Frequency and timing of all-cause deaths in visits involving suspected transfusion reactions, and the significance of cardiopulmonary disturbances.

BACKGROUND/OBJECTIVES: Transfusion reactions (TRs) may cause or contribute to death. Cardiopulmonary TRs are distressing, and collectively account for most transfusion fatalities, though the degree to which they alter survival more broadly is unclear. Deaths (and their timing) after TRs may provide further insights. MATERIALS/METHODS: Adult (tri-hospital network) haemovigilance data (2013-2016) recorded referrals with conclusions ranging from unrelated to transfusion (UTR) to entities such as: septic TRs, serologic/haemolytic reactions, transfusion-associated circulatory overload (TACO), transfusion-associated dyspnoea (TAD), transfusion-related acute lung injury (TRALI), allergic transfusion reaction (ATR), and others. For (in- or out-patient) visits involving suspected TRs (VISTRs), all-cause mortalities (% [95% confidence interval]) and associated time-to-death (TTD) (median days, [interquartile range]) were compared. Diagnoses were defined inclusively (possible-to-definite) or strictly (probable-to-definite). RESULTS: Of 1144 events, rank order VISTR mortality following (possible-to-definite) TRs, and associated TTDs, were led by: DHTR 33% [6-19], 1 death at 123d; TRALI 32% [15-54], 6 deaths: 3d [2-20]; BaCon 21% [14-31], 17 deaths: 10d [3-28]; TACO 18% [12-26], 23 deaths: 16d [6-28]; TAD 17% [11-26]: 18 deaths, 6d [3-12]. Higher-certainty TRs ranked similarly (DHTR 50% [9-91]; BaCon 29% [12-55], 4 deaths: 12d [3-22]; and TACO 25% [16-38], 15 deaths: 21d [6-28]). VISTR mortality after TACO or TRALI significantly exceeded ATR (3·3% [2·4-5·8], P < 0·00001) but was not different from UTR events (P = 0·3). CONCLUSIONS: Only half of cardiopulmonary TRs constituted high certainty diagnoses. Nevertheless, cardiopulmonary TRs and suspected BaCon marked higher VISTR mortality with shorter TTDs. Short (<1 week) TTDs in TAD, BaCon or TRALI imply either contributing roles in death, treatment refractoriness and/or applicable TR susceptibilities in the dying.

Serologic assessments in acute transfusion reactions: practices and yields.

BACKGROUND & OBJECTIVES: Serologic testing after transfusion reactions (TRs) aims to find accountable immune haemolytic incompatibility. Our hospital policies recommend serologic testing in all TR, except for low-risk fevers (subclinical temperature <39°C) or uncomplicated allergic reactions. Assessing compliance with these guidelines and serologic testing yields may provide insights on quality of practice and value. MATERIALS & METHODS: Interrogation of two haemovigilance databases identified all possible-to-definite TR over a 4-year period (2013-2016) at four academic hospitals. We reviewed the performance and outcome of serologic testing by site, year, reaction type, implicated product and service. RESULTS: Serologic testing occurred in 769 (55%) of 1408 referrals, with 1153 (82%) compliant with guidelines. Similar proportions deviated to overtesting (85/550 [15%]) and undertesting (174/858 [20%]), with undertesting seen most often in atypical TR. Overall, 30 (4.4%) of 769 cases had a new finding, but only 2 (0.3%) reflected host-derived antibodies. Overall, the number needed to test to discover an unexpected allospecificity was 385, or 253 if limited to high-risk fevers. Reaction- and product-specific yields ranged from 0% to 48%. The yield in complicated allergic reactions was low at 2%, constituting only predictable passive isohaemagglutinin(s) in retrospect. Investigated IVIG TR accounted for most of this cohort's signal by passive isohaemagglutinins in 48%. CONCLUSION: The performance of post-TR serologic testing revealed practice gaps and expected context-specific yields. Tailored serologic testing (i.e. indirect antiglobulin tests for alloantibodies in post-RBC/high-risk febrile reactions, ± isoagglutinin-focused tests after IVIG or ABO-minor-mismatched platelets) may improve value and liberate resources for other unmet needs in TR investigation.

Feeling the burn: the significant burden of febrile nonhemolytic transfusion reactions.

BACKGROUND: Febrile nonhemolytic transfusion reactions (FNHTRs) are characterized by a post-transfusion temperature rise (of ≥ 1°C, to ≥ 38°C) or chills/rigors unrelated to the underlying condition. FNHTRs are provoked by inflammatory cytokines in the product or by host antileukocyte antibodies against residual donor leukocytes. FNHTRs are among the most commonly reported transfusion disturbances and are generally deemed nonserious events. However, their impact on patients and hospitals may be underestimated. STUDY DESIGN AND METHODS: A search through two hemovigilance databases identified all known possible-to-definite FNHTRs over 3 years (2013-2015) at four academic hospitals using prestorage leukoreduced components. FNHTRs were assessed for frequency by product (red blood cells [RBCs], platelets [PLTs], intravenous immunoglobulin), diagnostics (bedside, chest imaging, serology, microbiology), and management (medications, disposition change). The definition of FNHTR was derived from Canada's Transfusion-Transmitted Injuries Surveillance System. RESULTS: For 437 FNHTRs, the overall per-product rate across all sites was 0.24%, or 0.17% with RBCs alone and 0.25% with PLTs alone. One-third of patients had significant fevers (≥ 39.0°C or a rise by ≥ 2.0°C). Approximately one-quarter underwent chest imaging within 48 hours, and 79% had blood cultures. A hospital admission directly attributable to the FNHTR, to exclude other causes of fever, occurred in 15% of FNHTR outpatients. CONCLUSION: An analysis of FNHTRs reveals a substantial burden of postreaction clinical activity in addition to the disturbance itself. Efforts to avoid this adverse event may save resources, reduce patient distress, and encourage compliance with more restrictive transfusion strategies.

Pilot study of ER utilization at Tulsa hospitals.

CONTEXT: The Tulsa Community Access Program (TCAP) project has been developed to address the health care needs of the uninsured and underinsured population in the Tulsa MSA. One of the objectives is the development of a 24/7 Care Center. OBJECTIVE: To complete a pilot study on patient demographics and utilization patterns in the Emergency Room (ER) to determine if a 24/7 clinic is needed or would be used as an alternative site for treatment. DESIGN: A survey was randomly administered to ER clients from March 2003 to April 2003 and included 45 open-ended, closed-ended, and Likert scale questions. Acuity level of care was verified by a hospital professional. SETTING: Hillcrest Medical Center, Tulsa, provides a significant amount of indigent care; four time periods were picked for peak and varied ER usage. PARTICIPANTS: 200 ER clients, 50 per time period; survey administration and data analysis were done by University of Oklahoma-Tulsa graduate students. MAIN OUTCOME MEASURE(S): Prior assumptions: uninsured/underinsured use the ER for primary care; demographics of the ER client are low income/unemployed, low education; majority of ER clients are non-urgent/non-emergency and would use a 24/7 clinic. RESULTS: 56% have used ER 1-2 times in past year; 89% of cases were non-urgent/non-emergency; 46% graduated high school; 48% work full-time; 42% have family income 15,000 dollars or less; 76.5% favor an after hours primary care clinic. CONCLUSIONS: This survey is an important tool to determine the needs and patterns of ER utilization; there is a need for a 24/7 clinic and it would be used by the target population.