Thromboembolism in Individuals with High-Oxygen-Affinity Hemoglobins - Systematic Review of the Literature

Background: High-oxygen-affinity hemoglobins (HOAH) are rare conditions, yet the most prevalent cause of constitutional erythrocytosis. Most data derive from individual and small case series and there is no consensus about their optimal therapeutic management. Thromboembolic events have been described in patients with HOAH-associated erythrocytosis but the actual risk is unknown. We conducted a review of literature in which we focused on initial clinical and biological presentations, thrombotic complications, and treatment strategies at diagnosis and during follow-up.

Methods: To conduct this review, we led a systematic search of the PubMed database with terms “hemoglobin”, “variant”, “high affinity”, “oxygen” and “erythrocytosis”. A Pubmed search with all HOAH variants described in the HbVar database was also conducted. Two reviewers independently screened all identified studies. The search was limited to human studies published in English and studies without any clinical data were removed. In the event of multiple case reports in one single publication, every patient was added individually to the database used for the analysis.

Results: We identified 632 patients with HOAH reported in the literature with a median age at diagnosis of 32 years (interquartile range [IQR]: 20-50) with 55% males and 45% females. We recorded 122 variants with 508 (81.2%) beta-chain and 118 (18.8%) alpha-chain variants. Erythrocytosis was observed in 82% of the patients with median hemoglobin (Hb) of 17.4 g/dL (IQR: 16.0-18.9) and median hematocrit (Hct) of 52% (IQR: 48.0-56.0). Median erythropoietin serum level was 12 UI/L (IQR: 8.0-22.0) while median p50 was 17.0 mmHg (IQR: 14.0-20.0).

Hyperviscosity symptoms were reported in 86 (17%) patients with Hb and Hct significantly higher in these patients (median Hb: 18.9 g/dL [IQR: 17.7-20.0] versus 17.2 g/dL [IQR: 15.9-18.6], p<0.001; median Hct 57% [IQR: 48-56] versus 52% [IQR: 48-56], p<0.001). Thrombosis at diagnosis occurred in 38 patients (7.3%), with 30 (62.5%) venous thrombosis (VT; 16 deep VT, 7 pulmonary embolisms, 3 superficial VT, 2 retinal vein occlusions, 1 temporal vein occlusion and 1 cerebral VT) and 18 (37.5%) arterial thrombosis (6 strokes, 7 myocardial infarctions, 3 transient ischemic attacks, 1 acute leg arterial occlusion and 1 renal artery obstruction). Age (55 years [IQR: 40-65] versus 31 years [IQR: 20-48], p<0.001), arterial hypertension (23% versus 7.1%, p=0.014), thrombophilia, (22% versus 1.4%, p<0.001; heterozygous factor V Leiden mutation [FVL] and heterozygous protein S [PS] deficiency), Hb (18.3 g/dL [IQR: 17.6-19.8] versus 17.4 g/dL [16.0-19.0], p=0.008), and Hct (56% [IQR: 53-62] versus 52% [IQR 48-57], p<0.001) were associated with increased thrombosis.

Antiplatelet therapy was initiated in 45 patients (34%) and anticoagulant in 25 patients (18%) while phlebotomy was started in 128 patients (31%) with missing data for 221 patients. Among the 121 patients for whom follow-up data regarding thrombosis was available, 12 (10%) developed thrombosis. Thrombophilia was significantly higher in this population: 3 patients (25%; heterozygous prothrombin gene G202101A and FVL mutations, heterozygous PS deficiency) versus 3 patients (3.7%; heterozygous FVL mutations) without thrombosis (p=0.026). No statistical difference in thrombosis onset during follow-up was observed between phlebotomized patients and those who were not, with respectively 7 and 5 thromboses (p=0.9). No statistical difference in hyperviscosity symptoms during follow-up was seen between phlebotomized patients and those who were not, with respectively 10 and 7 patients (p=0.7).

Conclusion: Due to the rarity of HOAH, with limited individual data available, we led a systematic review of the literature to gain insight on thromboembolism in these patients. While a large number of patients were asymptomatic and did not have any complication at diagnosis or during follow-up, thrombosis was observed in 7% of the patients at diagnosis and in 10% of patients during follow-up for whom it was available. Although Hb and Hct levels were associated with thrombosis, phlebotomies did not modify the risk of complications during follow-up. Because of the inherent risk of bias of our study design, more studies with systematic follow-up or with a prospective design are needed to assess the role of phlebotomy in individuals with HOAH.

No relevant conflicts of interest to declare.