Intensive Care Unit and Sepsis

The Transfusion Requirement in Critical Care trial (TRICC) was the first to challenge the widely held view that a threshold of 10 g/dL was required to recover from acute life-threatening illness. A total of 838 euvolemic intensive unit patients with a Hgb of less than 9 g/dL were randomly allocated to a 10 g/dL RBC transfusion threshold (liberal) group or a 7 g/dL RBC transfusion threshold (restrictive) group. The 30-day mortality was lower in patients in restrictive group (18.7%) than the 10 g/dL group (23.3%; P = .1) as were cardiovascular and pulmonary complications. The TRICC trial was the first to suggest that a 7 g/dL threshold was as safe and perhaps safer than a 10 g/dL threshold.

More recently, the Transfusion Requirements in Septic Shock (TRISS) trial, a randomized control trial in 1005 patients with septic shock in the ICU, found no difference in mortality or morbidity in individuals allocated to liberal (<9 g/dL) versus restrictive (<7 g/dL) transfusion strategies. Consensus criteria for sepsis were used and RBC transfusions were given as single units of leukoreduced RBCs. Other outcomes, including the need for mechanical ventilation, vasopressors, renal replacement therapy, or other ischemic events, were also similar between the 2 groups. Among the enrolled patients, 7.5% had a history of hematologic malignancies and there were no differences between findings in this subgroup and the remainder of the cohort.

There are 3 trials evaluating RBC transfusion thresholds in pediatric populations. The largest trial was performed in 637 critically ill children cared for in pediatric ICUs at 19 centers. Patients were enrolled with a Hgb level of less than 9.5 g/dL and randomly allocated to transfusion at 7 g/dL or 9.5 g/dL thresholds. The primary outcome of new or progressive multiple-organ dysfunction syndrome was similar between the 2 groups. Secondary outcomes, including transfusion reactions, respiratory and catheter-related infections, duration of stay, and mortality, were also not different between groups. Two smaller trials in premature infants that included neurocognitive outcomes have been published and a definitive trial (Transfusion of Prematures) is underway in this group of patients.

In total, the findings of these trials in adult and pediatric ICUs have helped established practice recommendations for using a restrictive strategy (<7 g/dL) for RBC transfusion in critically ill patients without ischemic or congenital heart disease.

Bleeding

Patients with malignancy may have a more profound blood loss anemia related to a number of cancer-specific causes. Management of bleeding from a tumor can be complex and may be exacerbated by concomitant coagulopathy and/or thrombocytopenia. Coagulopathy occurs with increased frequency in patients with malignancy and can be owing to acquired inhibitors of coagulation, disseminated intravascular coagulation, or therapeutic anticoagulation to treat thromboembolism. Thrombocytopenia is common in hematologic malignancies and may occur as a direct effect of bone marrow infiltration or a byproduct of chemotherapy.

Data on the management of transfusion in bleeding patients are from 2 clinical trials of upper gastrointestinal bleeding. The first study enrolled 921 patients in a trial comparing a 7 g/dL threshold (restrictive) with a 9 g/dL threshold (liberal). At 6 weeks, 5% of the patients in the restrictive group died compared with 9% of the patients in the liberal group (hazard ratio, 0.55; 95% CI, 0.33–0.92; P = .02). The restrictive strategy was also associated with less rebleeding and congestive heart failure. An increased portal pressure gradient in the liberal group ( P = .03) was hypothesized to explain the increased rate of rebleeding in the liberal transfusion group that included a significant number of patients with cirrhosis. This trial was the first to report a statistically lesser mortality rate in patients in a restrictive RBC transfusion group.

Another recently published clinical trial assessed the effectiveness of transfusion strategies for acute upper gastrointestinal bleeding. The Transfusion in Gastrointestinal Bleeding (TRIGGER) trial, a pragmatic, cluster randomized feasibility trial randomized patients to a 8 g/dL (restrictive) or 10 g/dL (liberal) transfusion threshold. In a 6-month period, 936 patients were enrolled across 6 university hospitals in the United Kingdom. The cluster design led to rapid recruitment and high protocol adherence but a nonsignificant reduction in RBC transfusion in the restrictive arm. No differences in clinical outcomes were noted in this pilot trial in which 10% of enrolled subjects had a history of malignancy.

The findings of these trials in patients with upper gastrointestinal bleeding raises the possibility that hemodynamically stable patients with other sites of bleeding and who are without cardiovascular comorbidities or thrombocytopenia could be managed with a restrictive transfusion strategy safely.

Bleeding risk in relation to platelet transfusion thresholds has been well-studied in patients with malignancy; however, the optimal Hgb thresholds in thrombocytopenic patients are not known. Neither of these 2 clinical trials enrolled a significant number of patients who required platelet transfusions. Preclinical studies suggest that concomitant anemia and thrombocytopenia may compound bleeding risk, and that hemostasis can be optimized in thrombocytopenic patients by maintaining a higher hematocrit.

Two clinical trials examine RBC transfusion thresholds in patients with hematologic malignancies where thrombocytopenia is common. A pilot trial was performed in 60 patients undergoing induction chemotherapy or stem cell transplantation. Patients were randomly allocated to receive 2 units of RBCs when the Hgb concentration was less than 8 g/dL or 2 units of RBCs when the Hgb concentration was less than 12 g/dL. The main hypothesis was that there would be less bleeding and less need for platelet transfusions in the group with the higher transfusion threshold. The pilot demonstrated that such a trial was feasible but was not powered to detect clinical differences in the groups. The second trial, Transfusion of Red Cells in Hematopoietic Stem Cell Transplantation (TRIST), is enrolling patients undergoing hematopoietic stem cell transplantation, and compares transfusion thresholds of at 7 and 9 g/dL. The results of this trial, which is in progress, will include quality of life in addition to clinical outcomes.

Postoperative Patients

In parallel with clinical trial data of bleeding patients, our understanding of best practice in postoperative management of anemia comes from several studies of elderly patients with cardiovascular risk factors undergoing orthopedic surgery. Cancer patients who are candidates for surgical resection often undergo potentially life-saving procedures with the understanding that advanced age and cardiovascular comorbidities put them at increased risk of adverse outcomes as a result of significant blood loss.

The Functional Outcomes in Cardiovascular Patients Undergoing Surgical Hip Fracture Repair (FOCUS) trial enrolled elderly patients with underlying cardiovascular disease or risk factors who underwent surgical repair of hip fracture. FOCUS compared a 10 g/dL transfusion threshold with an 8 g/dL or symptoms threshold. No difference was found in the primary outcome of death or inability to walk across a room unassisted; 35.2% in the liberal group and 34.7% in the restrictive group. There were also no differences in secondary outcomes including mortality at 60 days (liberal, 7.6% vs restrictive, 6.6%), infection, function, or duration of hospital stay. Furthermore, the composite outcome of acute myocardial infarction, unstable angina, or in-hospital mortality was not different in the liberal group (4.3%) and the restrictive group (5.2%). FOCUS was the first trial to provide evidence that patients with preexisting cardiovascular disease or risk factors could be safely managed postoperatively using a restrictive transfusion strategy.

The FOCUS trial was also significant in that it included functional status in postoperative patients as an outcome of different transfusion strategies. There was no difference in the ability to walk 10 feet or across a room without assistance at the 60-day evaluation. A systematic review of this trial and 5 others in patients with orthopedic fracture came to similar conclusions regarding the safety of restrictive transfusion practice.

More recently, a small clinical trial of RBC transfusion practice enrolled 198 adult patients who underwent surgery for abdominal cancer (mostly gastrointestinal, pancreatic, or urogenital) and required postoperative intensive care. Patients were randomized to a transfusion threshold of 7 g/dL (restrictive) or 9 g/dL (liberal) during their ICU stay; 21% and 42% of patients, respectively, were transfused in the 2 groups. The primary endpoint (death or severe complication at 30 days) occurred significantly more often in the restrictive group than in the liberal group (36% vs 20%). Several individual adverse outcomes also occurred more frequently with the restrictive strategy: 30-day mortality (23% vs 8%), 60-day mortality (24% vs 11%), cardiovascular complications (14% vs 5%), and abdominal infection (15% vs 5%).

The authors of this clinical trial suggested that maintaining a Hgb concentration of greater than 9 g/dL in postoperative cancer surgery patients is beneficial, although the study was limited by sample size and power. Their findings are contrary to the larger TRICC and TRISS trials in patients in the ICU, and the FOCUS trial in postoperative patients, which favored restrictive transfusion. In addition, a systematic review of clinical trials reporting data on in-hospital infections found a reduced risk among patients transfused with restrictive versus liberal transfusion strategies (risk ratio, 0.88; 95% CI, 0.78–0.99), and this finding was most striking in postoperative patients. Although this apparent lower risk of infection may be reversed with publication of a clinical trial in cardiac surgery, it has nonetheless increased focus on postoperative complications of surgery as they relate to transfusion practice. Additional studies are required to clarify differences in outcomes that may occur in surgical oncology patients.

Cardiovascular Events

Cardiovascular events occur with increased frequency in cancer patients than the general population as a result of the toxicity associated with therapies and the prevalence of cardiovascular risk factors in older individuals. Patients with malignancy and coexistent coronary artery disease may be particularly prone to adverse outcomes–related cardiac ischemia, and whether they may benefit from a higher Hgb level remains debatable. In addition, patients with malignancy or those receiving chemotherapy are at greater risk of bleeding and may not be appropriate candidates for anticoagulation or antiplatelet agents in the setting of bleeding, thrombocytopenia, and/or coagulopathy.

In the absence of established treatments for myocardial ischemia, RBC transfusion may be beneficial by sustaining oxygen delivery to myocardial cells and decreasing myocardial oxygen demand. However, RBC transfusion could also worsen outcomes as a result of increased risk of circulatory overload or thrombogenicity with higher Hgb levels. Few clinical trials have focused on patients with malignancy, and transfusion strategies for the management of cardiovascular events in cancer patients can be extrapolated from the limited available data. Prior studies of cardiac surgery supported the safety of restrictive transfusion practice. However, a more recent clinical trial of cardiac surgery patients found greater long-term mortality in the restrictive threshold group than in the liberal group (4.2% vs 2.6%; hazard ratio, 1.64; 95% CI, 1.00–2.67; P = .045).

There have been 2 small clinical trials published that enrolled patients with acute coronary syndrome. Both trials compared transfusion triggers of 8 g/dL and 10 g/dL. In the Conservative Versus Liberal Red Cell Transfusion in Myocardial Infarction Trial (CRIT) trial, which included 45 patients, there was a higher incidence of congestive heart failure in the liberal group. In the Myocardial Ischemia and Transfusion (MINT) trial in 110 patients, there was a trend toward fewer major cardiac events and deaths in the liberal group (7 deaths in restrictive strategy and 1 death in liberal strategy; P = .03). Combining the 2 trials in acute coronary syndrome, there were 9 deaths in the restrictive group and 2 deaths in the liberal group. These trials are the first to signal that liberal transfusion might be superior to restrictive transfusion in the setting of acute coronary syndrome. However, these preliminary findings await a definitive answer in a large clinical trial.