High-Oxygen-Affinity Hemoglobinopathy-Associated Erythrocytosis: Clinical Outcomes and Impact of Therapy in 41 Cases

Background

Approximately 100 high oxygen affinity (HOA) hemoglobin (Hgb) variants have been reported to date; review of the Mayo Clinic laboratory database (1974-2018) identified 80 distinct variants including 12 novel variants (60 β, 20 α). One-third of HOA Hgb variants result in secondary erythrocytosis, provoking concern regarding increased risk for thrombosis. Current management guidelines lack supporting evidence regarding the utilization of phlebotomy in such cases. We describe presenting features, treatment strategies and follow-up events involving 41 consecutive cases seen at our institution.

Methods

Study patients were recruited from our institutional laboratory and clinical Hgb variant database. Initial evaluation for Hgb variants was conducted by capillary electrophoresis and high-performance liquid chromatography. Additional testing which included mass spectrometry and isoelectric focusing was pursued as necessary. Since the majority of variants were difficult to identify by protein studies, DNA sequencing of HBB, or HBA1, HBA2 genes was performed for confirmation. Symptoms and thrombosis, both at presentation and during follow up, were recorded. Therapeutic interventions were based on physician discretion and mostly included phlebotomy and/or aspirin therapy; a careful response assessment was performed to determine the impact of each therapy on symptoms and/or thrombosis.

Results

A total of 41 patients with HOA Hgb variant-associated erythrocytosis (median age 39 years, range 1-81; 54% males) were seen at our institution between January 1973 and February 2020. The majority of the patients carried β-chain variants (n=34; 83%), common variants being Hgb Malmo (n=13), Olympia (n=4), San Diego (n=3), and Wood (n=2). Among the 7 patients with α-chain variants, Hgb Dallas was the most frequent (n=4).

Presenting median values (range) for Hgb/Hct, serum erythropoietin and p50 were 18 g/dl/52.9%(16-21.9g/dl/48-66%) 10.4 mIU (4-36.3 mIU), and 20 mmHg (12-25 mmHg), respectively. Family history was documented in 34 patients, of which 24 (71%) reported one or more affected family members. Family history of thrombosis was documented in 7 patients (21%). CV risk factors were present in half of the patients; by contrast, history of thrombosis prior to or at diagnosis was documented in only two patients (5%). Of 23 pregnancies reported in 12 women, live birth rate was 78% (n=18); none of the fetal losses were attributed to erythrocytosis. Active therapies at the time of initial referral consisted of phlebotomy (n=12), aspirin (n=11) and systemic anticoagulation (n=1).

At a median follow-up of 10 years (range; 0.04-44), 23 patients had reported one or more symptoms, attributed to hyper-viscosity, such as headaches, fatigue, and lightheadedness. Neither Hct level at diagnosis (p=0.32) nor phlebotomy (p=0.16; 75% patients on vs 52% not on phlebotomy) or aspirin therapy (p=0.75; 55% patients on vs 60% not on aspirin) appeared to influence the occurrence of symptoms. Phlebotomy relieved symptoms in 7 (42%) symptomatic patients; however, 7 (30%) of 23 patients on phlebotomy reported one or more adverse symptoms that were attributed to phlebotomy-induced iron deficiency.

Ten patients (24%) experienced thrombosis prior to or following diagnosis: 6 arterial and 4 venous. Median age at thrombotic event was 51 years and median hematocrit 52%; active therapies at the time of event included phlebotomy in 5 patients, aspirin in 4, and systemic anticoagulation in 2. Hct level at diagnosis (p=0.10) or the time of event (p=0.67) did not correlate with occurrence of thrombosis. Additionally, the incidence of thrombosis was no different among patients receiving or not receiving phlebotomy (5/23(22%) vs 5/18(28%), respectively; p=0.66). The presence of CV risk factors was predictive of arterial events (p=0.002). Two of the 4 venous events developed in the context of concomitant thrombophilia. Eight patients (20%), median age 28 years, without CV risk factors, were observed without therapy for a median of 9.5 years (range; 0.4-21) and have not experienced any thrombosis to date.

Conclusions

We found no association between Hct level and either thrombotic or non-thrombotic symptoms in HOA hemoglobinopathy-associated erythrocytosis; furthermore, implementation of aggressive phlebotomy did not provide a clear benefit with respect to thrombosis risk reduction.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal