OTTAWA—A hemoglobin cutoff of 70 g/L is generally as safe as a cutoff of 100 g/L in determining which critically ill patients with cardiovascular disease need red blood cell transfusions. That was the conclusion of Canadian researchers who compared the two transfusion strategies in 357 such patients.[1]

Overall mortality and multiple organ dysfunction (MOD) scores were similar with the two approaches, Claudio Martin, MD, one of the researchers, told PULMONARY REVIEWS. He cautioned, however, that there was a trend toward decreased survival in patients with severe ischemic disease, and therefore the higher cutoff may still be appropriate for patients with myocardial infarction, unstable angina, or severe peripheral arterial disease. But for many patients with cardiovascular disease, the lower cutoff can markedly diminish the need for red blood cell transfusions and thereby decrease the risk of transfusion-related complications.

The patients in this study were a subgroup of the Transfusion Requirements in Critical Care (TRICC) trial, a randomized, controlled, multicenter clinical investigation into lower and higher transfusion thresholds in critically ill patients.[2] The subgroup analysis occurred after publication of the overall trial results. It included all TRICC subjects with a cardiac or vascular condition as the primary or secondary diagnosis and all of those with known acute or chronic coronary artery disease. The most common primary cardiac diagnoses were ischemic heart disease, peripheral vascular disease, congestive heart failure, and cardiogenic shock. Thirty-day all-cause mortality was the primary outcome measure.

All participants in the TRICC trial had a hemoglobin concentration of 90 g/L or less within 72 hours of intensive care unit (ICU) admission. They were randomized to one of two transfusion strategies. Those in the restrictive group received a unit of red blood cells when their hemoglobin fell to 70 g/L, and that level was subsequently maintained between 70 and 90 g/L.

Patients in the liberal group were transfused with a unit of red blood cells at a hemoglobin level of 100 g/L, which was then kept between 100 and 120 g/L.

The transfusion protocol was only in effect during the patient’s stay in the ICU. When the patients were discharged to a hospital ward, a copy of the American College of Physicians’ transfusion guidelines was attached to their charts with a request that the physician providing further care follow the guidelines as closely as possible.

Among the 357 patients with cardiovascular disease, 160 were randomized to the restrictive group and 197 to the liberal group. Other than moderate variations in cardiac and anesthetic medication use, the two groups had similar baseline characteristics.

During their ICU stays, the two groups were also similar with regard to variables that could affect oxygen delivery, including vasoactive drug administration, use of pulmonary artery catheterization, and daily fluid balance. In addition, their overall medication use, number of days of dialysis and mechanical ventilation, and average number of surgical procedures were comparable. Daily hemoglobin concentrations averaged 85 g/L in the restrictive group and 103 g/L in the liberal group—a difference that was significant during the 30-day follow-up.

The mean number of red blood cell units given to each patient was much lower in the restrictive group (2.4, vs 5.2 in the liberal group). Conversely, the rate of physician nonadherence to the transfusion protocol—defined as hemoglobin levels outside the prespecified ranges for at least 48 hours—was higher in the liberal group (4.1%, vs 1.4% in the restrictive group).

Thirty-day mortality was 23% in both groups, and secondary end points (60-day, ICU, and hospital mortality) were also about the same. MOD scores were not significantly different between groups during the study period. However, the change in MOD score from baseline was significantly lower in the restrictive group than in the liberal group (0.2 vs 1.3, respectively).

The mean length of ICU stay was two days lower in the restrictive group, but this difference was nonsignificant. Except for acute pulmonary edema, which was less common in the restrictive group, the two groups had similar rates of complications, such as shock, myocardial infarction, unstable angina, and cardiac arrest.

In the subset of 257 patients with ischemic heart disease, there was no significant difference in 30-day, 60-day, or ICU mortality. However, the authors did detect a nonsignificant decline in overall survival among the patients with confirmed ischemic heart disease, severe peripheral vascular disease, or severe comorbid cardiac disease who were in the restrictive group.

“Our results are somewhat soft because of the retrospective nature of the study,” acknowledged Dr. Martin, of the Department of Critical Care Medicine at the University of Western Ontario in Ottawa. He and his colleagues nevertheless believe that restrictive transfusion should be the approach of choice for most critically ill patients with cardiovascular disease. This strategy is preferable not only because it was safe but also because it decreased the average number of red blood cell units transfused by 53%.

Their results should ease fears about anemia-related increases in mortality and organ damage at the lower threshold.

“[T]olerating anemia to a hemoglobin of [70 g/L] and minimizing exposure to red blood cell transfusion are clearly appropriate among … most critically ill patients with cardiovascular disease,” concurred Stephen D. Surgenor, MD, Marcus J. Hampers, MD, and Howard L. Corwin, MD, specialists in critical care and anesthesiology at Dartmouth-Hitchcock Medical Center in Lebanon, New Hampshire, in an accompanying editorial.[3] However, they agreed with Dr. Martin and his colleagues that the 100-g/L cutoff may be more appropriate for patients with severe ischemic disease, at least until further research is done.

—Timothy Begany

References
1. Hébert PC, Yetisir E, Martin C, et al. Is a low transfusion threshold safe in critically ill patients with cardiovascular diseases? Crit Care Med. 2001;29:227-234.

2. Hébert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. N Engl J Med. 1999;340:409-417.

3. Surgenor SD, Hampers MJ, Corwin HL. Is blood transfusion good for the heart? [review]. Crit Care Med. 2001;29:442-444.

Return to table of contents