The weight of stress: Allostatic load as a predictor of vaso-occlusive episodes in adults with sickle cell disease

Abstract Introduction: Sickle cell disease (SCD) is a debilitating hemoglobinopathy caused by a point mutation in the HBB gene, inducing hemoglobin S polymerization and erythrocyte sickling under hypoxia. Clinical manifestations include recurrent vaso-occlusive episodes (VOEs) leading to severe pain, organ damage, and reduced life expectancy. Emerging pre-clinical studies suggest psychological stress may exacerbate VOEs through inflammatory-mediated pathways, yet clinical tools to measure chronic stress are lacking. Allostatic load (AL), a multisystem index of physiological stress, quantifies the cumulative biological toll of chronic stress but has not been studied in SCD. This study evaluated the feasibility of AL in adults with SCD and whether high AL is associated with increased VOE frequency. Methods: We conducted a retrospective cohort study of adults (≥18 years) with SCD, regardless of genotype or disease-modifying treatment, receiving care at Augusta University. Twenty-seven clinical biomarkers representing cardiovascular, metabolic, hematologic, and renal/hepatic systems were used to evaluate AL. Biomarkers included pulse, systolic and diastolic blood pressure, body mass index, sodium, potassium, chloride, carbon dioxide, glucose, calcium, total protein, albumin, white blood cell count, red blood cell count, hematocrit, hemoglobin, mean corpuscular volume, platelet count, neutrophils, reticulocyte count, aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, bilirubin, blood urea nitrogen, creatinine, and estimated glomerular filtration rate. Each biomarker was scored as 0 if within and 1 if outside the normal clinical range. The sum of abnormal biomarkers determined the AL score. Subjects dichotomized into high and low AL groups based on the cohort median (<9 vs. ≥10). VOE frequency was determined over 24 months, defined as pain crises requiring emergency department (ED) or inpatient medical care. The VOE burden dichotomized at the median. Chi-square tests, odds ratios (ORs), and logistic regression models determined associations between variables. Results: Preliminary analyses included 43 individuals, all African American, with a median age of 35 years (range: 18–67), of whom 58.1% were female. Most had hemoglobin SS genotype (90.7%) and were not receiving disease-modifying therapy (79.1%). The AL scores ranged from 5 to 14 (median 10). The median number of VOEs was 2 (range: 0–29); median ED visits and hospitalizations was 1 (range: 0–15 and 0–22, respectively). Patients in the high AL group experienced significantly more frequent VOEs (82.6% vs. 17.4%, p=0.01) and ED visits (78.3% vs. 21.7%, p=0.01) compared to those with low AL. Hospitalizations were also more common in the high AL group (69.6% vs. 30.4%) but were not statistically significant (p=0.32). In univariate logistic regression models, high AL was associated with a greater than seven-fold increased odds of frequent VOEs (OR 7.1, 95% CI 1.7–28.9, p=0.02) and five-fold increased odds of ED visits (OR 5.4, 95% CI 1.4–20.5, p=0.01). The association with hospitalization remained nonsignificant. To explore potential confounders, we conducted a multivariate logistic regression analysis including fetal hemoglobin (HbF) levels and therapy status (hydroxyurea, voxelotor, or crizanlizumab). High AL remained a strong and independent predictor of pain episodes (OR=7, 95% CI: 1.5–32.1, p=0.01), even after adjusting for these factors. Neither HbF nor therapy use was significantly linked to pain in this model. A separate analysis evaluating predictors of ED utilization revealed similar patterns. Elevated AL was associated with substantially higher odds of ED visits (OR=6.43, 95% CI: 1.33–31.22, p=0.02). Treatment with disease-modifying agents also showed an association with ED use (OR=15.8, 95% CI: 1.1–235.9, p=0.05), while HbF ≥15% again did not reach statistical significance. Conclusions: Our preliminary findings suggest that the AL index is a feasible and practical tool for assessing cumulative stress burden in adults with SCD, as it relies on biomarkers routinely collected in clinical care. This pilot study provides early evidence that higher AL is independently associated with increased VOE frequency and ED utilization. Incorporating AL into clinical assessments may help identify high-risk individuals who could benefit from disease-modifying therapy. Enrollment is ongoing (target n=180); updated results will be presented at the annual meeting.