Transfusion Improves Outcome in Anemic Elderly Patients with Acute MI
Abstract & Commentary
Synopsis: For elderly patients with acute myocardial infarction who are anemic, transfusion appears to improve short-term mortality.
Source: Wu W-C, et al. Blood transfusion in elderly patients with acute myocardial infarction. N Engl J Med. 2001;345(17):1230-1236.
The cooperative cardiovascular project was a national program of the former Health Care Financing Administration (now called the Centers for Medicare and Medicaid Services) that gathered data on 234,769 Medicare beneficiaries hospitalized with acute myocardial infarction (MI) in 1994 and 1995. Patients included in the cohort all had a primary discharge diagnosis of acute myocardial infarction. In the present study, Wu and associates examined this large cohort for any relationship between admission hematocrit and short-term mortality, and also for any effect of blood transfusion during hospitalization.
Wu et al excluded patients younger than 65, those whose diagnosis of acute MI could not be confirmed, those readmitted for another MI during the study period, and patients transferred into or out of the study hospitals during the index admission. Patients with no recorded admission hematocrit values, or values above the normal range, were excluded from analysis, as were those known to have metastatic cancer or other terminal illness, those who underwent cardiac surgery, those with gastrointestinal bleeding during the hospitalization, and patients meeting several other small exclusion criteria. The resulting cohort consisted of 78,974, or 34% of the original group.
A total of 34,275 of the study patients (43%) had hematocrit values lower than 39% on admission, and 3324 (4.2%) had values below 30%. Lower admission hematocrit values were associated with more frequent in-hospital events such as shock, heart failure, and death, and also with an increased length of hospital stay (see Table).
A total of 3680 patients (4.7%) received red cell transfusions during the hospitalization. The association between transfusion and outcomes varied according to the admission hematocrit value. Transfusion appeared to improve survival for patients who were definitely anemic: adjusted odds ratios for 30-day mortality among patients who received transfusions were 0.22 (95% confidence interval, 0.11-0.45) for admission hematocrits of 5 to 24%; 0.48 (95% CI, 0.34-0.69) for admission hematocrits of 24-27%; 0.60 (95% CI, 0.47-0.76) for values of 27-30%; and 0.69 (95% CI, 0.53-0.89) for admission hematocrits of 30-33%. However, transfusion was associated with an increased 30-day mortality for those with admission hematocrits above 36% (adjusted odds ratio, 1.13 to 1.46).
| Table | |||||||
| Outcomes of Acute MI in Elderly Patients as a Function of Admission Hematocrit* | |||||||
| Hematocrit (%) | 5-24 | 24-27 | 27-30 | 30-33 | 33-36 | 36-39 | 39-48 |
| Number of patients | 380 | 838 | 2106 | 4848 | 9885 | 16,218 | 44,699 |
| Shock (%) | 15.0 | 10.3 | 10.9 | 9.5 | 8.6 | 7.5 | 6.2 |
| Heart failure (%) | 61.1 | 63.0 | 63.0 | 58.3 | 51.6 | 45.1 | 40.5 |
| 30-day mortality (%) | 38.7 | 35.2 | 35.9 | 30.0 | 25.6 | 20.9 | 17.2 |
| Length of stay (days) | 9.9 | 10.6 | 10.0 | 9.6 | 9.1 | 8.6 | 8.3 |
|
* all trends statistically significant; P < 0.001 |
|||||||
Comment by David J. Pierson, MD
This study found that there was a high prevalence of anemia among elderly patients admitted with acute MI, and that anemic patients had higher 30-day mortality rates than nonanemic patients. Importantly, it also found that blood transfusion reduced this short-term mortality rate, in proportion to the degree of initial anemia. That is, in the large cohort used as a database, transfusion improved survival among patients with admission hematocrit values of 30% or lower, and may have been effective in lowering mortality in patients with hematocrits as high as 33%.
In order to really prove a cause-and-effect relationship here, a randomized trial would have to be done in which half of all anemic elderly MI patients received blood transfusions and the other half did not, using strict criteria and making all other aspects of management the same. It is unlikely that such a controlled trial will be done. However, because of the size of the study population and the robust methods used, the present study’s findings should probably be taken seriously.
As pointed out in the editorial accompanying this study, the mortality rate among patients with admission hematocrit values of 27% or lower who did not receive transfusions approached 50%, and was nearly 3 times the mortality rate among patients with admission hematocrits over 39%. Transfusion was infrequently given to the patients in this study, even when they had substantial anemia. Only 24% of the patients whose admission hematocrits were 33% or lower received transfusions. If there really is a cause-and-effect relationship here, administering transfusions to all the anemic patients in this study would have saved a substantial number of lives.
A lot of attention has been focused on transfusion practices in the ICU, and most of it has come with the message that we give too many transfusions, not too few. A multicenter, prospective, randomized controlled trial of transfusion thresholds in critically ill patients recently showed that the use of a lower cut-off (a hemoglobin concentration of 7.0 rather than 10.0 g/dL) was clinically safe.1 An exception appears to be acute MI. Follow-up analysis of the data from the same study showed that there was a trend toward a higher mortality among 257 patients in a subgroup with ischemic heart disease.2 Thus, patients with acute MI seem to be an important exception to the new transfusion rules being implemented in ICUs everywhere. If a patient admitted with an acute MI has a hematocrit level of 33% or less on admission, the risks and costs of transfusion would seem to be outweighed by the potential gain in reduced complications, length of stay, and short-term mortality.
References
1. Hebert PC, et al. N Engl J Med. 1999;340:409-417.
2. Hebert PC, et al. Crit Care Med. 2001;29:227-234.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content