ABSTRACT
RATIONALE: A genetic locus at chromosome-3 was recently identified as a risk factor for respiratory failure during COVID-19, and preliminary in-silico analyses demonstrate a strong association between this locus, elevated serum lactate dehydrogenase (LDH), and respiratory failure. To understand if this locus may affect infection-mediated acute lung injury in other contexts, we tested whether serum LDH and the chromosome-3 locus were associated with ARDS risk and severity in critically ill adults with non-COVID infections. METHODS: We studied 553 critically ill adults with sepsis enrolled in the Validating biomarkers in Acute Lung Injury for Diagnosis (VALID) study. All patients were prospectively phenotyped for ARDS (Berlin Definition) during the first 4 ICU days by two physician investigators. For genetic analyses, we used a convenience sample of 289 septic VALID patients with genomewide genotyping from prior studies. We extracted data on the lead single nucleotide polymorphism in chromosome-3 (rs35081325;wild-type: A;risk variant: T) identified in the COVID-19 Host Genetics Initiative's analysis A2 (COVID-19 with severe respiratory failure). We assessed severity of illness by APACHE-II scores and oxygenation impairment by SpO2:FiO2 ratio. Between-group comparisons were performed using linear regression for continuous outcomes and logistic regression for categorical outcomes. We report effect odds ratios (OR) and 95% confidence intervals (CI). An inverse normal quantile transformation was applied to clinically ascertained LDH values to account for skewness and non-normality;associations are reported per normalized standard deviation (SD) of the transformed LDH value. RESULTS: Serum LDH was higher among patients with ARDS than without ARDS when controlling for age, gender, and APACHE-II (OR=1.20;95%CI: 1.01-1.43;P=0.04). Serum LDH was also associated with oxygenation impairment, particularly among patients with ARDS (FIGURE). Among genotyped patients, the rs35081325 T-allele was associated with higher serum LDH levels (0.62 normalized SD higher LDH;95%CI: 0.16-1.08;P=0.009), and exhibited trends for higher severity of illness (2.10 higher APACHE-II score per T-allele;95%CI:-0.56 to 4.77;P=0.19), increased ARDS risk (OR=1.55 per T-allele;95%CI 0.70-3.44;P=0.28;FIGURE) and increased in-hospital mortality (OR=2.24 per T-allele;95%CI: 1.00-5.05;P=0.05). CONCLUSION: Serum LDH and rs35081325 on chromosome-3 were associated with ARDS risk and oxygenation impairment in a large cohort of septic adults, suggesting a shared host genetic risk for severe respiratory failure among COVID-19 and other etiologies of infection-mediated acute lung injury. As this SNP is near several genes involved in chemokine function, autophagy, and solute transport, further mechanistic investigation is necessary to identify the causative gene(s).