Portal de Periódicos da CAPES

Low‐dose prophylactic platelet transfusions: time for further study, but too early for routine clinical practice

2004; Wiley; Volume: 44; Issue: 12 Linguagem: Inglês

10.1111/j.0041-1132.2004.04378.x

1537-2995

Ronald G. Strauss,

Hematopoietic Stem Cell Transplantation

In this issue of TRANSFUSION, Tinmouth and colleauges1 report results of a randomized, controlled clinical trial in which adult patients undergoing autologous hematopoietic progenitor cell (HPC) transplantation (n = 77) or induction chemotherapy for acute leukemia (n = 34) received prophylactic platelet transfusions (PTX) at either a low dose (a pool of three whole blood–derived platelet [PLT] units or a partial apheresis unit) or a standard dose (a pool of five whole blood–derived PLT units or a complete apheresis unit). Leukoreduced, whole blood–derived PLT units were prepared by the PLT-rich plasma method followed by prestorage filtration, and leukoreduced, apheresis PLT units were prepared using a blood separator (Spectra, Gambro Canada Ltd, Richmond Hill, Ontario, Canada). Unfortunately, neither the actual number of PLTs transfused per each transfusion nor the immediate posttransfusion blood PLT count increments were reported. Per quality control data, however, each low-dose PTX was estimated to contain approximately 2 × 1011 PLTs and the standard PTX dose slightly less than 4 × 1011 PLTs. Based on rates of minor bleeding (occurring in 20% of low-dose patients vs. 40% of standard-dose patients) and major bleeding (10.7% of low-dose patients vs. 7.3% of standard-dose patients) plus the transfusion of 25 percent fewer PLT units to low-dose patients, the authors concluded that low-dose prophylactic PTX appeared to be effective and safe, while reducing PLT unit utilization.1 In general, the experimental design, conduct of the study, and the conclusions seem reasonable. As is true for many clinical trials involving complex patients, however, the results are not definitive, and the authors rightfully recommended that larger, well-designed clinical trials are still needed to determine optimal PTX doses for clinical practice. Some of the possible shortcomings of the trial are as follows: Of the 111 patients enrolled on an intention to treat policy, 14 of 56 (25%) in the low-dose group and 10 of 55 (18%) in the standard-dose group actually received no PTX. In addition, one patient in the low-dose group and four patients in the standard-dose group were withdrawn before reaching study endpoints. Thus, the comparative effects of the two different PLT doses were assessed in study groups of only 41 patients per group with a variety of underlying diagnoses (i.e., fairly small numbers of heterogeneous patients) — despite stated sizes (i.e., “denominators” for calculations) of 56 and 55 patients for low and standard groups, respectively.1 Although prophylactic PTXs were to be ordered if the pretransfusion blood PLT count was not greater than 10 × 109 per L, PTX could also be ordered at higher PLT counts for minor bleeding or invasive procedures. In addition, standard-dose PTXs were given to patients — even if assigned to the low-dose group — if they experienced major bleeding, required additional chemotherapy, failed to engraft, or were admitted to an intensive care unit.1 This flexibility in ordering PTX plus the fact that prescribing physicians were not blinded to PTX doses given make it difficult to be convinced that low- and standard-dose prophylactic PTXs were proven definitively to be truly comparable. Finally, the use of sequential Bayesian design and statistical analyses will create problems — owing to lack of familiarity — for many readers as they try to judge whether or not the results, conclusions, and recommendations can be justified by the data. Because of all of these factors, it seems premature to embrace low-dose prophylactic PTX as a standard of practice for patients with severe thrombocytopenia attributed to marrow failure. About 2 million PTXs are given annually in the US at an estimated cost exceeding 1 billion dollars.2 Although PTXs are prescribed frequently for patients with severe thrombocytopenia attributed to marrow failure, specific details of PTX practices remain controversial and vary widely among individual physicians. Despite quibbling about details of the fine points of practice, the general guiding principles for PTX listed below likely would be judged as being reasonable by most practitioners (reviewed in Heal and Blumberg,2 Strauss,3 and Slichter4). Therapeutic PTXs clearly are efficacious to treat overt bleeding in patients with severe thrombocytopenia attributed to marrow failure. Before PTX became available, nearly all leukemic patients undergoing chemotherapy exhibited bleeding to some degree with more than 50 percent experiencing severe or fatal hemorrhages. After availability of PTX, life-threatening and fatal bleeding during initial induction and/or consolidation chemotherapy or following HPC transplantation has been reduced to less than 15 percent. The usual goal for therapeutic PTX is to maintain the blood PLT count of at least 40 × 109 per L until bleeding ceases. Prophylactic PTX delay the onset and/or lessen the likelihood of serious spontaneous bleeding in some patients with severe thrombocytopenia attributed to marrow failure. The efficacy of prophylactic PTXs, however, has not been convincingly demonstrated by large, well-designed clinical trials in patients treated with modern chemotherapy or peripheral blood hematopoietic progenitor infusions, in which the period of severe thrombocytopenia often is relatively brief — compared to the longer periods of thrombocytopenia following less effective chemotherapy (i.e., at the time earlier prophylactic PTX studies were performed) or marrow transplants. The customary goal of prophylactic PTX has been to maintain the blood PLT count of at least 20 × 109 per L. Several studies, however, have suggested lower goals of 10 × 109 and 5 × 109 per L for certain patients who are clinically stable and free of factors that place them at relatively high risk for bleeding. When transfused in comparable doses, PTXs using whole blood–derived and apheresis PLT units have equivalent efficacy and, when prepared similarly (i.e., leukoreduced, similar storage and transfusion techniques, monitored for bacterial contamination, pathogen-reduction), have similar costs.5 In contrast to the generally acceptable principles stated above, the optimal PLT dose for PTX is quite controversial.2-5 For therapeutic PTX given to stop overt bleeding or to prevent it during invasive procedures, all agree that the posttransfusion blood PLT counts should be at least 40 × 109 per L whenever achievable — with the controversy being that some physicians desire PLT counts as high as 100 × 109 per L. Regarding the desirability of high posttransfusion blood PLT counts, I have been unable to find data from properly designed clinical trials to justify the benefit of maintaining blood PLT counts of greater than 50 × 109 per L in patients in whom bleeding is caused by thrombocytopenia. For prophylactic PTXs, the controversy revolves around two related issues: 1) criteria that define “stable, low-risk” patients; and 2) the pretransfusion blood PLT count at which prophylactic PTXs should be given. As reviewed recently,4 several reports have suggested that clinically stable patients, without factors placing them at high risk of spontaneous bleeding, need lower doses of PLTs to prevent serious bleeding than unstable patients who may exhibit or be exposed to risk factors such as existing mucocutaneous bleeding, infection and antimicrobial drugs, graft-versus-host disease, liver and/or renal dysfunction, or abnormal coagulation test results. In practice, some of these factors are evident in nearly all patients to some degree. Moreover, the factors often are fairly subjective and not easily quantitated. Thus, it is difficult to get all physicians to categorize patients uniformly. This categorization of patients is important because several clinical trials have documented that prophylactic PTXs can be given safely, effectively, and more economically to clinically stable patients at a pretransfusion blood PLT count of 10 × 109 per L, rather than at 20 × 109 per L.4 Some studies have suggested that a pretransfusion blood PLT count of 5 × 109 per L might be satisfactory for some clinically stable patients (reviewed in Slichter4). Thus, identifying stable, low-risk patients is important when prescribing PTXs per PLT dose. Inherent in the acceptance of relatively low pretransfusion blood PLT counts of 5 × 109 to 10 × 109 per L is acknowledgment that relatively high posttransfusion blood PLT counts are not required to prevent spontaneous, serious bleeding in clinically stable thrombocytopenic patients. This is consistent with the suggestion that simply transfusing the low number of PLTs needed to “plug” the endothelium (i.e., 7.1 × 109 per L of patient blood volume per day) will suffice as a prophylactic PTX for clinically stable patients.4 These factors, of course, support the conclusions of Tinmouth and coworkers1 favoring low-dose prophylactic PTX.1 In contrast, those who favor higher dose prophylactic PTX feel that relatively high posttransfusion blood PLT counts are more likely to protect patients from serious spontaneous bleeding and, because blood PLT counts are maintained at levels of at least 20 × 109 per L for relatively long periods of time, permit fairly long intervals between individual PTX (reviewed in Strauss3). In an attempt to resolve the controversy of lower versus higher prophylactic PTX practices and to definitively establish the optimal PLT dose for prophylactic PTXs in patients with severe thrombocytopenia attributed to marrow failure, the multicenter Transfusion Medicine/Hemostasis Network, sponsored by the National Heart, Lung, & Blood Institute of the National Institutes of Health, is in the midst of a prospective, randomized, blinded, clinical trial in which patients with thrombocytopenia will be assigned to receive prophylactic PTXs at lower, medium, or higher doses. A total of 1350 patients, expected to require at least five PTXs for pretransfusion blood PLT counts of less than 10 × 109 per L, will be given PTXs each containing an average of 1.1 × 1011 PLTs per square meter of body surface area (lower dose), 2.2 × 1011 per m2 (medium dose), or 4.4 × 1011 per m2 (higher dose). The primary endpoint of the trial is the percentage of patients in each treatment arm who develop at least 1 day of Grade 2 or higher bleeding (modified World Health Organization bleeding scale). Also, several secondary endpoints will provide additional information pertaining to PTX efficacy, adverse effects, and application to clinical practices. Based on all of the information presented above, I recommend two approaches. First, for routine clinical practice, therapeutic and prophylactic PTXs should continue to be given in the standard or high doses necessary to achieve the posttransfusion blood PLT counts deemed efficacious for each patient's clinical status. As a practical point, the posttransfusion blood PLT count can be roughly estimated — based on the principles of the corrected count increment calculations — that each 1 × 1011 PLTs transfused per each square meter of body surface area will increase the blood PLT count by 8 × 109 to 10 × 109 per L 1 hour or so after transfusion.3 Thus, to increase the blood PLT count by 50 × 109 per L, a 1.5-m2 adult patient would need a PTX containing approximately 7 × 1011 PLTs. Second, because of information reported by several investigators including Tinmouth et al.,1,2,4 it is quite reasonable to study the feasibility, efficacy, and safety of low-dose prophylactic PTXs in patients who have given informed consent. If lower dose prophylactic PTXs are determined to be efficacious and safe by the Transfusion Medicine/Hemostasis Network Clinical Trial, a logical next step is to study whether prophylactic PTX are needed at all in clinically stable patients without risk factors for bleeding (i.e., low-dose PTX vs. no prophylactic PTX).