Ongoing Problems with Research on Vitamin D in Critical Illness. Comment on Admission 25‐hydroxyvitamin D levels are associated with functional status at time of discharge from intensive care unit in critically ill surgical patients

Abstract

Brook and others in a recent issue of our journal 1 reported that admission 25-hydroxy vitamin D (25OHD) level in critically ill patients was associated with discharge functional status score. Higher scores implying better functional status were associated with higher initial 25OHD levels. The authors suggest that studies are needed to determine if correcting low 25OHD levels (to a level of about 30 ng/mL) can improve discharge functional status of intensive care unit (ICU) patients.The study group included patients specifically recruited for this study as well as a smaller number from previously published studies on the association of vitamin D status and mechanical ventilation and mortality. In the present combined report, blood samples were collected in 300 patients within 24 hours of admission to the ICU.Many publications have reported low vitamin D levels in critically ill patients, as detailed in the present paper, but the significance of vitamin D status and the need to correct it are not well understood. As we have pointed out in an earlier publication, 2 25OHD levels obtained after ICU admission do not accurately reflect premorbid vitamin D status, as fluid resuscitation causing significant hemodilution most often begins in the emergency department even prior to transfer to an actual ICU bed. Levels further decrease during continued fluid administration, leading to false low levels when blood samples are collected 24 hours later. Although newer recommendations are to limit fluid administration in postsurgical patients and in septic shock, adequate resuscitation with fluids is the first and immediate line of therapy for hemodynamically unstable patients. Thus, many (but certainly not all) of the previous publications on the role of vitamin D in critical illness have not accurately identified deficiency states. Patients who were never vitamin D deficient are wrongly classified as such. Decreases in levels of other micronutrients (eg, selenium, thiamine) after cardiac surgery, where massive fluid shifts are inevitable, have also been reported. 3,4 In a previous study, we obtained blood samples at the time of induction of anesthesia in cardiac surgery patients, providing data for the genuine premorbid vitamin D status.We found no association between 25OHD levels and postoperative cardiac function, dysglycemia, and length of stay, all of which can affect eventual functional status. 5 The authors conclude that "admission vitamin D status may be a modifiable risk factor" to influence better functional status at discharge. We submit that such a study is not feasible, unless we have truly premorbid (ie, prior to hospital visit or transfer to ICU from a floor bed) levels, something that is quite difficult, if not impossible, to predict and obtain. References 1. Brook K, Otero TMN, Yeh, DD et al. Admission 25-hydroxyvitamin D levels are associated with functional status at time of discharge from intensive care unit in critically ill surgical patients. Nutr Clin Pract. 2019;34(4):572-580. 2. Sriram K, Meguid MM. The...

JPEN Journal Club 78. Response‐adaptive randomization

“…His belief was that the baseline (and allegedly low) levels in these trials may be artifacts related to hemodilution. 3 This perspective suggested to me that the patients were not really vitamin D deficient and that one might not expect any benefit from vitamin D provision. Although this is a consideration in recently hydrated patients (recently admitted critically ill ones), such a situation did not exist in these older adult individuals.…”

“…Dr Sriram provided me with his view of low serum levels of vitamin D, albeit his comments were directed at critically ill patients. His belief was that the baseline (and allegedly low) levels in these trials may be artifacts related to hemodilution 3 . This perspective suggested to me that the patients were not really vitamin D deficient and that one might not expect any benefit from vitamin D provision.…”