A comprehensive blood management program for orthopedic surgery patients is not only effective in reducing overall blood utilization but is also associated with reduced length of stay (LOS) and readmissions, with no increase in adverse outcomes, a study has concluded.
“One year ago, the AABB [American Association of Blood Banks] released its clinical practice guidelines [JAMA 2016;316:2025-2035], suggesting that patients be transfused at a hemoglobin trigger of 7 g/dL unless they were cardiac surgery patients, orthopedic surgery patients, or had existing cardiovascular disease, in which case 8 g/dL was recommended,” said Pranjal B. Gupta, BE, a second-year medical student at the Johns Hopkins School of Medicine, in Baltimore. “The orthopedic surgery guidelines come from a few studies, primarily focused on elderly, anemic patients and hip fracture patients, which doesn’t necessarily represent all orthopedic patients.
“With that in mind, we believe it remains unclear whether all orthopedic surgery patients need a more liberal transfusion trigger of 8 g/dL, since lower hemoglobin triggers were not even tested.”
Indeed, recent advances in surgical techniques and antifibrinolytic usage seem to have reduced transfusion requirements for routine orthopedic surgery procedures. As a result, the institution implemented a systemwide patient blood management program aimed at reducing overall blood utilization in January 2015. The program comprised a number of initiatives, including provider education, tranexamic acid, a new surgical blood order schedule, electronic best practice advisories, a campaign advocating for single-unit red blood cell transfusions, and audits with provider feedback for guideline compliance.
“This study is a database analysis two years before and two years after that program, to analyze blood utilization and outcomes in orthopedic patients,” Mr. Gupta said.
A total of 2,951 orthopedic surgery patients were treated at the institution between 2013 and 2017. Patients were divided into pre– or post–blood management cohorts based on surgical date. The two cohorts were compared using in-hospital databases to assess Case Mix Index, nadir (trigger) and discharge (target) hemoglobin in transfused patients, transfusion rates and postoperative outcomes. Adverse outcomes included hospital LOS, 30-day readmission rates, a composite morbidity/mortality score (including infectious, ischemic, renal, respiratory and thrombotic complications), and individual morbidity and mortality rates.
As Mr. Gupta reported at ANESTHESIOLOGY 2017 (abstract A3101), patient characteristics were similar between the two cohorts with the exception of aggregate patient complexity and age, both of which were slightly—but significantly—higher in the post–blood management group.
Not surprisingly, blood utilization decreased with the implementation of the program, as trigger hemoglobin dropped from 7.8±1.0 to 6.9±1.1 g/dL (P<0.0001) and target hemoglobin dropped from 9.0±1.1 to 8.4±1.1 g/dL (P<0.0001). The study also revealed a 38% decrease in the percentage of patients transfused (242/1,513 vs. 144/1,438; P<0.0001) and a 25% decrease in mean red blood cell units per patient (0.34±0.98 vs. 0.26±1.10; P=0.036).
“With respect to clinical outcomes, we found that four of the five outcomes we measured improved after patient blood management,” Mr. Gupta continued. Median LOS fell from three days (range, 1-4) to two days (range, 1-3; P<0.0001), while 30-day readmission decreased from 8.8% (133/1,513) to 6.0% (86/1,438; P=0.007).
“Our composite variable of morbidity and mortality also improved,” Mr. Gupta said, “as did individual morbidity. And mortality was unchanged. So basically, giving less blood improved clinical outcomes, with no change in mortality.”
To test the strength of the findings, the researchers also performed a risk adjustment for potential confounders, such as age, Case Mix Index and sex. “Even after adjusting for some of these variables, patient blood management was still significantly associated with a lower odds ratio of our adverse composite outcome of grouped morbidity and mortality rate,” he revealed. A subsequent subgroup analysis of patients over 65 years of age demonstrated that these individuals fared as well or better under a restrictive transfusion regimen.
“In conclusion, we report reduced blood utilization and improved outcomes after implementing patient blood management for orthopedic surgery patients, even after risk adjustment of confounding variables,” Mr. Gupta said. “So the AABB guidelines recommending a hemoglobin trigger of 8.0 g/dL for all orthopedic patients may be inappropriate, given our favorable findings with a trigger under 7 g/dL.”
“A lot of things changed clinically,” added co-investigator Steven M. Frank, MD, a professor of anesthesiology and critical care medicine at the institution. “So all we can really say is that we’re giving less blood in the context of the whole clinical scenario. And we can give them less blood, and they do as well or better.”