ABSTRACT: Postpartum hemorrhage causes approximately 11% of maternal deaths in the United States and is the leading cause of death that occurs on the day of birth. Importantly, 54–93% of maternal deaths due to obstetric hemorrhage may be preventable. Studies that have evaluated factors associated with identification and treatment of postpartum hemorrhage have found that imprecise health care provider estimation of actual blood loss during birth and the immediate postpartum period is a leading cause of delayed response to hemorrhage. Although current data do not support any one method of quantifying blood loss as superior to another, quantification of blood loss, such as using graduated drapes or weighing, provides a more accurate assessment of actual blood loss than visual estimation; however, the effectiveness of quantitative blood loss measurement on clinical outcomes has not been demonstrated. Successful obstetric hemorrhage bundle implementation is associated with improved outcome measures related to obstetric hemorrhage. However, further research is necessary to better evaluate the particular effect of quantitative blood loss measurement in reducing maternal hemorrhage- associated morbidity in the United States.

Recommendations and Conclusions

The American College of Obstetricians and Gynecologists makes the following recommendations and conclusions:

• Quantitative methods of measuring obstetric blood loss have been shown to be more accurate than visual estimation in determining obstetric blood loss.

• Studies that have compared visual estimation to quantitative measurement have found that visual estimation is more likely to underestimate the actual blood loss when volumes are high and overestimate when volumes are low.

• Although quantitative measurement is more accurate than visual estimation for identifying obstetric blood loss, the effectiveness of quantitative blood loss measurement on clinical outcomes has not been demonstrated.

• Implementation of quantitative assessment of blood loss includes the following two items: 1) use of direct measurement of obstetric blood loss (quantitative blood loss) and 2) protocols for collecting and reporting a cumulative record of blood loss postdelivery.

• Interprofessional protocols for the assessment of blood loss, including quantitative assessment, for both vaginal and cesarean births are best developed by a multidisciplinary team.

• Successful obstetric hemorrhage bundle implementation is associated with improved outcome measures related to obstetric hemorrhage. However, further research is necessary to better evaluate the particular effect of quantitative blood loss measurement in reducing maternal hemorrhage-associated morbidity in the United States.

Purpose

The purpose of this Committee Opinion is to review and clarify the current evidence regarding the accuracy of methods available for determining obstetric blood loss, including quantitative and visual estimated blood loss methods, and to identify research gaps. The American College of Obstetricians and Gynecologists’ support for the Alliance for Innovation on Maternal Health’s Obstetric Hemorrhage Patient Safety Bundle is well established, and this document is not intended to describe how to implement the bundle. Rather, it is intended to help facilities understand the evidence that supports different approaches to measuring obstetric blood loss. Although this document offers general guidance, technicalities on how to implement blood loss measurement can be found elsewhere and is, therefore, not included here (see the For More Information section).

Introduction

Obstetric hemorrhage is a major cause of maternal morbidity 1. Postpartum hemorrhage causes approximately 11% of maternal deaths in the United States and is the leading cause of death that occurs on the day of birth 2 3 4 5. Hemorrhage that requires a blood transfusion is also the leading cause of significant maternal morbidity 4 5 6 7. Importantly, 54–93% of maternal deaths due to obstetric hemorrhage may be preventable 3 8 9 10. Studies that have evaluated factors associated with identification and treatment of postpartum hemorrhage have found that imprecise health care provider estimation of actual blood loss during birth and the immediate postpartum period is a leading cause of delayed response to hemorrhage 10 11 12 13. Quantitative methods of measuring obstetric blood loss have been shown to be more accurate than visual estimation in determining obstetric blood loss 14 15 16 17 18 19. Some studies found that the use of quantitative methods resulted in a higher likelihood that women who experienced a postpartum hemorrhage were identified 15 17 20. However, other studies have not found that quantitative blood loss better predicts postpartum hemoglobin values 21 or changes the incidence of postpartum blood transfusion 22 23, and thus, the effect on clinical outcomes is less clear 20.

Prevention of Postpartum Hemorrhage

Recent efforts to decrease the incidence of maternal mortality and morbidity secondary to obstetric hemorrhage have focused on development of interdisciplinary team-based protocols that facilitate early diagnosis and treatment 24 25. The Alliance for Innovation on Maternal Health has developed an Obstetric Hemorrhage Patient Safety Bundle that is being increasingly adopted in hospitals within the United States 24 25 26 27 28. Quantitative and cumulative assessment of blood loss is one of several components of the Obstetric Hemorrhage Patient Safety Bundle. The California Maternal Quality Care Collaborative, regional hospital systems, and individual hospitals have reported reductions in severe maternal morbidity among patients who experience obstetric hemorrhage after implementation of this bundle, although it remains unclear whether these improvements are due to specific practices within the bundle or implementation of the bundle in total 24 28 29 30.

Visual Estimation of Obstetric Blood Loss

Historically, visual estimation of blood loss during and after childbirth has been the primary method of determining obstetric blood loss. Visual estimation is subjective and imprecise 31 32 33 34. Studies that have compared visual estimation to quantitative measurement have found that visual estimation is more likely to underestimate the actual blood loss when volumes are high and overestimate when volumes are low 19 32 33 34 35 36. Attempts to improve visual estimation of blood loss using visual tools for volume comparisons have been studied 32 34. These tools have not been found to consistently improve the accuracy of visual estimation 34 37. Although one study demonstrated improved accuracy with visual estimation of blood loss through a training program 37, a more recent study demonstrated skill decay within 9 months of training completion 34. Furthermore, visual estimation of blood loss does not appear to improve with health care provider specialty, age, or clinical experience 14 18 33.

Quantitative Measurement of Obstetric Blood Loss

Visual estimation has been compared to quantitative methods in both clinical and simulated scenarios 14 17 18 32 35 36. The accuracy of blood loss assessment is improved with quantitative measurement techniques 14 15 16 18 19 32 36. For example, one study compared visual estimation to a gravimetric measurement in a prospective cohort study that included 150 women. In this study, blood-soaked items were weighed, and the dry weight of the items was subtracted to obtain blood loss volume. Visual estimation of blood loss compared with the gravimetric technique was associated with an error of approximately 30% (gravimetric mean blood loss was 304.1 mL versus nurse- and physician-estimated mean blood loss was 213 mL and 214.3 mL, respectively) 14.

Studies that have compared visual estimation versus quantitative methods in clinical settings have also found that quantitative methods are more likely to accurately detect postpartum hemorrhage 15 16 17 38. An evaluation of low risk women after vaginal birth (n=286) conducted in Singapore found that mean estimated blood loss was 31% less accurate compared with mean measured blood loss 15. In this study the incidence of blood loss greater than 500 mL was 3.5% in the visual estimation group versus 9.1% in the direct measurement group. Only 34.6% of women with a blood loss greater than 500 mL were accurately diagnosed with visual estimation 15. Another, small study had similar findings among a cohort of low risk women after vaginal birth wherein only one woman out of eight who had a measured blood loss of greater than 500 mL was accurately identified by visual estimation 17.

Studies that used simulated blood have had similar results. One study conducted a randomized trial of simulated vaginal delivery and compared obstetric care providers’ estimation of blood loss using calibrated versus noncalibrated vaginal delivery drapes followed by crossover 18. Visual blood loss estimation with noncalibrated drapes underestimated blood loss, with worsening accuracy at larger volumes (16% error at 300 mL and 41% at 2,000 mL). The error was less than 15% at all volumes when the calibrated drape was used.

Recent developments with the use of artificial intelligence-enabled technology platforms appear promising for quantifying blood loss. These artificial intelligence platforms use mobile technology and image recognition algorithms. The tablet camera is used to take a picture of surgical sponges and canisters. The mobile app performs a colorimetric analysis, and the images are uploaded to a cloud-based machine learning program that uses algorithms to quantify hemoglobin and blood loss in real-time. One retrospective cohort study of 2,781 women demonstrated differences in estimated blood loss with an artificial intelligence-enabled platform for real time monitoring of blood loss versus traditional visual estimation for women having a cesarean birth 16. The study found that blood loss greater than 1,000 mL was more frequently detected using the artificial intelligence technology (14.1% vs 3.5% respectively; P<.0001), but transfusion rates were similar between the groups 16. Validation of these findings with additional research is needed.

Quantitative Assessment of Obstetric Blood Loss in Obstetric Hemorrhage Bundles

Analysis of root causes in maternal mortality reviews have consistently found missed or delayed diagnosis and delay in initiating treatment are recurrent problems in care of women with excessive obstetric blood loss. Thus, addressing more accurate and timely diagnosis and treatment of postpartum hemorrhage represents an important quality improvement opportunity for prevention 10 13 39. Obstetric hemorrhage bundles include “measure of cumulative blood loss (formal, as quantitative as possible)” as a component. Implementation of these bundles in U.S.-based birth settings has been found to significantly reduce maternal morbidity in participating hospitals 24 29 30. One study reported data from the California Maternal Quality Care Collaborative state-wide hemorrhage quality improvement initiative that involved collaborative learning with hospital mentorship, rapid response data, and quality improvement support. The study used before-and-after methodology to compare outcomes from women who had an obstetric hemorrhage in hospitals that implemented an obstetric hemorrhage bundle (N=99 hospitals) versus the outcomes of women in comparison hospitals (n=48 hospitals) 24. Women who experienced an obstetric hemorrhage in the collaborative hospitals had a 20.8% reduction in severe maternal morbidity while women in comparison hospitals had a 1.2% reduction ( P<.0001) when maternal outcomes from before the project was implemented (January 2011 to December 2014) were compared with outcomes during the last 6 months of the 18-month project (October 2015 to March 2016). In addition to state-level quality improvement initiatives, two studies from single institutions found successful hemorrhage bundle implementation was associated with a significant reduction in adverse hemorrhage-related outcomes 29 30.

Success of quality improvement initiatives is dependent upon many factors including efficacy of the intervention, duration of the project, and extent of adoption. Reports from multihospital collaboratives have had conflicting results. Early outcome data from a study of regional hospitals working on nursing team’s implementation of bundle elements did not show a reduction in hemorrhage-related severe maternal morbidity but reported that additional time was needed for implementation given no participating hospitals had yet completed implementation of the strategies 27. Additionally, a recent analysis from New York’s Safe Motherhood Initiative did not show a difference in hemorrhage-related morbidity 1 year after initiation of a hemorrhage initiative 40. Conversely, a mandated implementation of an eight-component hemorrhage protocol based on the hemorrhage safety bundle that included quantitative blood loss measurement in 29 hospitals within a multistate regional health system found a significant reduction in use of blood products (−25.9% per 1,000 births, P<.01) when assessed 10 months after implementation of the protocol 28.

Overall, implementation of the California Maternal Quality Care Collaborative obstetric hemorrhage bundle or similar obstetric quality improvement bundle as a state-wide initiative as well as in some individual hospitals and health systems has shown successful obstetric hemorrhage bundle implementation is associated with improved outcome measures related to obstetric hemorrhage. These outcomes may provide evidence of the effectiveness of quantitative blood loss measurement when it is included as a component of an obstetric hemorrhage bundle. However, further research is necessary to better evaluate the particular effect of quantitative blood loss measurement in reducing maternal hemorrhage-associated morbidity in the United States, as well as resources and cost-effectiveness across diverse hospital settings.

Processes for Quantification of Blood Loss

Quantification of maternal blood loss requires a team effort and can represent a cultural shift from health care provider visual estimation of blood loss to a process that involves all clinical team members at delivery, including obstetric care providers and nursing staff. Interprofessional protocols for the assessment of blood loss, including quantitative assessment, for both vaginal and cesarean births are best developed by a multidisciplinary team. Box 1 and Box 2 present example process maps for quantification of blood loss during vaginal and cesarean delivery, respectively.

Implementation of quantitative assessment of blood loss includes the following two items: 1) use of direct measurement of blood loss (quantitative blood loss) and 2) protocols for collecting and reporting a cumulative record of blood loss postdelivery 25. The process for quantification of blood loss at the time of vaginal birth is slightly different than for cesarean birth. To collect all fluids lost during a vaginal birth, a calibrated under-buttocks drape is used, whereas a calibrated suction canister is used during a cesarean birth. In both instances, the volume of fluid collected before delivery of the placenta is largely composed of amniotic fluid and urine (in the case of a vaginal delivery only) and is subtracted from the total volume of fluid collected after completion of the birth to determine the volume of blood lost during birth. Additionally, the amount of any fluid used for irrigation during either type of birth is subtracted from this volume. Finally, total cumulative blood loss in milliliters is determined by adding the weight in grams of blood-soaked materials (eg, laparotomy sponges, 4 × 4 sponges, bedsheets, disposable underpads) minus the dry weight of those materials.

There is insufficient evidence to recommend a specific timeframe to continue blood loss assessment postpartum. However, it is suggested that ongoing blood loss assessment should continue as long as active bleeding is present, or as long as the patient is unstable after a blood loss of more than 1,000 mL, including the postpartum care setting 41 42.

The equipment needed for quantification of blood loss is easily available and includes the following items: calibrated under-buttocks drapes, laminated cards that denote dry weights for delivery items, and a scale to weigh delivery items that become blood soaked. The entire delivery care team participates in implementation of these strategies and is empowered to identify additional resources as needed for individual sites. Obstetric nurses play a critical role in tracking quantitative and cumulative blood loss. The Association of Women's Health, Obstetric and Neonatal Nurses has developed a Postpartum Hemorrhage Project toolkit with support material that includes a video on implementation of quantitative blood loss assessment for obstetric clinical care teams. High-fidelity options, such as the integration of data into the electronic medical record, are available but not necessary to accomplish quantitative and cumulative assessment of blood loss. State collaborative organizations, such as the California Maternal Quality Care Collaborative, the Florida Perinatal Quality Collaborative, and the Oklahoma Perinatal Quality Improvement Collaborative, provide free resources that can assist in the development of facility-specific protocols and policies. Please see the For More Information section for additional resources on quantitative blood loss implementation and processes.

Conclusion

Given approximately 40% of postpartum hemorrhage occurs in low-risk women, every woman giving birth is at risk for obstetric hemorrhage 25. Hemorrhage is a major contributing factor to maternal morbidity and mortality. Although current data do not support any one method of quantifying blood loss as superior to another, quantification of blood loss, such as using graduated drapes or weighing, provides a more accurate assessment of actual blood loss than visual estimation. When quantitative blood loss is included as a component of a bundle of practices that focus on prevention and early diagnosis of excessive blood loss, it may improve situational awareness and thereby improve hemorrhage diagnosis and response time.

Hospitals that participate in quality improvement activities to improve hemorrhage outcomes should monitor compliance and effectiveness of these strategies. Additional research is needed to demonstrate the effect of quantitative assessment of blood loss on clinical outcomes and whether widespread implementation of quantitative blood loss measurement strategies, either as an independent strategy or alongside other hemorrhage bundle components, will decrease maternal severe morbidity and mortality in cases of obstetric hemorrhage.

For More Information

The American College of Obstetricians and Gynecologists has identified additional resources on topics related to this document that may be helpful for ob-gyns, other health care providers, and patients. You may view these resources at www.acog.org/More-Info/QuantitativeBloodLoss.

These resources are for information only and are not meant to be comprehensive. Referral to these resources does not imply the American College of Obstetricians and Gynecologists’ endorsement of the organization, the organization’s website, or the content of the resource. The resources may change without notice.