Delayed hemolytic transfusion reaction in children with sickle cell disease: first 5-year retrospective study in mainland France

Delayed hemolytic transfusion reaction (DHTR) is increasingly being reported as a serious complication of red blood cell (RBC) transfusion in patients with sickle cell disease (SCD).^1,2 Alloimmunization against RBC antigens has been identified as its leading underlying cause.^3,4

DHTR remains a major challenge as it may be under-recognized, leading to patient receiving inappropriate further transfusion that may result in life-threatening exacerbation of hemolysis.^1,5 Moreover there is no consensus on its optimal management. Immunoglobulins (IVIG) and erythropoïetin (EPO) are considered beneficial, and eculizumab has recently emerged as a possible alternative treatment. It has been recommended that RBC transfusions be withheld during DHTR as this might aggravate hemolysis, whereas the use of corticosteroids (CS) is still controversial because of its suspected vaso-occlusive triggering effect.^6-9 However, in the setting of acute severe DHTR, the potential risks of steroids associated with RBC transfusion must be balanced against their potential life-saving effects. For patients requiring further transfusion away from the DHTR acute episode, rituximab is recommended to prevent the production of additional alloantibodies.⁹ However, there is currently a lack of evidence to guide the best management of these “untransfusable” patients.

Our main objective was therefore to update the clinical description of DHTR, immuno-hematology findings, laboratory explorations and immediate management, at nationwide scale for children with SCD, over a 5-year period beginning in 2015, a period of improved awareness of DHTR. The secondary objective was to describe future patient management, when a subsequent RBC transfusion was required, later following the resolution of DHTR.

This study was based on a national survey performed through the French national SCD network in which all cases of DHTR in children (age <18 years) diagnosed with SCD, occurring between January 1, 2015 and December 31, 2020 and followed-up in mainland France were collected. Relevant patient data were retrospectively collected from medical files and anonymized. The data collected included clinical and laboratory data at the time of the triggering transfusion (TT) and during the DHTR episode (including evidence of the selective destruction of transfused cells over self-sickle RBC), management and clinical course. Transfusion and antibody history data from the French Blood Agency information system were also analyzed. Following national recommendations, all French centers deliver cross-matched, leukocyte-reduced, RBC units matched for ABO, RH and K. This study was performed in accordance with the Declaration of Helsinki.

DHTR was defined as the occurence at least 3 days after a transfusion, of signs indicating accelerated hemolysis together with a significant decrease in hemoglobin (Hb) (<30%), particularly HbA, indicating preferential hemolysis of the transfused RBC. Whenever possible, the previously published nomogram was applied to confirm the diagnosis.¹⁰ Forty-one DHTR were reported by all but one center of the SCD network. Four were excluded due to alternative diagnoses of splenic sequestration in one case, and autoimmune hemolytic anemia in the other three. For these four cases, sequential determinations of hemoglobin A indicated that the increased hemolysis did not selectively target transfused donor RBC. We, thus, considered, 37 DHTR episodes in total, at 18 pediatric centers in mainland France, in 37 children including 35 with HbSS, one with HbSβ^°, and one with HbSβ⁺ diseases. Median age at DHTR diagnosis was 9 years (range, 3-15). Twenty-two patients developed DHTR while on hydroxyurea therapy. For 34 children, the TT was delivered as an occasional transfusion episode, mostly during an inflammatory state (vaso-occlusive crisis [VOC], acute chest syndrom [ACS], infection, pre-operative setting). Three patients were on chronic transfusion programs, initiated 3 to 6 months before the DHTR episode.

Clinical symptoms and biological data collected during DHTR are detailed in Table 1. Pain mimicking VOC was the leading symptom followed by dark urine, indicating of hemoglobinuria. Half the patients developed fever during the DHTR episode. No death or stroke occurred in our series, but we observed posterior reversible leukoencephalopathy syndrome in one child, acute tubular necrosis in another, and ACS in 12 children. All complications occurred before further transfusion.

Post-TT determinations of Hb level and HbA% were monitored in only 20 and nine cases, respectively. During DHTR, Hb electrophoresis was performed for only 16 children, suggesting that the diagnosis of DHTR was initially overlooked for 21 of 37 children. Hb and HbA% were assessed both shortly after the TT and at the time of DHTR symptoms in only nine children. Using nomogram, there was a high risk of ongoing DHTR in these nine episodes.¹⁰

For other episodes, signs of accelerated hemolysis, including an inappropriate marked drop in HbA levels after the TT were used to diagnose DHTR. Hyperhemolysis (HH) with Hb levels falling below pretransfusion values, occurred in 32 of 37 children. Reticulocyte counts were significantly lower in the subgroup of 17 children receiving an additional transfusion, suggesting that profound erythropoiesis impairment might have contributed to the decision to transfuse during DHTR. None of the other clinical or biological parameters differed significantly between the subgroups receiving only supportive care including EPO (n=11), immunosuppression without transfusion (CS, IVIG, and/or rituximab) (n=9) or further transfusion (n=17). Interestingly, HbA levels were unavailable for the eight children who received further transfusions without immunosuppression, suggesting that DHTR was undiagnosed at the time of retransfusion.

Table 1.Clinical and laboratory data during delayed hemolytic transfusion reaction and by treatment.

Patient management during DHTR is detailed in Table 2. Seventeen patients received an additional transfusion. When no immunosuppressive agent was associated with transfusion, it clearly and consistently worsened hemolysis and resulted in HH episodes. The addition of CS at 1 mg/kg, prescribed for two children was ineffective. When higher CS dose (2 mg/kg) were prescribed for seven children in addition to transfusion, post-transfusion Hb levels were effectively maintained. Interestingly, three of these children had previously experienced a DHTR recurrence when transfusion was attempted with 1 mg/kg of CS for one or without any CS cover for the two others. Eleven patients received IVIG, among which eight did not require further transfusion. For one child, IVIG was the only immunosuppression prescribed, precluding any conclusions on its efficacy to mitigate DHTR.

We were able to obtain previous transfusion records and antibody testing results for all but six children, these remaining children having received transfusion outside France (Table 3). Consistent with adult data showing a history of DHTR and/or alloimmunization to be associated with a higher risk of DHTR, 16 children in our cohort had a history of alloimmunization and/or DHTR.^11-13 The number of RBC units transfused before the TT was low (median 4) but seven children developed DHTR after receiving ≥ 12 RBC units. Nine of the 21 patients with no immunization history, therefore receiving RH/K-matched units, developed antibodies. Sixteen patients were already immunized, and were transfused with extended (FY, JK, Ss) matched RBC units. Six of these patients, developed new antibodies, which were anti-M in three cases (M compatibility was not taken into account). For five patients, antibodies were detected only during later tests (3 weeks to 3 months after DHTR).

Eleven children received subsequent transfusions, a median of 3 months (range, 10 days-3 years) after resolution of the DHTR episode (Online Supplementary Figure S1). All three alloimmunized children received rituximab for non-specific prevention, and RBC units compatible in FY, JK and MNS systems, DHTR recurred in one child.

Table 2.Delayed hemolytic transfusion reaction management and outcome.

Table 3.Immunohematologic characteristics of the patients.

Among the eight children without alloimmunization history, four were managed with rituximab and extensively matched RBC, with an uneventful course for all children. Without rituximab, three of the other four non-alloimmunized children experienced DHTR recurrence.

This analysis of 37 children presenting a DHTR episode during a recent 5-year period in mainland France shows, that DHTR complications are regularly encountered in the pediatric SCD population. Some milder cases almost certainly passed unnoticed, but we can confirm that DHTR morbidity is high in children, as in adults, with SCD-related severe complications occurring in 12 of 37 children.^2,11-14 Whenever possible, we used the diagnostic nomogram developed for adults¹⁰ to assess the likelihood of DHTR. We recommend the systematic determination of Hb and HbA concentrations within 48 hours of every occasional transfusion, and repeatedly in case of any symptoms occurring in a context of a recent transfusion. Given the non-specific symptoms observed at DHTR presentation, this would facilitate the timely diagnosis of DHTR. Once DHTR has been recognized, the timing of antibody testing is also of key importance: in our study, late screening allowed to capture five among the 15 newly identified alloantibodies, highlighting the need for sequential testing during follow-up.

Further transfusion should be avoided during the acute episode, as it can lead to HH, as observed in eight children in our study with apparently overlooked DHTR. However, the SCD-related complications, as well as life-threatening anemia and intravascular hemolysis reactions that can arise during DHTR are not without risks.^2,11-14 Timely transfusion, before irreversible multiorgan failure or stroke has occurred, should be considered, to improve oxygen delivery, together with the use of corticosteroids to reduce inflammation.^8,9 The efficacy of more recent therapeutic interventions such as eculizumab or tocilizumab needs to be confirmed by additionnal reports.^{7,1 5} In the acute DHTR setting, as during future patient management in situations which subsequent RBC transfusion may be required, shared decision-making is crucial. We propose national multidisciplinary meetings, to facilitate close communication between SCD physicians and transfusion medicine specialists for the discussion of preventive strategies.¹⁶

Footnotes

• Received July 4, 2022
• Accepted October 26, 2022

Correspondence

C. FALGUIÈRE - claire.falguiere@chicreteil.fr

Disclosures

CoP reports honoraria and expert consultancy for Addmedica and Novartis. All other authors have no conflicts of interest to disclose.

Contributions

CF and CoP cared for some patients, designed and initiated the study, collected and interpreted the data, carried out the statistical analysis, wrote the paper, and had final responsibility for the decision to submit for publication; FP was involved in the collection of transfusion data, interpretation of results, and writing of the manuscript; SA followed up patients, was involved in the interpretation of results, and writing of the manuscript; BK, CA, AK, MB, CoG, MHO, SPCD, CeG, NG, AG, PM, VSB, LL, EM, AD, BP, CaP, JFB and ML cared for the patients and contributed to the writing of the paper. All authors reviewed the paper and approved the final manuscript.

Data-sharing statement

All data generated or analyzed during this study are included in this published article (and its Online Supplementary Appendix).