Evaluation of serum 25‐hydroxyvitamin D concentrations in a heterogeneous canine ICU population

OBJECTIVE: To identify the prevalence of low serum 25-hydroxyvitamin D (25-OH-D) concentrations in critically ill dogs and determine whether vitamin D concentration measured at admission correlates with illness severity and patient outcome. DESIGN: Prospective observational study from December 2013 to July 2014. SETTING: Private referral teaching hospital ICU. ANIMALS: One hundred eleven critically ill dogs admitted to the ICU, and 28 apparently healthy control dogs. INTERVENTIONS: Measurement of 25-OH-D concentration and assessment of physiological parameters required to calculate illness severity scores. MEASUREMENTS AND MAIN RESULTS: Blood was sampled to measure serum 25-OH-D concentration, point-of-care laboratory data (packed cell volume, total plasma protein, venous blood gas, electrolytes, plasma lactate, and blood glucose concentration), platelet count, and serum albumin within 12 hours of admission. Primary disease etiologies were determined. Mentation score, Acute Patient Physiologic and Laboratory Evaluation (APPLE) fast score, length of ICU hospitalization, and survival to discharge were recorded. Low serum 25-OH-D concentrations were detected in 25% (28/111) of the ICU cohort; 25-OH-D concentrations (median; IQR) (89.2; 26.3 ng/mL [222; 140.0 nmol/L]) were significantly lower in hospitalized dogs compared with the healthy cohort (127.5; 44.5 ng/mL [318; 111.0 nmol/L]) (P < 0.0001). Serum 25-OH-D concentrations had a weak positive correlation with albumin (r = 0.34, P = 0.0003), but not with any other blood analyte evaluated. Serum concentration of 25-OH-D was also weakly correlated with APPLEfast score (r = 0.31, P = 0.001). 25-OH-D concentrations (median; IQR) were significantly higher in ICU dogs that survived to discharge (95.5; 55 ng/mL [238; 137 nmol/L]) compared with nonsurvivors (60.0; 54 ng/mL [149; 135 nmol/L]; P < 0.001). CONCLUSIONS: Low serum 25-OH-D was recorded in one-quarter of critically ill dogs, and was significantly lower in nonsurvivors in comparison to survivors. There was weak correlation between serum 25-OH-D and illness severity. Further studies are required to clarify relationships between vitamin D status and outcome.