Hot and bothered: management and outcomes for patients with febrile nonhemolytic transfusion reactions

We've all been there—as clinical staff called to see a transfused patient with a fever, trying to work out whether it's serious, and whether it's related to the transfusion at all; as laboratory staff receiving a report and samples on a febrile transfused patient working through the process of eliminating serious transfusion-related causes; or after the event as part of the team trying to work out what happened and wondering (again!) why the available notes are so unhelpful. Patients do not always read the textbooks, so trying to assign a diagnosis for a particular constellation of signs and symptoms that may not exactly match any of the published definitions of reactions can feel like trying to fit a square peg into a round hole. Although superficially simple and apparently without serious consequence, febrile nonhemolytic transfusion reactions (FNHTRs) present their own challenges—not least because they are diagnosed by excluding other causes to explain a patient's fever in the context of a transfusion. The International Society of Blood Transfusion and International Haemovigilance Network (IHN) define FNHTR with the occurrence of fever (≥38°C oral or equivalent and/or a change of ≥1°C from pretransfusion value) and/or chills/rigors, which may be accompanied by headache and nausea, occurring during or within 4 hours after transfusion, without any other cause such as hemolytic transfusion reaction, bacterial contamination, or underlying condition.1 Cases may be further classified as serious if fever is 39°C or more and a change of 2°C or more from pretransfusion value. FNHTR can also be diagnosed in the absence of fever (solely on the presence of chills or rigors). No definitive diagnostic test is available—which can make case assignment, and assessment of imputability, incidence, and outcomes, difficult. Current thinking is that FNHTRs are caused by accumulation of proinflammatory white blood cell (WBC)- and platelet (PLT)-derived cytokines during storage and/or by interactions between WBCs and WBC antibodies present in donors and patients.2, 3 Donor, product, and patient factors likely all contribute, and studies have reported the effectiveness of prestorage leukoreduction and other measures in reducing the incidence of FNHTR.2-6 However, although febrile events in the context of transfusion are probably less common than in the past, for the reasons above, in day-to-day practice they are still a frequent reason for transfusions to be interrupted for urgent clinical assessment and investigations, and for the use of medication to treat or attempt to prevent further events.2, 3, 7 FNHTRs also still account for a substantial proportion of reactions reported to hemovigilance programs, including 32% of the cases in IHN's ISTARE database and 10% of the serious reactions.8-11 This observation is notable since many febrile reactions are either not recognized or recorded (especially likely in the outpatient transfusion setting), judged as not serious, not reported to hospital transfusion services, and/or not reportable to hemovigilance programs and therefore they are likely to be underreported. Despite their frequency, and the obvious potential for impact on patient well-being and hospital resources, relatively few data are available on the clinical and other downstream consequences of these reactions. In this issue of TRANSFUSION, Cohen and colleagues12 examined FNHTRs and their impact on individual patients and institutions at four academic hospitals in Canada. The participating institutions provide a wide range of services, reflecting the diversity of clinical scenarios in which transfusion may be required and the range of clinical staff from different disciplines managing the patients and the investigations. All blood components were leukoreduced. The authors identified 437 reported FNHTRs in 407 patients during the study period, for an overall incidence of one FNHTR in every 409 products transfused, slightly higher for PLTs (0.25%) than red blood cells (0.17%). One-third of events were classified as severe. Many patients had other adverse features in addition to fever and chills or rigors. Twelve percent had dyspnea, while wheeze, nausea and/or vomiting, and pain and other symptoms or signs were not uncommon. Several patients had urticaria or rash and lip or tongue swelling. This admixture of signs and symptoms makes the exclusion of other transfusion-related causes, and the immediate management of the patient, more difficult. If we add in the complexities of assessing new or recurrent fever in a patient with profound neutropenia with positive blood cultures, or a lymphoma patient with disease-related fevers, it can be exceedingly difficult at the bedside to work out which signs may be due to a transfusion reaction and which are due to other causes. These variable presentations also help to explain the wide range of additional medications the patients received, from acetaminophen for fever and meperidine for chills, to corticosteroids, diuretics, antihistamines, and new antibiotics. The fact that 27% of the patients had severe neutropenia at the time of transfusion, and nine febrile reactions met criteria for possible or probable bacterial contamination during the study period, underscores the need for vigilance and early intervention in high-risk cases. Just over half the reactions had their onset during the transfusion, and nearly all of these were interrupted for investigation but only 15% were then restarted, and only 58% of transfusions were completed in full. More than one-third of patients were outpatients at the time of transfusion, and 15% of patients required admission, for a median stay of 2 days, or transfer of inpatient care to a higher-acuity setting. Multiple additional investigations were required to rule out other causes and one in three received empiric antibiotics. The results paint an impressive picture of the disruption to planned care and patients’ (and their families’) lives, as well as the exposures to additional investigations and medications, which carry their own small but real risks, as well as costs. The authors also attempt to quantify the costs associated with FNHTRs, in particular for diagnostics. Of necessity, in the setting of incomplete information, they have been conservative in their estimates. What is exceedingly difficult to capture (and expensive) is the cost of nursing and junior medical staff time for initial assessment, with assistance from more senior staff especially if the diagnosis is unclear, or the patient is very unwell or “high risk” and needs emergency department evaluation and/or admission. It was not possible to accurately estimate the total of these “non-test diagnostic” costs, although it is clear that they are significant. The interruption to planned activities and the time required to urgently assess and investigate patients with fever during or after transfusion is also very disruptive for staff. The study has limitations, in that it was retrospective, conducted only at four sites in Canada, with no diagnostic test available to confirm clinical diagnoses made by a number of different clinical evaluators, and some information was not available. However, it is very helpful as far as it goes, and the results highlight the need for additional research into a problem that continues to occur and which not surprisingly persists in the era of routine prestorage leukoreduction in many countries. The authors identify the limited availability of high-quality studies of prevention strategies and also suggest the potential role of additional manufacturing steps, such as pathogen inactivation, to assist in reducing FNHTR occurrence. Nevertheless, prevention is still key—once a transfused patient develops a fever, then life-threatening conditions such as bacterial infection or hemolytic transfusion reactions will be important differential diagnoses, precipitating a sequence of urgent clinical and laboratory assessments. Finding the balance to protect patients from uncommon but potentially lethal complications while at the same time avoiding transfusion interruptions, unnecessary tests, antibiotic administration, and admissions is not easy. Clinical experience can help, but FNHTR remains a diagnosis of exclusion. Therefore, the best way to minimize these risks and costs (and many others) is to avoid unnecessary transfusions. Cohen and colleagues provide additional useful evidence to support the fundamental principles of patient blood management and remind us of the importance of carefully considering the risk–benefit equation for each and every transfusion for each and every patient and of capturing meaningful clinical transfusion outcomes data to inform our practice. The authors have disclosed no conflicts of interest. Erica M. Wood, MBBS, FRACP, FRCPA1,2 e-mail: erica.wood@monash.edu Lucy C. Fox, BCom/BSci, MBBS1,3 1Transfusion Research Unit School of Public Health and Preventive Medicine Monash University 2Department of Clinical Haematology Monash Health 3Department of Clinical Haematology Epworth Health and Victorian Comprehensive Cancer Centre Melbourne, Australia

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.