Practice-Defining Clinical Trials in Nonmalignant Hematology

During 2015, two important myths from the nonmalignant hematology world were busted (to use a verb from the title of a recently popular television series, “Myth Busters”). In both cases, the myth was based on biological theory and observational data from uncontrolled studies, while the definitive evidence emerged from one or more prospective, randomized controlled trials.

The first myth consists of the idea that patients who are anticoagulated for so-called high-risk (but hemodynamically stable) pulmonary embolism (PE) might benefit from the placement of a retrievable inferior vena cava (IVC) filter. Indeed, the general notion that more aggressive interventions (over and above anticoagulant therapy) might benefit high-risk PE patients is not new. For example, more than one randomized controlled trial of thrombolysis plus anticoagulation versus anticoagulation alone has been conducted in this population. For some clinicians, the retrievable IVC filter, an intervention with no bleeding risk that theoretically could protect an already tenuous right ventricle from the next embolism, seemed even more attractive, especially after an observational study in PE patients documented an association between IVC filter placement and improved survival.¹  Thankfully, a group of investigators was able to carry out a prospective, randomized controlled trial in which patients with high-risk PE were randomly assigned to have (or not have) a retrievable IVC filter placed; all participants received anticoagulation.²  The patients who received a filter had no improvement on any of the efficacy outcomes, and a handful experienced some of the complications known to occur after IVC filter placement such as access site hematoma, filter thrombosis, and retrieval failure. This study, by Dr. Patrick Mismetti and colleagues, should serve as a reminder to all hematologists that, when administered with careful monitoring at a high-quality center, anticoagulation therapy is as effective as more expensive and potentially hazardous treatment options, sometimes considered for patients with high-risk pulmonary embolism. I suspect that Drs. D.W. Barritt and Stanley C. Jordan, the authors who performed the landmark randomized trial of heparin vs. no heparin in patients with PE more than 50 years ago,³  would be pleased to learn that the therapy they tested remains the gold standard in 2015.

The second myth to be challenged was that, among patients who require red cell transfusions, those who receive “fresher” units of packed red blood cells (RBCs) fare better than patients who receive RBC units of “standard” age. This idea that fresher blood might be better has a basis in biology⁴  and observational data.⁵  However, in 2015, two randomized trials involving a cumulative total of more than 3,500 patients^6,7  failed to demonstrate any evidence of a mortality (or other clinical) benefit among those who received “fresher” RBCs. In one study, the mean ages of RBCs were six versus 22 days, and in the other, the median ages were seven versus 28 days. The studies involved critically ill adult patients and patients (12 years or older) undergoing complex cardiac surgery; all RBCs administered were leukoreduced. Like any data from a clinical trial, the evidence provided by these large trials can certainly be criticized: What if more RBC units older than 35 days had been tested? Should the variable of interest (age of RBCs) have been treated as continuous rather than categorical (dichotomous)? Those reservations notwithstanding, the sheer size of these trials, along with the degree to which the findings agree with previous controlled studies, indicates that if there is any benefit from RBCs with storage time longer than 28 days, it is small, and the additional costs of providing “fresher” blood to all patients would be very difficult to justify.