Wu and associates did a retrospective cohort study using the Veterans Administration National Surgical Quality Improvement Program database (NSQIP) to examine the relationship between pre operative hemoglobin and 30 day post operative death or cardiac events after non cardiac surgery in patients older than 65. A total of 310,311 patients operated upon between 1997 and 2004 were studied. In this predominately male population, normal hemoglobin was defined as a hematocrit between 39 – 53.9%, anemia was defined as a hematocrit less than 39%, and polycythemia was defined as a hemoglobin greater than 54%. Examples of operations included were inguinal hernioraphy, prostatectomy, angioplasty, total knee replacement, and partial colectomy. Additional procedures within 30 days of the first were excluded. Some very low risk procedures such as eye or nasal surgery are not part of the NSQIP data base. The overall 30 day mortality was 3.89%. Thirty day mortality and cardiac event rates increased with either positive or negative deviations from normal hematocrit levels. There was a 1.6% increase in 30 day postoperative mortality with every percentage point increase or decrease from the normal range. In a separate study they noted similar results for cardiac surgery. The authors concluded that in this population of patients greater than 65 years old, even mild degrees of pre operative anemia or polycythemia were associated with increased risk of mortality or a cardiac event. What this paper does not help with is what to do with these findings. As the authors state, “if anemia is a modifiable risk factor and not simply a marker of other conditions that confer increased risk, then preoperative transfusion might be considered.” However neither this nor other similar papers provide that guidance. Ref: Preoperative hematocrit levels and postoperative outcomes in older patients undergoing noncardiac surgery. JAMA 2007;297:2481 – 2488 (attached).
Karkouti et al examined the effects of acute anemia on adverse outcomes after cardiac surgery in 10,179 patients and found that the incidence of post operative complications as indicated by a composite outcome variable of in hospital death, stroke or kidney failure was related to the percent change in hemoglobin concentration from baseline and not the absolute value of that variable. Regardless of starting hemoglobin or lowest recorded intraoperative hemoglobin a 50% decrease in hemoglobin was associated with markedly increased risk of death, stroke or kidney failure. Dr. Augoustides opines that “this study is consistent with my clinical experience that the safely tolerated degree of acute anemia is a function of baseline hemoglobin concentration. The findings suggest a new rationale to assist in perioperative red blood cell transfusion decision making in cardiac surgery.” Ref: The influence of baseline hemoglobin concentration on tolerance of anemia in cardiac surgery. Transfusion. 2008;48:666-672 (attached)
DSS note -- Both of these papers challenge the current dogma concerning transfusion thresholds. Though the accepted transfusion threshold of 6 - 7 g/dl may be acceptable for healthy patients, there are now two subgroups, the elderly and patients undergoing cardiac surgery for whom lower hemoglobin concentrations are associated with poorer outcomes.
Feedback -- Dr. Sarani sent the following comment “In regards to the beneficial effects of transfusion in elderly or cardiac pts, I respectfully disagree. Study after study has shown that transfusion is associated with worse outcomes in patients with asymptomatic anemia. Please see a recently published article which shows that although anemia is harmful in traumatic brain injury patients, transfusing anemic pts is even more harmful! I think that anemia is an index of illness, not the cause of illness in non-hemorrhaging patients and fixing it does little overall, unless there is evidence of end-organ ischemia. This is akin to giving albumin to hypoalbuminemic patients.” Doctor Sarani provided two papers which are attached: Rao SV et al: Relationship of blood transfusion and clinical outcomes in patients with acute coronary syndromes. JAMA 2004;292:1555-1562. Salim A et al: Role of anemia in traumatic brain injury. J Am Coll Surg 2008;207:398-406.
DSS note -- The first two papers abstracted examined the relationship between anemia and outcome in very different patient populations than those in the papers provided by Dr. Sarani. Also the initial two papers did not examine the effect of transfusion. Thus it is possible that anemia is a marker but transfusion will not improve outcome.
Joni Augoustides M.D., FASE is Associate Professor of Anesthesiology and Critical Care at the Hospital of the University of Pennsylvania
Barbak Sarani M.D. is Assistant Professor of Surgery, University of Pennsylvania School of Medicine
David S. Smith, M.D., Ph.D.