Richard M. Kaufman, MD; Benjamin Djulbegovic, MD, PhD; Terry Gernsheimer, MD; Steven Kleinman, MD; Alan T. Tinmouth, MD; Kelley E. Capocelli, MD; Mark D. Cipolle, MD, PhD; Claudia S. Cohn, MD, PhD; Mark K. Fung, MD, PhD; Brenda J. Grossman, MD, MPH; Paul D. Mintz, MD; Barbara A. O'Malley, MD; Deborah A. Sesok-Pizzini, MD; Aryeh Shander, MD; Gary E. Stack, MD, PhD; Kathryn E. Webert, MD, MSc; Robert Weinstein, MD; Babu G. Welch, MD; Glenn J. Whitman, MD; Edward C. Wong, MD; Aaron A.R. Tobian, MD, PhD
* This article was published online first at www.annals.org on 11 November 2014.
Acknowledgment: The authors thank Theresa Wiegmann for her outstanding skill and dedication in guiding this project and Jacqlyn Riposo for her superb logistic support.
Disclosures: Disclosures can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M14-1589.
Requests for Single Reprints: Richard M. Kaufman, MD, Department of Pathology, Brigham and Women's Hospital, Blood Bank, Amory 260, 75 Francis Street, Boston, MA 02115; e-mail, email@example.com.
Current Author Addresses: Dr. Kaufman: Department of Pathology, Brigham and Women's Hospital, Blood Bank, Amory 260, 75 Francis Street, Boston, MA 02115.
Dr. Djulbegovic: University of South Florida, 3515 East Fletcher Avenue, Health/Therapy 1201, Health/College of Medicine 27, Tampa, FL 33612.
Dr. Gernsheimer: University of Washington, 1959 NE Pacific Street, Box 356330, Seattle, WA 98195.
Dr. Kleinman: University of British Columbia, 1281 Rockcrest Avenue, Victoria, British Columbia V9A 4W4, Canada.
Dr. Tinmouth: Clinical Epidemiology Research Unit, Ottawa Hospital Research Institute, General Campus, Box 201, Room 1812-C, 501 Smyth Road, Ottawa, Ontario K1H 8L6, Canada.
Dr. Capocelli: Department of Pathology, Children's Hospital Colorado, B120, Aurora, CO 80045.
Dr. Cipolle: Christiana Care Health System, Surgical and Critical Care Associates, 4755 Ogletown-Stanton Road, Suite 1320, Newark, DE 19713.
Dr. Cohn: Department of Laboratory Medicine and Pathology, University of Minnesota, Mayo D242, Mayo Mail Code 609, 420 Delaware Street Southeast, Minneapolis, MN 55455.
Dr. Fung: Department of Pathology, University of Vermont and Fletcher Allen Health Care, 111 Colchester Avenue, Burlington, VT 05401.
Dr. Grossman: Department of Pathology and Immunology, Washington University School of Medicine, 660 South Euclid Avenue, Campus Box 8118, St. Louis, MO 63110.
Dr. Mintz: Division of Hematology Clinical Review, Center for Biologics Evaluation and Research, U.S. Food and Drug Administration, 10903 New Hampshire Avenue, Silver Spring, MD 20993.
Dr. O'Malley: Department of Pathology, Wayne State University School of Medicine, 3990 John R. Road, Harper University Hospital, Detroit Medical Center, Detroit, MI 48202.
Dr. Sesok-Pizzini: Children's Hospital of Philadelphia, 5136 Main Hospital, 34th Street and Civic Center Boulevard, Philadelphia, PA 19104-4399.
Dr. Shander: Department of Anesthesiology and Critical Care Medicine, Englewood Hospital and Medical Center, 350 Engle Street, Englewood, NJ 07631.
Dr. Stack: Yale School of Medicine, Pathology and Laboratory Medicine Service/113, 950 Campbell Avenue, West Haven, CT 06516-2770.
Dr. Webert: Canadian Blood Services, 35 Stone Church Road, Suite 200, Ancaster, Ontario L9K 1S5, Canada.
Dr. Weinstein: University of Massachusetts Medical School, 55 Lake Avenue North, LA-113, Worcester, MA 01655.
Dr. Welch: University of Texas Southwestern Medical Center, 5161 Harry Hines Boulevard, CS5.112, Dallas, TX 75390-8855.
Dr. Whitman: Division of Cardiac Surgery, Johns Hopkins University, Suite 7107/Zayed Tower, 1800 Orleans Street, Baltimore, MD 21287.
Dr. Wong: Division of Laboratory Medicine, Children's National Medical Center, 111 Michigan Avenue NW, Washington, DC 20010.
Dr. Tobian: Department of Pathology, Division of Transfusion Medicine, Johns Hopkins University, Carnegie 437, 600 North Wolfe Street, Baltimore, MD 21287.
Author Contributions: Conception and design: R.M. Kaufman, B. Djulbegovic, T. Gernsheimer, S. Kleinman, A.T. Tinmouth, B.J. Grossman, P.D. Mintz, D.A. Sesok-Pizzini, G.E. Stack, K.E. Webert, R. Weinstein, A.A.R. Tobian.
Analysis and interpretation of the data: R.M. Kaufman, B. Djulbegovic, T. Gernsheimer, S. Kleinman, A.T. Tinmouth, K.E. Capocelli, C.S. Cohn, M.K. Fung, B.J. Grossman, P.D. Mintz, B.A. O'Malley, D.A. Sesok-Pizzini, A. Shander, G.E. Stack, K.E. Webert, R. Weinstein, B.G. Welch, G.J. Whitman, E.C. Wong, A.A.R. Tobian.
Drafting of the article: R.M. Kaufman, B. Djulbegovic, T. Gernsheimer, S. Kleinman, A.T. Tinmouth, K.E. Capocelli, M.D. Cipolle, D.A. Sesok-Pizzini, A. Shander, B.G. Welch, A.A.R. Tobian.
Critical revision of the article for important intellectual content: R.M. Kaufman, B. Djulbegovic, T. Gernsheimer, S. Kleinman, A.T. Tinmouth, K.E. Capocelli, M.D. Cipolle, M.K. Fung, B.J. Grossman, P.D. Mintz, B.A. O'Malley, D.A. Sesok-Pizzini, A. Shander, K.E. Webert, R. Weinstein, G.J. Whitman, E.C. Wong, A.A.R. Tobian.
Final approval of the article: R.M. Kaufman, B. Djulbegovic, T. Gernsheimer, S. Kleinman, A.T. Tinmouth, K.E. Capocelli, C.S. Cohn, M.K. Fung, B.J. Grossman, P.D. Mintz, B.A. O'Malley, D.A. Sesok-Pizzini, A. Shander, K.E. Webert, R. Weinstein, B.G. Welch, E.C. Wong, A.A.R. Tobian.
Provision of study materials or patients: B. Djulbegovic.
Statistical expertise: B. Djulbegovic.
Administrative, technical, or logistic support: M.K. Fung, A.A.R. Tobian.
Collection and assembly of data: B. Djulbegovic, G.E. Stack, A.A.R. Tobian.
Kaufman RM, Djulbegovic B, Gernsheimer T, Kleinman S, Tinmouth AT, Capocelli KE, et al. Platelet Transfusion: A Clinical Practice Guideline From the AABB. Ann Intern Med. 2015;162:205-213. doi: 10.7326/M14-1589
Download citation file:
Published: Ann Intern Med. 2015;162(3):205-213.
The AABB (formerly, the American Association of Blood Banks) developed this guideline on appropriate use of platelet transfusion in adult patients.
These guidelines are based on a systematic review of randomized, clinical trials and observational studies (1900 to September 2014) that reported clinical outcomes on patients receiving prophylactic or therapeutic platelet transfusions. An expert panel reviewed the data and developed recommendations using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework.
The AABB recommends that platelets should be transfused prophylactically to reduce the risk for spontaneous bleeding in hospitalized adult patients with therapy-induced hypoproliferative thrombocytopenia. The AABB recommends transfusing hospitalized adult patients with a platelet count of 10 × 109 cells/L or less to reduce the risk for spontaneous bleeding. The AABB recommends transfusing up to a single apheresis unit or equivalent. Greater doses are not more effective, and lower doses equal to one half of a standard apheresis unit are equally effective. (Grade: strong recommendation; moderate-quality evidence)
The AABB suggests prophylactic platelet transfusion for patients having elective central venous catheter placement with a platelet count less than 20 × 109 cells/L. (Grade: weak recommendation; low-quality evidence)
The AABB suggests prophylactic platelet transfusion for patients having elective diagnostic lumbar puncture with a platelet count less than 50 × 109 cells/L. (Grade: weak recommendation; very-low-quality evidence)
The AABB suggests prophylactic platelet transfusion for patients having major elective nonneuraxial surgery with a platelet count less than 50 × 109 cells/L. (Grade: weak recommendation; very-low-quality evidence)
The AABB recommends against routine prophylactic platelet transfusion for patients who are nonthrombocytopenic and have cardiac surgery with cardiopulmonary bypass. The AABB suggests platelet transfusion for patients having bypass who exhibit perioperative bleeding with thrombocytopenia and/or evidence of platelet dysfunction. (Grade: weak recommendation; very-low-quality evidence)
The AABB cannot recommend for or against platelet transfusion for patients receiving antiplatelet therapy who have intracranial hemorrhage (traumatic or spontaneous). (Grade: uncertain recommendation; very-low-quality evidence)
Approximately 2.2 million platelet doses are transfused annually in the United States (1). A high proportion of these platelet units are transfused prophylactically to reduce the risk for spontaneous bleeding in patients who are thrombocytopenic after chemotherapy or hematopoietic progenitor cell transplantation (HPCT) (1–3). Unlike other blood components, platelets must be stored at room temperature, limiting the shelf life of platelet units to only 5 days because of the risk for bacterial growth during storage. Therefore, maintaining hospital platelet inventories is logistically difficult and highly resource-intensive (4, 5). Platelet transfusion is associated with several risks to the recipient (Table 1), including allergic reactions and febrile nonhemolytic reactions. Sepsis from a bacterially contaminated platelet unit represents the most frequent infectious complication from any blood product today (8). In any situation where platelet transfusion is being considered, these risks must be balanced against the potential clinical benefits.
Table 1. Approximate Per-Unit Risks for Platelet Transfusion in the United States
These guidelines were designed to provide pragmatic recommendations, based on the best available published evidence, about when platelet transfusion may be appropriate in adult patients. For several common clinical situations, we attempted to identify a platelet count threshold below which platelet transfusion may improve hemostasis and above which platelet transfusion is unlikely to benefit the patient. We did not attempt to address all clinical situations in which platelets may be transfused, and these guidelines are not intended to serve as standards. Clinical judgment, and not a specific platelet count threshold, is paramount in deciding whether to transfuse platelets.
These guidelines provide advice for adult patients who are candidates for platelet transfusion.
The AABB commissioned and funded the development of these guidelines.
A panel of 21 experts was convened. Fifteen participants were members of the Clinical Transfusion Medicine Committee of the AABB, all of whom were hematologists or pathologists with expertise in transfusion medicine. Five additional panel members included a neurosurgeon, a cardiac surgeon, a critical care specialist, an anesthesiologist, and a hematologist, representing the American Association of Neurological Surgeons, the Society of Thoracic Surgeons, the Society of Critical Care Medicine, the American Society of Anesthesiologists, and the American Society of Hematology, respectively. The final panel member was a Grading of Recommendations Assessment, Development and Evaluation (GRADE) methodologist. Committee members had no substantial conflicts of interest as defined by the AABB conflict of interest policy. Pursuant to the policy, individual members were required to disclose actual and apparent financial, professional, or personal conflicts (Appendix Table 1).
Appendix Table 1. Panel Members' Conflicts of Interest
The guidelines were developed on the basis of a recent systematic review of the literature on platelet transfusions, published separately (11). The search strategy is provided in Appendix Table 2. We searched PubMed from 1946 to the first week of April 2013, and the Cochrane Central Register of Controlled Trials and Web of Science from 1900 to the first week of April 2013 (1024 studies identified). An updated search of these databases was done from the first week of April 2013 to the first week of September 2014. Randomized, controlled trials (RCTs) and observational studies (prospective or retrospective cohort studies, case–control studies, and those with no control group) were eligible for inclusion. Outcomes of interest included all-cause mortality, bleeding-related mortality, bleeding, and number of platelet units transfused. Although all observational studies meeting the inclusion criteria were reviewed, data from observational studies were not used when more than 2 RCTs addressed a particular question. There were no language restrictions. After exclusions, 17 RCTs and 53 observational studies were included in the final systematic review. Only 1 relevant observational study (12) from the updated search was identified, and evidence from this study did not change our GRADE judgments of evidence quality or recommendation strength.
Appendix Table 2. Search Strategy Used for Systematic Review of the Literature
The GRADE method was used to assess the quality of the evidence and determine the strength of recommendations (13, 14). The recommendations were developed by consensus at an in-person panel meeting. Panel member judgments on 4 GRADE factors (quality of evidence, balance between the intervention's benefits and harms, resource use, and patient values and preferences) and ratings of the strength of recommendations were validated using an online survey tool 1 week after the meeting.
In this guideline, a platelet unit refers to 1 apheresis platelet unit or a pool of 4 to 6 whole blood–derived platelet concentrates, typically containing 3 to 4 × 1011 platelets. Thrombocytopenia refers to a platelet count below the lower limit of the normal range used by the laboratory performing the count. Seven platelet trials included in the systematic review (15–21) used a variation of the World Health Organization scale (22) to assess patient bleeding outcomes (23). A summary of the modified World Health Organization scale is provided in Table 2 .
Table 2. Summary of the Modified WHO Bleeding Scale
Recommendation 1: The AABB recommends that platelets should be transfused prophylactically to reduce the risk for spontaneous bleeding in adult patients with therapy-induced hypoproliferative thrombocytopenia.
The AABB recommends transfusing hospitalized adult patients with a platelet count of 10 × 109 cells/L or less to reduce the risk for spontaneous bleeding.
The AABB recommends transfusing up to a single apheresis unit or equivalent. Greater doses are not more effective, and lower doses equal to one half of a standard apheresis unit are equally effective.
Quality of evidence: moderate; strength of recommendation: strong.
Three RCTs (n = 1047) compared bleeding outcomes in hospitalized patients with radiation and/or chemotherapy-induced hypoproliferative thrombocytopenia assigned to receive or not receive prophylactic platelet transfusions (Appendix Table 3) (19, 21, 24, 25). All patients had hematologic malignancy treated with chemotherapy or HPCT. Prophylactic platelet transfusions were found to significantly reduce the risk for spontaneous grade 2 or greater bleeding (odds ratio [OR], 0.53 [95% CI, 0.32 to 0.87]). Most bleeding events were classified as grade 2. In the 2 largest trials (19, 21), grade 2 or greater bleeding in patients assigned to the group that did not receive prophylaxis occurred more frequently among patients receiving chemotherapy for acute leukemia compared with autologous HPCT recipients (58% vs. 47% [19, 25]; 51% vs. 28% ).
Appendix Table 3. Prophylactic Platelet Transfusion Versus No Prophylactic Platelet Transfusion in Therapy-Induced Hypoproliferative Thrombocytopenia
The threshold platelet count at which platelets should be transfused prophylactically to reduce the bleeding risk in hospitalized patients with therapy-induced hypoproliferative thrombocytopenia was examined in 4 RCTs (n = 658) (Appendix Table 4). Patients were assigned to receive prophylactic platelet transfusion for a morning platelet count less than 10 × 109 versus 20 × 109 cells/L (26–28) or 30 × 109 cells/L (15). A greater platelet count threshold (20 × 109 or 30 × 109 cells/L) was not associated with a significantly lower incidence of grade 2 or greater bleeding (OR, 0.74 [CI, 0.41 to 1.35]) or bleeding-related mortality (OR, 0.37 [CI, 0.02 to 9.22]). The total number of days with bleeding was greater in the 10 × 109–cells/L threshold group. The 10 × 109–cells/L threshold was associated with lower platelet usage and fewer transfusion reactions.
Appendix Table 4. Higher Versus Lower Platelet Count Thresholds for Prophylactic Platelet Transfusions in Therapy-Induced Hypoproliferative Thrombocytopenia
Four RCTs (n = 1132) (Appendix Table 5) examined whether prophylactic transfusion of low-dose platelets (defined as approximately one half of the standard dose of 3 to 4 × 1011 platelets) would provide hemostasis equal to that of standard-dose platelets in patients with therapy-induced hypoproliferative thrombocytopenia (16, 18, 20, 29). There was no difference in grade 2 or greater bleeding in recipients of standard-dose versus low-dose platelets (OR, 0.91 [CI, 0.70 to 1.19]). High-dose platelets (approximately double the standard dose) were compared with standard-dose platelets in 2 RCTs (n = 951) (Appendix Table 6) (17, 18). Prophylactic transfusion of high-dose platelets did not reduce the risk for bleeding compared with standard-dose platelets (OR, 1.05 [CI, 0.79 to 1.40]).
Appendix Table 5. Standard-Dose Versus Low-Dose Prophylactic Platelet Transfusions in Therapy-Induced Hypoproliferative Thrombocytopenia
Appendix Table 6. High-Dose Versus Standard-Dose Prophylactic Platelet Transfusions in Therapy-Induced Hypoproliferative Thrombocytopenia
Before routine platelet prophylaxis was introduced, severe hemorrhage was a common cause of death among patients receiving high-dose chemotherapy (30, 31). Today, severe hemorrhage is rarely encountered in this setting. The original studies of platelet prophylaxis were done decades ago, and both chemotherapy and supportive care for patients with cancer have changed dramatically over time. Therefore, the randomized trials reported by Wandt (21) and Stanworth (19) and their colleagues were designed to answer the question of whether a prophylactic as compared with a therapeutic platelet transfusion strategy provides benefit in contemporary cancer care. In the study by Wandt and colleagues (21), grade 2 or greater bleeding was seen in 42% of patients assigned to receive therapeutic platelet transfusions only, compared with 19% of patients assigned to receive prophylactic platelet transfusion for a platelet count of 10 × 109 cells/L or less (P < 0.001). In the subset of patients with acute myelogenous leukemia, intracerebral bleeding (grade 4) occurred significantly more often in the therapeutic platelet group compared with the prophylactic platelet group (7% vs. 2%; P = 0.010). In 11 of 13 cases, intracerebral bleeding was detectable on CT scan, but there were no apparent clinical sequelae. Computed tomography scans to investigate new headache or other cerebral symptoms were required only for patients in the therapeutic platelet group, so subclinical intracerebral hemorrhage in the prophylactic platelet group may have been underdiagnosed. In the Trial of Prophylactic Platelets (19), subtler differences in bleeding outcomes were seen between the study groups. Grade 2 or greater bleeding occurred in 50% of patients assigned to the group that did not receive prophylaxis, compared with 43% of patients receiving prophylactic platelet transfusions (P = 0.06 for noninferiority). In patients receiving chemotherapy (not HPCT), there was a significant increase in grade 2 or greater bleeding in the group that did not receive prophylaxis (risk difference, 20% [90% CI, 7.9% to 32.2%]). There was also a nonsignificant trend toward increased grade 3 and 4 bleeding for all patients in the group that did not receive prophylaxis. Thus, both the Wandt trial and the Trial of Prophylactic Platelets support the continued use of prophylactic platelet transfusions in patients with therapy-induced hypoproliferative thrombocytopenia. In this population, we recommend prophylactic platelet transfusion for a morning platelet count of 10 × 109 cells/L or less. Some data suggest that the risk for spontaneous bleeding does not increase until the platelet count decreases to less than approximately 6 × 109 cells/L (18, 32), but the 10 × 109–cells/L platelet count threshold seems to provide a good balance of safety and practicality, and the accuracy of extremely low platelet count measurements is questionable (33, 34). The recommendation for prophylactic platelet transfusion based on a 10 × 109–cells/L platelet count threshold applies to hospitalized patients only. Prophylactic platelet transfusion based on a more liberal (greater) platelet count threshold may be appropriate when treating outpatients, for reasons of practicality (fewer clinic visits).
The Platelet Dose study (18) established that patients receiving low-dose prophylactic platelet transfusions for a morning platelet count of 10 × 109 cells/L or less had the same bleeding risk as patients receiving standard- or high-dose platelets. However, low-dose platelets did need to be transfused more often because they provided a lower increment. It is safe to provide low-dose platelet prophylaxis to patients with therapy-induced hypoproliferative thrombocytopenia, either routinely or as a temporary maneuver in times of platelet shortage. High-dose prophylactic platelet transfusions have not been shown to provide additional benefit, so they are not recommended as routine therapy for inpatients.
Recommendation 2: The AABB suggests prophylactic platelet transfusion for patients having elective central venous catheter placement with a platelet count less than 20 × 109 cells/L.
Quality of evidence: low; strength of recommendation: weak.
Recommendation 3: The AABB suggests prophylactic platelet transfusion for patients having elective diagnostic lumbar puncture with a platelet count less than 50 × 109 cells/L.
Quality of evidence: very low; strength of recommendation: weak.
Eight observational studies of central venous catheter (CVC) placement in the setting of thrombocytopenia were identified (n = 1311 cannulations) (Appendix Table 7) (12, 35–41). Many patients had acute leukemia or were having HPCT; however, patients with renal failure, critically ill patients, and others were included. Overall bleeding complication rates were low, ranging from 0% to 9% of catheter placements. The largest series of nontunneled CVC placements included 604 cannulations in 193 consecutive patients (41). In multivariate analysis, only patients with preprocedure platelet counts less than 20 × 109 cells/L (n = 93) were at increased risk for bleeding compared with patients with platelet counts greater than 100 × 109 cells/L. Ninety-six percent of bleeding events were grade 1, and the remaining 4% of bleeding events were grade 2, requiring only local compression. In another single-center study, bleeding outcomes were reported on 3170 tunneled CVCs placed under ultrasonography guidance in 2512 patients (38). No bleeding complications occurred in the 344 CVC placements performed with a preprocedure platelet count less than 50 × 109 cells/L, including 42 cases with a platelet count less than 25 × 109 cells/L.
Appendix Table 7. Prophylactic Platelet Transfusion for Central Venous Catheter Placement
Data from 7 observational studies of children or adults who were thrombocytopenic and had diagnostic or therapeutic lumbar puncture (LP) were evaluated (Appendix Table 8) (42–49). The largest was a single-center observational study of 5223 LPs in 956 pediatric patients with acute lymphoblastic leukemia (45). A total of 199 LPs were performed with platelet counts of 20 × 109 cells/L or less, and 742 LPs were performed with platelet counts between 21 × 109 cells/L and 50 × 109 cells/L. No bleeding complications were seen, regardless of platelet count. The upper 95% CI for serious complications was 1.75% for patients with platelet counts of 20 × 109 cells/L or less and 0.37% for patients with platelet counts of 50 × 109 cells/L or less. Traumatic LP (>500 red blood cells per high-power field) occurred in 10.5% of procedures but was not associated with adverse clinical outcomes. The largest reported series in adults included 195 diagnostic or therapeutic LPs in 66 adult patients with acute leukemia and thrombocytopenia (49). Patients were prophylactically transfused with platelets for a preprocedure platelet count less than 20 × 109 cells/L. Thirty-five LPs were performed in patients with platelet counts of 20 × 109 to 30 × 109 cells/L, and 40 were done with platelet counts of 31 × 109 to 50 × 109 cells/L. No bleeding complications were seen.
Appendix Table 8. Prophylactic Platelet Transfusion Versus No Prophylactic Platelet Transfusion for Lumbar Puncture
Serious bleeding complications after CVC placement are rare, and when they occur, they are often unrelated to the platelet count (such as accidental arterial puncture). In aggregate, the existing data support the use of a 20 × 109–cells/L platelet count threshold for CVC placement. The reported studies included patients with a wide range of primary diagnoses; this recommendation is intended to be broadly applicable to adult patients with hypoproliferative thrombocytopenia.
Bleeding complications are rare with LPs, but hemorrhage anywhere in the central nervous system has the potential to cause devastating neurologic sequelae. In the absence of better published data supporting the safety of a lower threshold in adult patients, a fairly liberal platelet count threshold for LPs (that is, 50 × 109 cells/L) seems prudent. The 50 × 109–cells/L threshold is intended for simple diagnostic or therapeutic LPs only. Despite a lack of supportive data, a greater platelet count is often recommended for other procedures, such as epidural anesthesia (50, 51).
Recommendation 4: The AABB suggests prophylactic platelet transfusion for patients having major elective nonneuraxial surgery with a platelet count less than 50 × 109 cells/L.
Recommendation 5: The AABB recommends against routine prophylactic platelet transfusion for patients who are nonthrombocytopenic and have cardiac surgery with cardiopulmonary bypass (CPB). The AABB suggests platelet transfusion for patients having CPB who exhibit perioperative bleeding with thrombocytopenia and/or with evidence of platelet dysfunction.
In 1 series (Appendix Table 9) (52), 95 patients with acute leukemia and thrombocytopenia had 167 invasive procedures, including 29 major surgeries (such as thoracotomy) and 24 moderately invasive procedures (such as arteriovenous fistula construction). Platelet prophylaxis was given before the 130 procedures in which the preoperative platelet count was less than 50 × 109 cells/L. The median postoperative platelet count in these cases was 56 × 109 cells/L. Intraoperative blood loss greater than 500 mL occurred in only 7% of all operations, and there were no deaths due to bleeding. Preoperative platelet count was not significantly associated with intraoperative or postoperative bleeding.
Appendix Table 9. Prophylactic Platelet Transfusion Versus No Prophylactic Platelet Transfusion for Surgery
In a meta-analysis of 6 RCTs and a single pilot study conducted during the licensure of aprotinin, adverse outcome data were compared between cardiac surgical patients who received (n = 284) or did not receive (n = 1436) perioperative platelet transfusions (Appendix Table 10) (53). Platelet transfusion was identified as an independent predictor of adverse outcomes, including mortality (OR, 4.76 [CI, 1.65 to 13.73]). It is possible that platelet transfusion served at least in part as a surrogate marker of sicker patients in this analysis, rather than as a direct cause of adverse outcomes (that is, confounding by indication).
Appendix Table 10. Platelet Transfusion Versus No Platelet Transfusion for Coronary Artery Bypass Graft Surgery
The consensus opinion of the panel is that platelet counts of 50 × 109 cells/L and greater are safe for major nonneuraxial surgery. There is no evidence of increased perioperative bleeding risk in thrombocytopenic patients with platelet counts greater than 50 × 109 cells/L. We recommend that platelet transfusion be withheld in nonbleeding surgical patients when the platelet count is greater than 50 × 109 cells/L and there is no evidence of coagulopathy. In contrast, we suggest that platelet transfusion should be considered in cardiac surgical patients with perioperative bleeding and thrombocytopenia (see the Definitions section) and/or suspected qualitative platelet abnormalities, which often result from exposure of platelets to the CPB circuit (54). Platelet transfusions are often administered to nonbleeding cardiac surgical patients (55). There are no data supporting this practice, and it should be discouraged.
Recommendation 6: The AABB cannot recommend for or against platelet transfusion for patients receiving antiplatelet therapy who have intracranial hemorrhage (traumatic or spontaneous).
Quality of evidence: very low; strength of recommendation: uncertain.
Five observational studies (n = 635) examined clinical outcomes among patients receiving antiplatelet agents who present with traumatic brain injury (Appendix Table 11) (56). One study reported a greater mortality rate for patients who received transfusions with platelets (relative risk, 2.4 [CI, 1.2 to 4.9]) (57), and a second study reported a lower mortality rate for patients receiving platelets (relative risk, 0.21 [CI, 0.05 to 0.95]) (58). Three studies showed no significant effect on mortality rates when patients received transfusions with platelets (59–61). One additional observational study (n = 88) reported that patients with traumatic brain injury and moderate thrombocytopenia (50 × 109 to 107 × 109 cells/L) who were transfused with platelets had poorer survival than those who were not transfused with platelets (62). In all of these studies, it was not possible to establish a causal relationship between platelet transfusion and clinical outcomes, and confounding by indication was possible.
Appendix Table 11. Platelet Transfusion Versus No Platelet Transfusion in Patients Who Were Thrombocytopenic and Had a Traumatic Brain Injury
In patients with intracerebral hemorrhage who are receiving antiplatelet agents, the decision to transfuse platelets requires an individual clinical decision based on various clinical factors, including the size of the bleeding and the patient's level of consciousness. For surgeries involving the central nervous system, platelets are conventionally transfused prophylactically for a preprocedure platelet count less than 80 × 109 to 100 × 109 cells/L, although only low-quality data supporting this threshold are available.
A large proportion of platelet transfusions are administered prophylactically to reduce the risk for spontaneous hemorrhage in patients receiving chemotherapy or HPCT (1–3). With data available from several RCTs (15–21, 24–29, 63), there is now a solid understanding of the role of platelet transfusions in this specific setting. Platelet prophylaxis, as compared with a therapeutic platelet transfusion strategy, reduces but does not eliminate the risk for bleeding in hospitalized patients with therapy-induced hypoproliferative thrombocytopenia. We recommend that these patients receive prophylactic platelet transfusions for a morning platelet count of 10 × 109 cells/L or less. Clinicians can be assured that prophylaxis with low-dose platelets provides hemostasis that is equal to standard- or high-dose platelets in patients with therapy-induced hypoproliferative thrombocytopenia. However, low-dose platelets must be transfused more often because they provide a lower platelet increment (18).
Only limited data are available to support transfusing platelets for indications other than prophylaxis against spontaneous bleeding in patients with therapy-induced hypoproliferative thrombocytopenia. Our panel took the position that it is appropriate for the AABB to address common and important clinical scenarios, such as the role of platelet transfusions in patients having invasive procedures, even as we await better data. Therefore, we decided to review observational data as a basis for platelet transfusion recommendations. The lower quality of data is reflected in the weak strength of recommendations outside of the hypoproliferative thrombocytopenia setting. In the specific case of CVC placement, our consensus opinion is that recent observational data (38, 41) support a platelet count transfusion threshold of 20 × 109 cells/L. This threshold seems to be reasonable even for the placement of large-bore catheters for apheresis in thrombocytopenic patients (12). Observational data were also used to inform the platelet transfusion recommendation for LP, for which we suggest a threshold platelet count of 50 × 109 cells/L. Most of the published data about the safety of performing diagnostic LP in the setting of thrombocytopenia comes from a single center's experience with pediatric patients (45); it is unclear how generalizable these data are to adult patients. Of 21 case reports of LP-associated spinal hematomas in adults, 17 (81%) occurred at a platelet count less than 50 × 109 cells/L. However, in all but 1 patient, other risk factors for bleeding were identified (50). We believe that clinical judgment should be used about the need for platelet transfusion in patients requiring LP with platelet counts in the range of 20 × 109 to 50 × 109 cells/L.
Our recommendation to provide prophylactic platelet transfusion at a platelet count of 10 × 109 cells/L or less for patients with therapy-induced hypoproliferative thrombocytopenia is consistent with the current standard of practice as reflected in other published transfusion guidelines (64–70). The recommendation of using a platelet count of 50 × 109 cells/L or greater as a safe level to perform LP in adults falls within the spectrum of other published guidelines, which have typically recommended platelet thresholds ranging from 20 × 109 to 50 × 109 cells/L (50, 65, 66). The recommendation of a 50 × 109–cells/L platelet transfusion threshold for major nonneuraxial procedures is also consistent with other guidelines (64–70). The suggestion to transfuse platelets to patients having CPB with perioperative bleeding and thrombocytopenia or suspected platelet dysfunction is concordant with the guideline from the Society of Thoracic Surgeons (71), which states, “It is reasonable to transfuse non-red cell hemostatic blood components based on clinical evidence of bleeding and preferably guided by specific point-of-care tests.” We consider coronary artery bypass graft to serve as a model for all surgeries requiring CPB. Our recommendation to use a platelet count threshold of 20 × 109 cells/L for CVC placement represents the most substantial break from other published guidelines (64–70, 72, 73). The 2012 Society of Interventional Radiology guideline, for example, recommends a minimum platelet count of 50 × 109 cells/L for CVC placement (73). We believe that existing observational data (38, 41) are sufficiently compelling to support using a lower platelet threshold. Adherence to this lower threshold should reduce transfusion risks while conserving resources.
Grade 2 bleeding remains very common among patients receiving marrow-suppressive therapy, even with routine platelet prophylaxis (18, 19, 21). Other means of preventing bleeding in this setting should be explored, such as using antifibrinolytic therapy. Serious or life-threatening bleeding (grade 3 or 4) is fortunately rare. When severe bleeding occurs in patients with therapy-induced hypoproliferative thrombocytopenia, it is often at a platelet count greater than the 10 × 109–cells/L threshold typically used for prophylaxis (25). Future studies should explore the role of platelet prophylaxis in patient subgroups that may have specific risk factors for bleeding.
Data addressing the question of a minimum safe platelet count for performing invasive procedures are limited and observational in nature. Randomized trials of prophylactic platelet transfusion for procedures would be valuable but would present logistic and ethical challenges. However, it would be straightforward to establish registries to document the outcomes of consecutive patients having specific procedures. We believe that this should be a high research priority.
Platelet count is the main laboratory measurement used to guide platelet transfusion; however, it provides no qualitative information about platelet hemostatic function. The clinical utility of in vitro platelet hemostasis testing, particularly at the point of care, remains a key area of exploration.
The ideal approach to platelet transfusion would be to administer sufficient platelets to optimize patient outcomes while avoiding unnecessary transfusions with their attendant risks and costs. The recommendations in this guideline reflect the AABB's current thinking on how platelet transfusions should be used in various clinical settings. These recommendations are not meant to be interpreted as strict standards but should provide a useful adjunct to providers' clinical judgment as individualized transfusion decisions are being made. We anticipate that these guidelines will be refined and improved over time, using new data from well-designed prospective trials.
The In the Clinic® slide sets are owned and copyrighted by the American College of Physicians (ACP). All text, graphics, trademarks, and other intellectual property incorporated into the slide sets remain the sole and exclusive property of the ACP. The slide sets may be used only by the person who downloads or purchases them and only for the purpose of presenting them during not-for-profit educational activities. Users may incorporate the entire slide set or selected individual slides into their own teaching presentations but may not alter the content of the slides in any way or remove the ACP copyright notice. Users may make print copies for use as hand-outs for the audience the user is personally addressing but may not otherwise reproduce or distribute the slides by any means or media, including but not limited to sending them as e-mail attachments, posting them on Internet or Intranet sites, publishing them in meeting proceedings, or making them available for sale or distribution in any unauthorized form, without the express written permission of the ACP. Unauthorized use of the In the Clinic slide sets will constitute copyright infringement.
James P. AuBuchon, Yanyun Wu
Puget Sound Blood Center
December 4, 2014
What is a "Low-Dose" Platelet Unit?
We commend the authors of these platelet guidelines for their careful evaluation of the literature and development of rational indications for platelet transfusion.We note that the guideline references the results of the Platelet-Dose (PLADO) Study showing that “low dose” platelets yield equivalent prophylactic protection against Grade 2 or greater hemorrhage in hospitalized patients with therapy-induced hypoproliferative thrombocytopenia. The guideline refers to this transfusion dose as a “half unit” of apheresis platelets, but simply dividing a unit into two halves may provide suboptimal dosing for many patients.The content of “low-dose” platelet units in the PLADO Study was defined as 1.1 x 1011/m^2. In the US, the average male body surface area (BSA) is 1.7 m2, and for females it is 1.4 m2.1 The FDA-required minimum is 3.0 x 10^11 platelets in an apheresis platelet unit; in the current era, when the majority of such units are collected as “multiple” units from a single donor, the content is usually found to be 3.5 – 4.0 x 10^11 platelets; in our center, the median is 3.7 x 10^11, and the average is 3.8 x 10^11 platelets.Half of an “average” unit would therefore provide the intended “low-dose” prophylactic platelet transfusion only if the patient’s BSA were less than 1.7m^2. This would correspond to a 5 foot 8 inch 135 pound male recipient (BMI: 22 kg/m^2), or a 5 foot 5 inch 145 pound female (BMI: 24.1 kg/m^2). For larger patients, and for the half of units below the median of 3.7 x 10^11 platelets in the full unit, the dose of platelets would be less than expected.Therefore, although the size of an adult platelet recipient is usually not taken into account when selecting units for transfusion, the size of the recipient and the actual content of the unit may be more important if lower doses are utilized. If a standard “low-dose” platelet were to be created, it might more safely be sized closer to two-thirds the average apheresis unit so as to avoid difficulties in unit selection and inadvertent (and unsuspected) under-dosing in larger patients. The effects of prophylactic platelet transfusion with doses below the “low-dose” definition used in the PLADO study have not been examined in clinical trials.James P. AuBuchon, MD, FCAP, FRCP(Edin)Yanyun Wu, MD, PhDPuget Sound Blood CenterSeattle, WashingtonReference1. McDowell MA, Fryar CD, Ogden CL, Flegal KM. Anthropomorphic reference data for children and adults: United States, 2003-2006. National Health Statistics Report. 10:1-45, October 22, 2008. DHHS Publication No. (PHS) 2009-1250. Centers for Disease Control and Prevention, Hyattsville, MD, 2008.
Richard Kaufman, MD
Brigham and Women's Hospital
March 18, 2015
In Reply: What is a "Low-Dose" Platelet Unit?
We wish to thank Drs. AuBuchon and Wu for raising the issue of the definition of a “low dose” platelet unit. We agree that for larger recipients, transfusing “half” apheresis platelet units may not provide a platelet dose equal to what was used in the low dose arm of the PLADO trial in every case. We think that this is an important point for blood banks to consider. The specific language chosen for the AABB clinical practice guideline on platelet transfusion was a simplification, and this was intentional. The primary target audience of our guideline was front-line clinicians who order platelets, rather than blood banks that prepare platelets.
The PLADO manuscript states:
“For a typical adult, the medium dose was equivalent to the standard dose currently used in clinical practice. The low dose was half the medium dose, and the high dose was twice the medium dose”(1).
This was also a simplification. But it concisely conveyed what “low dose” meant in the context of the PLADO trial: for a typical adult, “about half” of an apheresis unit, as opposed to e.g. “about 2%” of an apheresis unit. Similarly, we sought to express the concept of low dose platelets in terms that would be meaningful to prescribing clinicians. The alternative would have been to recommend doses of, for instance, 1.1 X 10^11 platelets/m^2 to providers who typically order platelets by the unit without considering recipient body surface area, and who may not be aware that the minimum required platelet content for an apheresis unit is 3 X 10^11 platelets (2).
Additionally, the acceptable range for a low dose platelet transfusion in the PLADO study was 1.1 X 10^11 platelets/m^2 ± 25% (1), a wider range than that cited by AuBuchon and Wu. A dose of 1.1 X 10^11 platelets per m^2 body surface area may not constitute an absolute lower limit of safety based on PLADO. Among 1223 acceptable “low dose” platelet transfusions given to adults in the PLADO study, 572 (46.8%) were in the range of 0.825 X 10^11 to 1.1 X 10^11 platelets/m^2 (Assmann SF, personal communication.) To date, hospitals have demonstrated relatively little interest in adopting low dose platelets. While a “3 doses from 2 apheresis units” strategy as suggested by AuBuchon and Wu might be a reasonable manufacturing approach, splitting apheresis units in half would be a simpler approach.
Results provided by:
Copyright © 2016 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use
This PDF is available to Subscribers Only