A Prospective Study of Clinical Outcomes in SARS CoV-2 Pneumonia Patients with Low Vitamin D State

Title: A Prospective Study of Clinical Outcomes in SARS CoV-2 Pneumonia Patients with Low Vitamin D State
Author(s): M. Youssef ¹ R. Barrett ² I. Shah¹ J. Ioana¹ A. Al Lawati ¹ A. Bukhari ¹ S. Hegarty ¹ L. Cormican ¹ E. Judge ¹ C. Burke ¹ C. Cody ⁴ J. Feely ⁵ K. Hutchinson ³ W. Tormey ⁵ E. O’ Neill ⁵ A. O’Shea ² M. Connolly ² D. McCartney ² J. L. Faul ¹
Institution: ¹ Respiratory Department, Connolly Hospital, Dublin, Ireland; ² School of Biological and Health Sciences, Technological University, Dublin, Ireland; ³ Eurofins Biomnis, Dublin, Ireland; ⁴ Department of Critical Care Medicine, Connolly Hospital, Dublin, Ireland; ⁵ Microbiology and Biochemistry Department, Connolly Hospital, Dublin, Ireland.
Poster: Click to view poster
Category: COVID 19
Abstract: Introduction: Covid-19 disease incidence, severity and mortality are higher in populations with low Vitamin D levels. Risk factors for a low vitamin D state overlap with the main susceptibility factors for Covid-19 (e.g. advanced age, obesity and darker skin pigmentation). It is unclear whether low Vitamin D status contributes causally to increased severity of Covid-19 disease, or it might simply be present by chance, or as a consequence of the disease process (e.g. catabolism of vitamin D by pronounced systemic inflammation (i.e. reverse causality)).Our own experience showed poor clinical outcomes in SARS-CoV-2 pneumonia patients with low vitamin D.This study aimed to establish if low vitamin D status was associated with more severe Covid- 19 disease and whether any identified relationship was causal in nature, or might be explained by confounders.

Methods: Prospective observational cohort study examining serum vitamin D levels, Covid -19 disease severity and outcomes in 232 consecutive patients from a total of 972 patients admitted to Connolly Hospital with SARS-CoV-2 pneumonia in 2020.All 232 patients had abnormal chest radiographs. Background data including age, gender, BMI and comorbidities were recorded. We examined clinical outcomes: length of admission, duration of supplementary oxygen therapy, admission to the intensive care unit and mortality. Biochemical and haematological laboratory parameters were recorded; serum 25(OH)D levels were checked at time of admission. Univariate analysis, binary and multivariate logistic analyses were used to determine whether measures of inflammation and clinical outcome were associated with vitamin D status.

Results: 232 patients, mean age 58 years (range 17-99), 60% male. 38% (n=88) had deficient (25(OH) D <30 nmol.l-1), 26% (n=60) had insufficient (25(OH) D of 30-49 nmol.l-1) and36% (n=84) had sufficient (25(OH) ≥50 nmol.l-1). Patients aged ≥70 years were significantly less likely to be vitamin D deficient (VDD) (OR: 0.520 (95%CI: 0.285- 0.951) (p = 0.034). Gender, presence of diabetes mellitus, presence of obesity (BMI ≥30 kg.m-2) and renal disease were not independently associated with VDD.

Biochemical Parameters: Vitamin D was not predictive of any haematological or inflammatory markers.

Oxygen (O2) Requirement: Male gender was a significant predictor of the need for supplemental O2 >24 hours (OR: 3.22 (95% CI: 1.26-8.22) (p = 0.014). After adjusting for male gender VDD was not a significant predictor of the need for supplemental O2 >24 hours (OR: 2.11 (95% CI: 0.85-5.39) (p = 0.116).

ICU Admission: Male gender (OR: 2.90 (95% CI: 1.29-6.51) (p = 0.01) and diabetes mellitus (OR: 2.82 (95% CI: 1.21-6.53) (p = 0.016) were independent predictors of ICU admission. After adjustment for age, male gender and diabetes Mellitus, VDD was not a significant predictor of ICU admission (OR 1.18 (95% CI: 0.58-2.39) (p = 0.632).

Mortality : After adjustment for age, gender, BMI and diabetes mellitus, the hazard ratio for mortality was 4.6 in VDD patients compared to patients with serum 25(OH)D above the 30nmol.l-1 threshold (HR:4.63; 95% CI:1.53-13.97 ) (p = 0.006). Age ≥70 was a strong predictor of mortality (OR: 13.32 (95% CI: 4.24-41.81) (p<0.001). Diabetes mellitus, male gender and obesity were not independently associated with mortality.

Conclusion: A low vitamin D in Caucasian adults with SARS-CoV-2 pneumonia is closely associated with mortality, but not with any common clinical marker of systemic inflammation. This suggests that a low vitamin D state precedes the severe inflammation which characterizes severe Covid-19 disease.