High‐Altitude Hypoxemia in Adults With Sickle Cell Disease ( <scp>SCD</scp> )

Sickle cell disease (SCD) is characterized by abnormal polymerization of deoxygenated hemoglobin S. This polymerization leads to the distinctive sickling of red blood cells (RBC), and resultant painful episodes and progressive multiorgan vasculopathy [1]. At high altitudes, barometric pressure drops, as does the partial pressure of oxygen (PaO2), also known as “hypobaric hypoxia” [2]. Above ≥ 8000 ft (the approximate air pressure of commercial air travel), atmospheric hypoxia can exacerbate or uncover occult hypoxemia in high-risk people, including those with SCD and other cardiopulmonary diseases [3-5]. As a way to assess the risk of hypoxemia (low blood oxygen) prior to travel, the hypoxia altitude simulation test (HAST) simulates high altitude conditions to assess for tolerance and for the need for prophylactic supplemental oxygen [6]. At our institution, following anecdotal reports from people living with SCD of symptom exacerbation with or following air travel, we have routinely offered HAST evaluations as part of pretravel evaluation for our patients. The goal of this cross-sectional study is to assess the prevalence of high-altitude hypoxemia in tested adults with SCD over a 5-year period, and to explore the association between high-altitude hypoxemia and disease characteristics. After institutional review board approval, we completed a retrospective chart review of patients at the University of North Carolina Comprehensive Sickle Cell Center. We included patients ≥ 18 years old who completed high-altitude simulation testing (HAST) between January 1, 2019, and June 30, 2025. This test is offered to all patients who inform us of plans for travel. The study's main objective was to assess the prevalence of high-altitude hypoxemia and the onset of hypoxemia in this select cohort. The secondary objective of the study was to investigate the relationship between high-altitude hypoxemia and participant characteristics. At our institution, HAST was performed with the use of a Hans Rudolph two-way non-rebreathing valve system, and oxygen saturation was monitored with a forehead oximeter. Compressed Air (10 L per minute) and Nitrogen (8 L per minute) were blended to an oxygen level of 15%, simulating an altitude of 8000 ft, the maximum allowable altitude in an aircraft, for 20 min [7]. High-altitude hypoxemia was defined as persistent oxygen saturation (SpO2) ≤ 88% at 15% inspired oxygen, requiring supplemental oxygen during the test. This definition was based on eligibility criteria for oxygen therapy under the Center for Medicare and Medicaid Services (CMS) at the time of the study. We also collected the time to onset of hypoxemia and the minimum SpO2 recorded during testing. Additional data extracted from electronic health records were demographics, comorbidities, medication and opioid prescriptions, healthcare utilization within 1 year of HAST, clinical labs obtained around the time of HAST testing, and echocardiography results (tricuspid regurgitant jet velocity, TRJV), if available. Clinical labs included complete blood count, hemoglobin electrophoresis, absolute reticulocyte count (ARC), lactate dehydrogenase (LDH), albumin-to-creatinine ratio (ACR), estimated glomerular filtration rate (eGFR), haptoglobin, total bilirubin, C-reactive protein (CRP), and ferritin. For the primary outcome measure, HAST results were reported as a binary variable (positive/negative) and continuous variable (time without hypoxemia). After assessing data for missingness and assessing continuous variables for normality, participant characteristics were reported as frequency (percent) and means (standard deviation) for categorical and continuous data, respectively. For bivariable analysis, comparison of participants with and without high-altitude hypoxemia was completed using Fisher's exact test for categorical variables and appropriate parametric t-tests for continuous variables. Comparison of the onset of hypoxemia by categorical and continuous variables was completed using 2 sample t-test/ANOVA and Pearson's correlations, respectively. For exploratory analyses, we used logistic regression to identify characteristics with independent associations with development of hypoxemia. p-values of < 0.05 were considered statistically significant. Statistical analyses were performed using Stata 18 (StataCorp LP, College Station, TX). We identified 31 adult patients who underwent HAST during the specified study period for analysis. Two participants completed multiple HAST evaluations, so we selected their most recent or abnormal test result in one each. Most tested participants were female (19/31 participants) and had Hemoglobin SS genotype (17/31). The mean age and BMI were 35.2 years and 25.9 kg/m2, respectively. At the time of testing, 6 participants had a diagnosis of mild obstructive sleep apnea, 2 participants had a diagnosis of pulmonary hypertension, and 1 participant was a current smoker. Twenty-one participants were prescribed opioids with a daily mean of 49.6 morphine milligram equivalents (MMEs). The most common sickle cell complications in this cohort were: avascular necrosis (16/31), depression/anxiety (15/31), sickle nephropathy (9/31), venous thromboembolism (8/31), and cerebrovascular accident (7/31). Fourteen participants were on hydroxyurea, and 3 participants were receiving routine exchange transfusions. > 3% of data for total bilirubin, ferritin, LDH, eGFR, CRP, Haptoglobin, TRJV was missing for this analysis; otherwise, participants had complete data available. During the 20-min HAST, 18 participants (58%) developed high-altitude hypoxemia and 13 participants did not. Most participants (14/18) that tested positive developed hypoxemia within 4 min (Figure 1) of testing. On average, for those who tested positive, it took 3.85 min (range 1–16 min) to develop hypoxemia; the minimum SpO2 recorded was 86% with a range of 86%–89%. Of note, once actionable hypoxemia (i.e., SpO2 ≤ 88%) developed during testing, supplemental oxygen was administered and an improved SpO2 was confirmed. Compared to those who tested negative for high-altitude hypoxemia, those who tested positive had lower BMI (mean 24.6 kg/m2 vs. 27.7 kg/m2, p = 0.046), and were likely to have sickle cell anemia (Hg SS or Sβ0) genotype (83.3% vs. 38.46%, p = 0.021). On average, participants with sickle cell anemia developed hypoxemia 3.88 min faster (95% CI –9.683, 1.930; p = 0.1759) than those with variant disease. The two groups did not differ significantly by other clinical characteristics, including history of obstructive sleep apnea, healthcare utilization, opioid prescription, and hydroxyurea use (Table 1). The participants that developed hypoxemia had lower baseline hemoglobin level (9.1 g/dL vs. 11 g/dL, p = 0.002), higher ARC (254.3 × 109/L vs. 153.0 × 109/L, p = 0.006), higher WBC (9.1 × 109/L vs. 6.7 × 109/L, p = 0.019), and higher total bilirubin (3.7 mg/dL vs. 1.7 mg/dL, p = 0.008) levels, when compared to those that tested negative. Other lab characteristics are listed in Table 1. In the participants that developed hypoxemia, the strongest correlations with a faster onset of hypoxemia were the following labs (Figure 2): WBC (r = −0.5575, p = 0.0162), ARC (r = −0.5004, p = 0.0344) and hemoglobin (r = 0.3462, p = 0.1594). As part of exploratory analysis, unadjusted logistic regression showed that the following characteristics increased the odds of developing hypoxemia: sickle cell anemia (OR 8, 95% CI 1.51, 42.45), hemoglobin level (OR 0.42, 95% CI 0.21, 0.81), white cell count (OR 1.48, 95% CI 1.03, 2.12), and total bilirubin (OR 2, 95% CI 1.10, 3.65). In a full multivariable logistical regression model, none of the characteristics independently predicted development of hypoxemia. Adults with SCD are at high risk for developing hypoxemia with altitude, as in air travel and mountainous destinations. To our knowledge, this is the first study to systematically examine hypoxemia with high altitude in this population. In participants undergoing pre-travel evaluation, we found evidence of actionable high-altitude hypoxemia in more than half of participants. Interestingly, we noted that the development of hypoxemia was rapid in many of these participants, raising the question of the underlying pathophysiological mechanism by which SCD impacts hypoxemia. Similar to other studies of occult hypoxemia [8-10] in people with SCD, we saw an association between the presence of high-altitude hypoxemia and increased disease activity. This was reflected by increased anemia, reticulocytosis, and inflammation in those with high-altitude hypoxemia. In this population, exposure to hypobaric hypoxia presumably increases the concentration of deoxygenated-Hb S and the polymerization of hemoglobin S in circulating RBCs. We believe that it is likely that people with labs suggestive of more disease activity (excessively increased LDH or reticulocytes, or excessively decreased hemoglobin) are at increased risk for clinical decompensation following exposure to high altitude. While a larger study will be needed to make generalizable conclusions, our results suggest that SCD providers should be proactive in inquiring about travel plans and consider assessment of patients' risk for hypoxemia during air travel or high-altitude exposure. Our study is limited by the retrospective design with small sample size, and a lack of patient reported outcomes about altitude associated symptoms or complications. Additionally, there may be a selection bias towards “healthier” adults with SCD, that is, those who are well enough to travel. In addition to completing this assessment on a larger and varied sample of adults with SCD, uncovering the potential mechanism of hypoxemia will be important. Mofiyin A. Obadina: conceptualization, data curation, formal analysis, methodology, visualization, writing – original draft, and writing – review and editing. Sherri Morris: conceptualization, data curation. Tara Alin: conceptualization, data curation. Barbara LeVarge: conceptualization, data curation. Jane A. Little: conceptualization, resources, writing – review and editing, supervision. Special thanks to our patients who contributed to the completion of this study. M.O.: National Heart, Lung, and Blood Institute (T32HL007149–47) and American Society of Hematology RTAF award. J.A.L.: American Society of Hematology and collaborations with UAB and JH (PCORI and NIH funded projects). This study was approved by local institutional review board (IRB # 21-0274), and the need for informed consent was waived due to the retrospective nature of the study. J.A.L.: Research support from Pfizer, NASCC and USC; Adjudication committee for FORMA's Hibiscus study, Patent holder for device commercialized by BioChip Labs and Hemex. The data that support the findings of this study are available from the corresponding author upon reasonable request.