Blood transfusion for acute decompensated heart failure—friend or foe?

Background

In acute coronary syndromes (ACSs), blood transfusion (BT) has been associated with worse outcomes. The impact of BT among patients with acute decompensated heart failure (ADHF) remains unknown.

Methods

Propensity score analysis of patients with ADHF with and without BT in a national heart failure (HF) survey was used in this study.

Results

Of the 4,102 enrolled patients, 2,335 had ADHF, of whom 166 (7.1%) received BT. These patients were older (75.6% vs 73.6%, P = .04), more likely to be females (54.8% vs 43.9%, P = .007), more likely to have diabetes (59.0% vs 51.1%, P = .04) and renal dysfunction (59.0% vs 40.2%, P < .001), and more likely to receive inotropes (16.9% vs 8.0%, P < .001), but they had similar rates of ACS (41.0% vs 39.4%, P = .69) and prior HF (64.5% vs 70.0%, P = .23). Nadir hemoglobin levels were commonly <10 g/dL in BT patients (92.7% vs 8.0%); 15 BT patients had bleeding complications, of which 10 are major bleeding. Major predictors for BT were ACS (OR 1.85, 95% CI 1.15-2.96), inotropes use (OR 2.36, 95% CI 1.22-4.55), and nadir hemoglobin (OR 0.18 per 1 g/dL increase, 95% CI 0.14-0.22). In-hospital, 30-day, 1-year, and 4-year unadjusted mortality rates were higher for BT patients (10.8% vs 5.2%, P = .02; 11.0% vs 8.5%, P = .27; 39.6% vs 28.5%, P = .03; 69.5% vs 59.5%, P = .01, respectively). However, in 103 propensity-matched pairs (c-statistic 0.97), short-term mortality tended to be lower with BT (8.7% vs 14.6%, P = .20; 9.7% vs 18.4%, P = .08; 38.8% vs 42.7%, P = .59; and 72.8% vs 76.7%, P = .52, respectively).

Conclusions

Acute decompensated HF patients receiving BT had worse clinical features and unadjusted outcomes, but BT per se seemed to be safe and perhaps even beneficial.