 min of completion of packed red blood cell transfusion. 4‐h and 12‐h refer to proscribed delays for laboratory testing after completion of transfusion. Comparisons were performed using paired t‐tests. There are no significant differences between immediate, 4‐h, and 12‐h hemoglobin levels in stable patients without ongoing bleeding.

TABLE 2.

Posttransfusion hemoglobin analysis.

Starting time point	Comparator time point	Shapiro–Wilk p value	Paired t‐test p value	Mean difference	95% CI Lower	95% CI Upper
All patients (N = 24)
Pre‐transfusion	Immediate	0.457	<0.001	1.39	0.96	1.83
Pre‐transfusion	4‐h	0.810	<0.001	1.38	1.04	1.72
Pre‐transfusion	12‐h	0.327	<0.001	1.48	1.15	1.81
Immediate	12‐h	0.263	0.661	0.08	−0.44	0.26
4‐h	12‐h	0.713	0.543	0.10	−0.41	0.21
Patients transfused one unit (N = 14)
Pre‐transfusion	Immediate	0.968	<0.001	0.97	0.48	1.45
Pre‐transfusion	4‐h	0.947	<0.001	1.02	0.65	1.39
Pre‐transfusion	12‐h	0.957	<0.001	1.11	0.78	1.43
Immediate	12‐h	0.896	0.626	0.13	−0.72	0.45
4‐h	12‐h	0.880	0.731	0.08	−0.61	0.44
Patients transfused two units (N = 9)
Pre‐transfusion	Immediate	0.826	<0.001	1.99	1.26	2.71
Pre‐transfusion	4‐h	0.867	<0.001	1.89	1.34	2.43
Pre‐transfusion	12‐h	0.839	<0.001	2.01	1.46	2.55
Immediate	12‐h	0.939	0.900	0.02	−0.37	0.33
4‐h	12‐h	0.984	0.329	0.12	−0.38	0.14

Note: Difference in measurements of hemoglobin concentration in g/dL at various time points. Paired analysis was performed with paired t‐tests, Shaprio–Wilk test was used to verify normal distribution.

Abbreviation: CI, confidence interval.

4. DISCUSSION

Our study demonstrates that, in stable postpartum patients who have undergone PRBC transfusion for acute blood loss at delivery, hemoglobin concentrations measured immediately after transfusion are not significantly different from those obtained at 4‐ or 12‐h posttransfusion.

Our findings are consistent with other studies of nonpregnant adults that have shown that laboratory assessment as early as 15 min after transfusion is as accurate as waiting for 4–6 h for equilibration [5, 6, 7]. Prior to our study, the applicability of these findings to postpartum patients remained uncertain due to the unique physiological changes and fluid shifts that occur following delivery. Our study addresses this gap and provides evidence that immediate posttransfusion hemoglobin assessment is reliable in the postpartum setting.

Hamm et al. reported the effectiveness of a single‐unit transfusion strategy compared to multiple units in obstetric patients with postpartum hemorrhage [13]. Their primary outcome was avoiding transfusion of the second unit of PRBC. A 4–6 h post‐transfusion lab assessment was used to decide if more PRBC units were needed after a single unit transfusion. In their study, 80% of women following this sequential‐unit protocol avoided a second transfusion without increased morbidity or changes in postpartum outcomes such as breastfeeding, depression, bonding, or fatigue. Our results suggest that the delay between first unit completion and lab testing can be shortened from 4–6 h to less than 15 min in postpartum patients, allowing faster care with this single‐unit transfusion strategy. Our study also supports the findings of a secondary analysis by Rush et al. [9] of the RCT by Hamm et al. [13] that the expected rise of hemoglobin concentration is 1 g/dL per unit of PRBC transfused in postpartum patients [9].

There are several strengths to this study, including its prospective design, rigorous protocol adherence, and comprehensive data collection. The study cohort was diverse with respect to demographic and clinical characteristics, enhancing the generalizability of the findings. Most patients had comorbidities such as anemia and hypertension, and the causes of postpartum hemorrhage reflected those seen in the general obstetric population.

However, certain limitations should be acknowledged. The sample size, while adequately powered for the primary, was modest, and the study was conducted at a single academic center. Additionally, patients with ongoing bleeding or hemodynamic instability were excluded, so the findings may not be generalizable to all postpartum patients requiring transfusion, such as those that require massive transfusion.

5. CONCLUSION

Our findings indicate that hemoglobin equilibration occurs rapidly following PRBC transfusion in stable postpartum patients, and that an arbitrary 4‐h delay after the completion of transfusion before laboratory reassessment is not necessary.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

FUNDING INFORMATION

The authors received no specific funding for this work.

ETHICS STATEMENT

The study was approved by the Institutional Review Board as an exempt quality improvement study and granted a waiver of individual informed consent (#HSC 2024‐2598)

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.