Clinical Validation of Risk Stratification Criteria for Peripartum Hemorrhage
Abstract & Commentary
By John C. Hobbins, MD
Professor, Department of Obstetrics and GynecologyUniversity of Colorado School of Medicine, Aurora
Dr. Hobbins reports no financial relationships relevant to this field of study.
Synopsis:A study from Pittsburgh testing the efficacy of the California Maternal Quality Care Collaborative method for "at risky" patients for possible blood transfusion validated this commonly used protocol and demonstrated that adding more high-risk categories did not add to its efficiency.
Source: Dilla AJ, et al. Clinical validation of risk stratification criteria for peripartum hemorrhage. Obstet Gynecol 2013;122:120-126.
Three years ago, the california maternal quality Care Collaborative drafted guidelines on how to identify and prepare for patients who are at varying risk for developing peripartum hemorrhage.1 Since then, these guidelines have been adopted by many hospitals around the country. Magee Women's Hospital in Pittsburgh is one of them, and recently authors from this institution published an instructive paper that tested the real-life efficacy of putting these guidelines into practice.2 Also, the authors addressed the concept of adding other high-risk criteria to the existing California guidelines in an effort to capture even more patients who might benefit from expanded preparation.
During 2011, 10,134 women delivered at this hospital. Data were available to assign every patient to one of the California risk categories — low, medium, high — according to the guidelines in Table 1. In addition, the authors added another nine categories to the list. One addition was a model for abnormal placentation (accreta and percreta). Their simple endpoint was all peripartum hemorrhage — defined as a significant hemorrhage requiring at least one unit of red blood cells (RBCs).
There were 7060 patients in the low-risk category, 2462 in the medium-risk category, and 412 in the high-risk category. The rate of hemorrhage requiring RBCs was 1.4% (139 patients) in the entire group, with a rate of hemorrhage of 0.8%, 2%, and 7.3% in the low-, medium-, and high-risk groups, respectively. Of the 139 patients needing blood, 22% came from a high-risk group. When the authors reclassified the groups by adding their additional nine high-risk criteria, the new modified high-risk group included 85% of those needing blood, but the chance of this happening in the expanded category was only 2.6%.
The authors, while being candid about the lack of benefit from their modifications to the existing guidelines, were able to put the California screening and preparation strategies into proper perspective. For example, while typing and screening patients requires about 45 minutes in most hospitals, cross-matching only takes an additional 5 minutes in those hospitals using an electronic matching system now commonly employed in larger institutions. An additional 30-40 minutes is needed in hospitals where this is not available.
We always worry about being unprepared for an unexpected hemorrhage, and this is why the authors tested the concept of adding categories to up-risk more patients into a "more likely" group. A very aggressive strategy to capture more patients needing transfusion would be to type and cross-match everyone in the medium-risk group for two units of RBCs (instead of simply typing and screening them as advocated now), but it would mean that this ritual would have to be performed in 263 patients for every patient who would actually need blood (19 of 5005 patients). The authors postulated that if a protocol for typing and cross-matching is to be effective, more than 5% of the high-risk group would need to be transfused, and the only group in the study falling into that category was the high-risk group (7.3%). This should validate the California protocol. Also, as the authors point out, just expanding the California protocol’s high-risk category to include any of the authors’ five best predictors of hemorrhage would add 10 times more patients to the high-risk group, and, even then, only 2.6% in this bloated group would need a transfusion. Last, the "adding more" concept needs to be put into perspective when one realizes that it only takes about 5 minutes to cross-match the blood in most hospitals and that uncrossed type O blood is usually immediately available with little risk of hemolysis in those clinical situations where waiting is not an option.
The obvious conclusion that can be drawn from this study is that the California protocol, drafted from expert opinion alone, has stood up to scrutiny in a single institution investigation involving large numbers of patients.
- Bingham D, et al. The California Maternal Quality Care Collaborative (CMQCC). CMQCC obstetric hemorrhage hospital level implementation guide. Palto Alto, CA: California Maternal Quality Care Collaborative; 2010.
- Dilla AJ, et al. Clinical validation of risk stratification criteria for peripartum hemorrhage. Obstet Gynecol 2013;122:120-126.