A case of hyperhemolysis syndrome in sickle cell disease and concomitant COVID-19

Abstract

Background

Hyperhemolysis syndrome (HHS) is an uncommon transfusion reaction described in several hematologic disorders, including sickle cell disease (SCD). HHS is characterized by a decline in hemoglobin (Hb) values below pre-transfusion levels following transfusion of red blood cells (RBCs), coupled with laboratory markers consistent with hemolysis. The proposed pathophysiologic mechanisms underlying HHS include increased phosphatidylserine expression, macrophage activation, and complement dysregulation. Many pathophysiologic mechanisms thought to contribute to HHS have been similarly described in cases of severe COVID-19.

Case report

A 28-year-old male with a history of HbSS presented with shortness of breath, right-sided chest pain, and a two-day history of fever. Polymerase chain reaction (PCR) detected SARS-CoV-2 infection with the omicron variant. The patient required an RBC transfusion (pre-transfusion hemoglobin [Hb]5.8 g/dL) with an immediate post-transfusion Hb of 6.3 g/dL. However, Hb rapidly declined to 1.7 g/dL, and lactate dehydrogenase (LDH) rose to 8701 u/L. The absolute reticulocyte count of 538 × 10⁹/L correspondingly fell to 29 × 10⁹/L. Despite additional RBC transfusions and initiation of immunosuppressive therapy, he expired on Day 9(D9).

Conclusion

Given the similarities in their proposed pathophysiology, patients with SCD and concomitant SARS-CoV-2 infection may be predisposed to developing HHS.

1. Introduction

HHS is a life-threatening complication associated with RBC transfusions that has been predominantly described in SCD [1], [2]. This condition is characterized by post-transfusion hemolysis and Hb levels below pre-transfusion levels, suggesting the destruction of both transfused and autologous RBCs [1]. In contrast to many other hemolytic transfusion reactions, cases of HHS rarely demonstrate alloantibodies, and direct antiglobulin testing is often negative [1].

Although the underlying pathophysiology of HHS remains uncertain, many proposed disease processes have also been similarly characterized in patients with SARS-CoV-2 infections. These include the externalization of phosphatidylserine (PS) on the surface of RBCs, resulting in macrophage erythrophagocytosis while also stimulating complement-mediated intravascular and extravascular hemolysis [1], [2], [3]. Abnormal activation of macrophages is similarly observed in SCD-associated HHS and SARS-CoV-2 infections. These activated macrophages drive maladaptive inflammation and coagulopathy in patients with severe SARS-CoV-2 while acting to significantly reduce the lifespan of RBCs in HHS [1], [4]. Additionally, dysregulation of the complement cascade is equally observed in SCD-associated HHS and SARS-CoV-2 infections, a finding associated with the degree of hemolysis and inflammation observed in both [2], [5], [6], [7].

The similarities driving each disease process suggest that patients with SCD and concomitant SARS-CoV-2 infection may be susceptible to developing HHS. Currently, limited data exist describing the association between COVID-19 and HHS in the context of SCD. [8], [9], [10] Herein, we present a case of a patient with SCD who developed HHS in the setting of COVID-19.

2. Case report

Laboratory values and clinical management are summarized in Table 1.

Table 1.

Laboratory Parameters.

Laboratory Parameters (Reference Range)	Day 1	Day 2	Day 3	Day 4	Day 5	Day 6a	Day 6b	Day 6c	Day 7d	Day 8
Hemoglobin (12.4–17.3 g/dL)	11.1	10.4	10.1	10	9.5	5.8	6.3	5.4	3.1	1.7
Hematocrit (37.0–50.0%)	29.7	28.6	27.9	26.7	25.8	16.6	17.7	14.7	8.3	5.2
Platelet (150–450 ×10^9/L)	289	247	259	270	149	68	42	36	62	9
Reticulocyte Count, Absolute (10–150 ×10^9/L)	616.8	388.8	416.8	538.3	333.8	-	-	-	-	29
Reticulocyte Count (0.2–2.5%)	20.6	12.5	13.8	18.3	10.7	-	-	-	-	4.9
Lactate Dehydrogenase (125–220 u/L)	348	292	377	381	2389	-	-	8701	5811	16,441
ALT (10–50 u/L)	64	61	81	110	-	-	-	712	742	-
AST (10–50 u/L)	95	76	107	123	-	-	-	1876	2533	7703
Total Bilirubin (0.2–1.3 mmol/L)	7.6	3.1	3.2	3.1	-	-	-	9	7.4	8.2
Direct Bilirubin (0.0–0.5 mmol/L)	2	-	-	-	-	-	-	5.7	4.2	4.8
Lactate (0.5–2.2 mmol/L)	1.9	-	-	-	-	-	-	-	9.6	28.2
Ferritin (22–275 ng/mL)	529	586	693	737	12,592	-	-	-	-	-
Albumin (3.5–5.2 g/dL)	4.1	3.6	3.5	3.6	3.9	-	-	2.3	2.9	2.5
PT (9.5–12.8 s)	11.9	-	11.2	-	13.5	19.8	-	24.7	24.3	-
PTT (23.0–32.5 s)	-	-	-	-	30.8	68.6	-	-	-	-
C-Reactive Protein ≤ 5 mg/L)	28.3	56.9	17.2	19	64.1	-	-	-	-	-
D-Dimer (≤0.59 mg/L FEU)	1.31	4.07	2.75	1.48	17.76	-	-	-	-	-

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; PT, prothrombin time; PTT, partial thromboplastin time; FEU, fibrinogen equivalent units.

^aDenotes Pre-transfusion sample.

^bDenotes Post-transfusion sample.

^cDenotes Pre-Therapeutic Plasma Exchange sample.

^dDenotes Post-Therapeutic Plasma Exchange sample.

A 28-year-old male with HbSS presented to the emergency department (Day 1) with shortness of breath, chest pain, and intermittent fevers for two days. He was unvaccinated for SARS-CoV-2 with a known exposure three days earlier. He was receiving hydroxyurea for his SCD (15 mg/kg/day). His history of SCD complications included episodes of acute chest syndrome 15 years earlier. Lifetime transfusion history consisted of one RBC unit without prior alloantibody or autoantibody formation and no reported history of transfusion reactions or episodes of HHS.

On presentation, he was hypoxic (oxygen saturation 88%), requiring 2 L supplemental oxygen by nasal cannula resulting in an increase in oxygen saturation to 95%. He was otherwise afebrile with normal hemodynamic status. Laboratory results revealed baseline renal function (creatinine 0.41 mg/dL, BUN 5 mg/dL) and mild transaminitis (ALT 28 u/L, AST 95 u/L) above his baseline. Lactate dehydrogenase 348 u/L and total bilirubin 7.6 mmol/L were abnormal but corresponded with prior vaso-occlusive crises (VOC). Hemoglobin electrophoresis showed: HbS 84.9%; HbF 11.8%. Reverse transcription-PCR was positive for SARS-CoV-2 with multiplex fragment analysis identifying the omicron variant (B.1.1.529). Computed tomography(CT) angiography revealed subpleural nodular consolidations consistent with multifocal pneumonia and/or pulmonary infarcts. Intravenous antibiotics were initiated for empiric coverage of community-acquired pneumonia, and remdesivir and dexamethasone were administered for COVID-19. Intravenous fluids and pain medications were administered to treat his acute VOC-associated pain.

Despite ongoing complaints of chest and back pain, his respiratory and hematologic status remained stable until Day 4, when he became febrile (38.2 °C) and tachycardic (107bpm), with a decline in Hb (10.5–9.5 g/dL) and worsening transaminitis (ALT 207 u/L and AST 467 u/L), prompting discontinuation of remdesivir due to concerns for drug-induced acute liver injury.

On Day 6, he developed cardiogenic and septic shock requiring intubation, vasopressor support, and transfer to the intensive care unit. Laboratory studies revealed: Hb decline (9.5 g/dL to 5.8 g/dL) without evidence of bleeding; platelet count 42 × 10⁹/L; LDH 2389 u/L; absolute reticulocyte count decline (333.8 ×10⁹/L to 18 ×10⁹/L); and lactate 17.7 mmol/L. Chest x-ray showed new bilateral opacities. One unit of Rh- and Kell-matched RBCs were transfused with Hb rise from 5.8 g/dL to 6.3 g/dL. Therapeutic plasma exchange (TPE) was initiated due to evidence of multi-organ dysfunction. He underwent a single TPE (1.2-volume, 8:1 whole blood to citrate ratio, with 5% albumin and 800 mL of plasma replacement fluid). A one-unit RBC prime was used given the low pre-procedure hematocrit (Hct) and concern for a transient intra-procedure drop in Hct. Laboratory evaluation 4.5 h following completion of RBC transfusion showed a Hb 5.4 g/dL (0.9 g/dL decline). Notably, the plasma removed by TPE was dark red, suggesting significant RBC hemolysis ( Fig. 1).

Fig. 1.

Evidence of hemolysis in waste plasma from therapeutic plasma exchange for a patient presenting with SCD and COVID-19 infection.

Given the drop in Hb and visual evidence of hemolysis, 1 g methylprednisolone and 0.4 g/kg intravenous immunoglobulin (IVIG) were administered due to concern for HHS. Subsequently, a femoral hematoma at the site of the apheresis catheter was noted, prompting the transfusion of one unit of apheresis platelets due to concern for coagulopathy. CT angiogram ruled out extravasation. Despite ongoing pressor support, the patient became severely hypotensive. This coincided with a decline in renal function with metabolic acidosis and hyperkalemia requiring dialysis. A continued decline in Hb (Day 7, 3.1 g/dL) prompted an increase in IVIG (1 g/kg) and initiation of eculizumab (900 mg) and darbepoetin alfa (200 mcg).

Despite these measures, subsequent laboratories on Day 8 were suggestive of ongoing hemolysis with a Hb nadir of 1.7 g/dL, lactate rise to 34.1 mmol/L, and decline in pH to 6.98, prompting the transfusion of two RBC units. Despite an improvement in his metabolic acidosis, he developed progressive hypotension and cardiac arrest and expired on Day 9.

3. Discussion

We present a case of SCD-associated HHS with concomitant SARS-CoV-2 infection. COVID-19 poses a significant risk to patients with SCD, given this patient population’s increased risk for infections, including COVID-19. [11], [12] Once patients with SCD are infected with SARS-CoV-2, they demonstrate higher rates of morbidity and mortality relative to other populations, a finding that may be partially explained by decreased access to care, vaccine hesitancy, and immune dysfunction. [13], [14] Moreover, the risks posed by COVID-19 to patients with SCD will likely persist as evidence suggests viral endemicity with the likelihood that future strains will demonstrate increased transmissibility and the capability to evade immunity. [15] Therefore, despite prior infection(s) or vaccination, it is likely that patients with SCD will remain susceptible to future SARS-CoV-2 infections and, correspondingly, infection-associated complications.

Several common pathophysiologic features may amplify the risk for HHS in patients with SCD and concurrent severe SARS-CoV-2. First, increased expression of PS on cell surfaces has been observed in SARS-CoV-2 infections, resulting from a virally-induced influx of intracellular calcium, stimulating the activation of scramblase with subsequent RBC externalization of PS. [3] RBC externalization of PS is also observed in SCD-associated HHS, driven by inflammation and oxidative injury stemming from byproducts of chronic hemolysis [16], [17]. PS, a marker of senescent RBCs, triggers complement binding to RBC surfaces. [18] Ultimately, this RBC-bound complement culminates in intravascular hemolysis or recognition by cognate macrophage receptors and erythrophagocytosis (extravascular hemolysis) [3], [16], [17], [18]. The chronically increased expression of PS in SCD-associated HHS results in a significantly shortened RBC lifespan [2]. Consequently, SCD plus concomitant COVID-19, stimulating PS externalization through different mechanisms, may be additive or synergistic in evoking the onset of HHS.

Macrophage hyperactivation has been characterized in SARS-CoV-2 infections and SCD-associated HHS. An excess of cytokines (i.e., cytokine storm) has been observed in severe SARS-CoV-2 infections and patients with HHS, with accompanying laboratory findings mirroring those seen in macrophage-activation syndrome (MAS), including elevated plasma concentrations of cytokines, ferritin, C-reactive protein, and D-dimer [10], [19]. The significant role of macrophage hyperactivation in severe SARS-CoV-2 infections is supported by clinical trials demonstrating significant improvements in morbidity and mortality in patients receiving tocilizumab, a monoclonal antibody that inhibits IL-6 interactions with IL-6 receptors, ultimately preventing the activation of macrophages [20], [21]. Correspondingly, case studies have described the successful use of tocilizumab in managing patients with SCD-associated HHS [10], [19]. Similar to findings reported in these cases, our patient was febrile at presentation with comparable laboratory findings (cytopenias, abnormal liver function testing, elevated ferritin) before the onset of HHS. Moreover, our patient’s subsequent clinical course, including coagulopathy, hypoalbuminemia, neurological dysfunction, progressively worsening cytopenia, continued decline in liver function, and ferritin, correspond with those observed in MAS [22]. The role of macrophage activation in SCD-associated HHS and COVID-19 in evoking HHS may also explain the lack of alloantibodies observed in our case.

Infections such as COVID-19 may lead to direct activation and dysregulation of the complement cascade with complement deposition on RBC surfaces and, ultimately, complement-mediated RBC destruction (bystander hemolysis) [7]. Infections may promote complement dysregulation through either stimulating autoantibody formation directly or secondary to molecular mimicry (antibody-dependent mechanisms) or through antibody-independent means with SARS-CoV-2 mechanisms including spike protein competitively inhibiting factor H, indirect activation of the alternative pathway, and direct activation of the direct pathway. [5], [6]. A recent case-control analysis (41 distinct patients, 54 episodes) showed that 22 of the 54 episodes (40.7%) of HHS were associated with an infection at the time of presentation [5]. Moreover, a recent case demonstrated the formation of multiple auto- and alloantibodies following transfusion in the setting of HHS following COVID-19, supporting the concept of molecular mimicry [10]. Contrastingly, our patient did not demonstrate appreciable serologic findings with no detectable RBC antibodies, suggesting a possible SARS-related antibody-independent complement dysfunction.

Collectively, increased PS expression, hyperactivation of macrophages, and complement dysregulation may place patients with SCD and concurrent COVID-19 at a higher likelihood of developing HHS.

4. Conclusion

HHS is a rare but potentially fatal complication that can develop in patients with SCD, which may be triggered or exacerbated by COVID-19. Further studies and reports of observations in patients with SCD are needed to understand this potential association better.

Author contributions

AG, HJ, CBW, and SGY collected the clinical and laboratory data; AG, HJ, and SGY analyzed the data; AG, HJ, and SGY performed the literature review and wrote the manuscript; AG, HJ, AN, RS, LMS, IFI, BDA, and SGY critically reviewed the manuscript.

CRediT authorship contribution statement

Allen Green: Investigation, Formal analysis, Writing – original draft, Writing – review & editing, Visualization. Heather Jones: Investigation, Formal analysis, Writing – review & editing, Visualization. Alecia Nero: Writing – review & editing. Ibrahim F. Ibrahim: Writing – review & editing. Ravi Sarode: Writing – review & editing. Lisa M. Scheid: Writing – review & editing. Christopher B. Webb: Investigation, Writing – review & editing. Brian D. Adkins: Writing – review & editing. Sean G. Yates: Conceptualization, Investigation, Formal analysis, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration.

Conflict of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.