Warm Autoimmune Hemolytic Anemia Presenting 21 Years After Liver Transplantation: A Case Report

Introduction

Autoimmune hemolytic anemia (AIHA) is caused by autoantibodies that bind to self-red blood cells, leading to premature erythrocyte destruction and anemia. In warm autoimmune hemolytic anemia (WAIHA), IgG antibodies bind optimally at or near body temperature and mediate extravascular hemolysis predominantly through splenic sequestration, whereas in cold agglutinin disease (CAD), IgM antibodies bind to red blood cells at lower temperatures and activate the complement pathway, resulting in both intravascular and extravascular hemolysis [1]. AIHA occurring after solid organ transplantation is considered uncommon, and the true incidence remains uncertain [2]. Both warm and cold autoimmune hemolytic anemia have been reported in the post-transplant setting. Cold agglutinin syndrome has been described in pediatric liver transplant recipients receiving tacrolimus-based immunosuppression [3], and in a case series of pediatric intestinal transplant recipients, cold agglutinin syndrome was the most frequently observed subtype, followed by warm AIHA and mixed forms [4]. Mixed presentations with concomitant warm and cold autoantibodies have also been reported [5]. Warm AIHA remains the most commonly reported subtype overall, and cases occurring many years after transplantation are rare [2,5].

One well-recognized mechanism of early post-transplant hemolytic anemia is passenger lymphocyte syndrome, in which donor-derived lymphocytes produce antibodies directed against recipient red blood cells [6]. In contrast, late-onset hemolytic anemia after solid organ transplantation can occur months to years after transplantation [2,5]. Reported late presentations have been attributed to infections, post-transplant lymphoproliferative disorders, calcineurin inhibitor-associated hemolysis, or unclear/idiopathic causes [1,2,5].

Here, we present a case of warm autoimmune hemolytic anemia occurring 21 years after liver transplantation in a patient receiving long-term immunosuppressive therapy with tacrolimus.

Case Report

A 30-year-old woman with a history of orthotopic liver transplantation at age 9 for biliary atresia complicated by ruptured esophageal varices (21 years prior), chronic kidney disease stage IIIa, and long-term immunosuppression with tacrolimus presented to the emergency department with several days of exertional dyspnea. The patient reported a 2-to-3-week history of progressively worsening shortness of breath, which she initially noticed with walking and mild exertion. Over this period, her dyspnea gradually increased in severity, eventually prompting her to seek evaluation in the emergency department. She also reported intermittent chest tightness, palpitations, headache, and burning retrosternal discomfort associated with reduced appetite. She denied cough, sore throat, sick contacts, fever, chills, abdominal pain, nausea, vomiting, diarrhea, or constipation.

She denied overt bleeding, including melena, hematochezia, hematuria, epistaxis, or easy bruising, and reported no excessive menstrual bleeding. She followed annually with hepatology and nephrology. Her home medications included tacrolimus 5 mg daily, allopurinol 100 mg daily, and bisoprolol 5 mg daily. She denied NSAID use, tobacco use, alcohol consumption, or illicit substance use. Physical examination was unremarkable. The abdomen was soft and nontender, with no palpable splenomegaly.

On presentation, she was hemodynamically stable with mild tachycardia. Electrocardiography demonstrated sinus tachycardia with non-specific T-wave inversions in the anterior leads. Initial laboratory testing revealed severe anemia (hemoglobin 7.2 g/dL) and her baseline hemoglobin was 12.8 g/dL about 2 months prior. Troponin was normal, D-dimer was negative (D-dimer 0.39 μg/mL), and a chest radiograph was unremarkable.

Given the degree of anemia and lack of overt bleeding, gastrointestinal blood loss and hemolysis were considered. Repeat laboratory testing later that evening showed worsening anemia (hemoglobin 5.9 g/dL; hematocrit 16%) with a normal platelet count (229×109/L). She received 1 unit of packed red blood cells, and gastroenterology and hematology were consulted.

Iron studies demonstrated ferritin 410 ng/mL, serum iron 211 μg/dL, transferrin 165 mg/dL, and total iron-binding capacity 231 μg/dL. The hemolysis work-up showed undetectable haptoglobin (<10 mg/dL) with reticulocytosis (7.4%), lactate dehydrogenase of 281 IU/L, fibrinogen 231 mg/dL, and leukocytosis (white blood cell count 22.3×109/L). Total bilirubin was 3.2 mg/dL and direct bilirubin was 1.1 mg/dL. Serum tacrolimus level was 4 ng/mL. Complete laboratory findings with reference ranges are presented in Table 1. Review of the peripheral blood smear did not reveal findings consistent with substantial hemolysis. No abnormal red blood cell or white blood cell morphologies were identified, although mild microcytosis and an increase in monocytes were noted. Peripheral blood parasite smear was negative. Viral testing for COVID-19, influenza, and RSV was negative, and TSH was within normal limits. Computed tomography of the abdomen and pelvis showed no evidence of intra-abdominal hemorrhage; incidental findings included hyperdense cysts within the left kidney, an 8-mm cystic lesion in the pancreatic head, pneumobilia consistent with prior liver transplantation, and mild splenomegaly (diameter of 11.7 cm) (Figure 1).

A direct antiglobulin (Coombs) test was positive for IgG and polyspecific, and negative for C3, supporting a diagnosis of warm autoimmune hemolytic anemia (WAIHA). The indirect antiglobulin test (antibody screen) was positive. Antibody identification demonstrated a warm autoantibody mimicking anti-e. Extended red cell antigen typing confirmed the patient was e-positive and c-positive, establishing that the apparent anti-e specificity was autoimmune rather than alloimmune in nature. No clinically significant underlying alloantibodies were identified. With the autoantibody fully characterized, crossmatch-compatible packed red blood cells were issued and transfused without reaction. Potential contributors that were considered included post-transplant lymphoproliferative disorder (PTLD), Epstein-Barr virus (EBV) infection, tacrolimus-associated WAIHA, and immune dysregulation related to solid organ transplantation. EBV PCR and serology were sent.

She was started on prednisone 1 mg/kg daily on hospital day 1, with partial improvement in hemoglobin. She required a third unit of packed red blood cells on hospital day 3. A bone marrow aspirate and biopsy were performed. Due to continued hemoglobin instability with a further decline on hospital day 5 (from 8.6 to 7.7 g/dL), rituximab was initiated weekly with a planned total of 4 doses. After discussion with her transplant team, the tacrolimus dose was reduced from 5 mg to 4 mg daily (for a goal tacrolimus trough level of 2 to 4 ng/mL). Peripheral blood flow cytometry showed no immunophenotypically abnormal cell population. She was discharged on hospital day 6 with plans for outpatient rituximab and close laboratory monitoring.

Bone marrow biopsy revealed a hypercellular marrow (approximately 90% cellularity) with trilineage hematopoiesis and no histological or immunophenotypic evidence of a hematolymphoid neoplasm. Flow cytometry from the marrow aspirate showed blasts <1%, polytypic B lymphocytes (kappa/lambda ratio 1.4), polytypic plasma cells, and no abnormal T-cell or NK-cell populations. EBV testing was negative.

She completed 4 weekly doses of rituximab. Prednisone tapering was initiated after the second rituximab dose. Prednisone was initiated at 1 mg/kg/day (60 mg daily) and subsequently tapered as follows: 40 mg daily for one week, 20 mg daily for one week, 15 mg daily for 2 weeks. Hemoglobin stabilized with resolution of hemolysis; at follow-up, hemoglobin was 10.6 g/dL with normal bilirubin. She underwent weekly monitoring with CBC, CMP, LDH, haptoglobin, and reticulocyte count. Over approximately 4 to 5 weeks, she achieved remission with hemoglobin improving to 13.6 g/dL, LDH to 164 U/L, haptoglobin to 120 mg/dL, and reticulocyte count to 4.7%, with normal total bilirubin. Prednisone was further tapered to 10 mg daily for 2 weeks, then 5 mg daily for 2 weeks, followed by a reduction of 1 mg every 2 weeks. Laboratory surveillance was spaced to every 2 weeks to monitor for recurrence. Serial trends in hemoglobin, lactate dehydrogenase, and haptoglobin during treatment are illustrated in Figure 2.

Discussion

In this case, the patient underwent liver transplantation during childhood for biliary atresia and had remained on tacrolimus-based immunosuppression for many years without prior major complications. She presented more than 2 decades later with symptomatic anemia and was diagnosed with warm autoimmune hemolytic anemia. Although AIHA is an uncommon complication of solid organ transplantation, reports of late-onset disease occurring many years after transplantation remain limited [1,2,5].

When hemolytic anemia occurs in the post-transplant setting – particularly late after transplantation – important etiologies include post-transplant lymphoproliferative disorders, infection-associated hemolysis, and medication-related immune hemolysis, including calcineurin inhibitor-associated hemolysis [1,7–9]. In our patient, post-transplant lymphoproliferative disorder was considered but was felt to be unlikely based on negative peripheral flow cytometry and a bone marrow biopsy without evidence of hematolymphoid neoplasm. Infection was also less likely given the negative EBV serology and no clinical evidence to suggest an alternative infectious trigger. Tacrolimus-associated autoimmune hemolysis remains a diagnosis of exclusion; however, given the clinical context and the elimination of other major causes, it remained a plausible contributing factor.

Importantly, calcineurin inhibitors such as tacrolimus have been associated with 2 distinct forms of hemolytic anemia in the post-transplant setting: autoimmune hemolytic anemia (AIHA) and thrombotic microangiopathy (TMA). Tacrolimus-associated TMA is a well-recognized complication of solid organ transplantation, with a reported incidence of 1% to 4.7%, and is mediated by direct endothelial injury leading to microvascular thrombosis and mechanical red blood cell fragmentation [10,11]. In contrast, tacrolimus-associated AIHA is an immune-mediated process in which autoantibodies target red blood cells, resulting in extravascular hemolysis. The 2 entities can be distinguished clinically and by laboratory evaluation. TMA is characterized by thrombocytopenia, schistocytes on peripheral blood smear, a negative direct antiglobulin test, renal dysfunction, and hypertension, whereas AIHA is characterized by a positive direct antiglobulin test, spherocytes on peripheral smear, preserved platelet count, and absence of schistocytes [7,10,11]. In our patient, a strongly positive DAT, normal platelet count (229×109/L), absence of schistocytes on peripheral smear, and lack of renal deterioration or hypertension supported the diagnosis of warm AIHA rather than TMA. Autoimmune cytopenias, including AIHA, have also been described in pediatric intestinal transplant recipients on calcineurin inhibitor-based regimens, further supporting the association between tacrolimus and immune-mediated hematologic complications [4].

Post-transplant management of AIHA reported in the literature frequently requires more than 1 therapeutic modality to achieve remission. High-dose corticosteroids (prednisone 1 mg/kg daily) are commonly used as first-line treatment, with escalation to intravenous immunoglobulin, rituximab, plasmapheresis, or splenectomy in refractory cases [1,4,5]. In patients with warm autoimmune hemolytic anemia who require treatment beyond corticosteroids, rituximab has been associated with clinical responses comparable to those historically reported for splenectomy [12]. Limited evidence suggests that combining rituximab with corticosteroids in first-line therapy can improve outcomes compared with corticosteroid monotherapy in selected patients [12].

In selected reports of treatment-refractory hemolysis, conversion from a calcineurin inhibitor (tacrolimus) to an alternative immunosuppressive regimen – including mTOR inhibitors (sirolimus or everolimus) or cyclosporine – has been associated with resolution of hemolysis, supporting a potential drug-related mechanism in at least a subset of patients [3,13]. This pattern suggests that calcineurin inhibitor-associated immune hemolysis is under-recognized.

In our patient, although tacrolimus-associated AIHA remained a possibility, remission was achieved without complete tacrolimus discontinuation. Instead, she responded to corticosteroids and early rituximab as the first-line therapy, alongside a modest reduction in tacrolimus dose after discussion with the transplant team. Reports describing successful remission while continuing tacrolimus – without switching to an alternative immunosuppressant or making major dose reductions – are limited [5]. Notably, our approach incorporated rituximab early rather than waiting for response to corticosteroids alone, with a favorable outcome. However, evidence comparing corticosteroid monotherapy versus early addition of rituximab as first-line therapy in selected severe cases remains sparse, and further research is needed to guide optimal initial treatment strategies.

Conclusions

This case underscores the importance of maintaining a high index of suspicion for post-transplant–related and tacrolimus-associated hemolytic anemia, even decades after solid organ transplantation. Early recognition, prompt diagnostic evaluation, and timely initiation of therapy are essential to achieve remission.

Our case also suggests that not all patients require conversion from tacrolimus to an alternative immunosuppressive agent, and durable remission can be achieved while continuing tacrolimus with careful clinical and laboratory monitoring.

Finally, there remain limited data regarding the effectiveness of corticosteroids alone as first-line therapy versus early addition of rituximab in selected or severe cases. Additional studies are needed to clarify optimal treatment sequencing and to standardize management strategies in this rare post-transplant complication.