When there is inadequate sunlight, dietary sources of vitamin D are required to maintain adequate tissue levels. Unfortunately, few foods (other than fish, mushrooms, and fortified milk products) naturally contain or are fortified with vitamin D. Twenty-five-hydroxyvitamin D (25OHD) is the major form of vitamin D in the circulation (plasma levels 30–100 ng/mL or 75–250 nmol/L) and has a half-life of approximately 2 weeks. Calcitriol is the most potent and primary active form of vitamin D (plasma levels 24–65 pg/mL or 58–156 pmol/L) and has a circulating half-life of approximately 4 to 15 hours. The 25-hydroxylase enzyme in the liver is not subject to endocrine regulation; as a result, levels of 25OHD reflect the supply of cholecalciferol and ergocalciferol and are used to assess patients for vitamin D substrate supply (from both skin and diet).
Vitamin D is transported primarily in blood bound to the specific plasma carrier protein, serum vitamin D–binding protein (DBP). DBP is a globulin protein synthesized by the liver and abundant in plasma. It has sequence homology to albumin and a high capacity for binding vitamin D. DBP binds approximately 88% of 25OHD in blood; the remaining 12% of 25OHD is bound by albumin. DBP and albumin decrease in patients after tissue injury and in patients with nephrotic syndrome. Loss of these proteins in the urine, along with bound vitamin D, may lead to vitamin D deficiency. Adipose tissue and muscle are major storage sites for vitamin D. Interestingly, obese patients may develop vitamin D deficiency as the result of sequestration of vitamin D in fat tissues.
Vitamin D has diverse actions throughout tissues.3,4 Most cells have a vitamin D receptor, and several cell types (e.g., bone, muscle, heart, vasculature, brain, prostate, breast, immune, and colon) in addition to kidney (Fig. 1) possess the enzymatic capability to convert circulating 25OHD to calcitriol.3 As a typical steroid hormone, calcitriol mediates most of its effects on the nucleus by regulating gene expression through the vitamin D receptor. More than 200 human genes are regulated by calcitriol,3 including those responsible for cellular proliferation, differentiation, apoptosis, and angiogenesis. As a part of the parathormone–vitamin D–Ca axis, the primary functions of vitamin D are to promote calcium absorption in the gut and reabsorption in the kidneys to maintain normal circulating levels of calcium and phosphorus and to promote bone mineralization and remodeling. In addition to the regulation of mineral and bone metabolism, vitamin D is implicated in muscle function, cardiovascular disease, cancer development, inflammation, immunity, and insulin sensitivity.4,5 Recent smaller studies have associated vitamin D replacement with improved mood state in ambulatory patients with depression; however, this effect could not be replicated in a larger study of acutely hospitalized patients.6
DEFINITIONS AND CAUSES OF VITAMIN D DEFICIENCY
A key issue discussed in the study by Turan et al.1 is the definition of vitamin D deficiency. There are 2 major concerns with the definition of vitamin D deficiency: the assays used may not reflect cellular exposure to vitamin D, and levels of vitamin D adequacy have not been well defined across populations and clinical end points.7
With respect to measurement, vitamin D status usually is estimated using a single measurement of the circulating levels of 25OHD. Circulating levels alone may not always predict deficiency of vitamin D at the cellular level but will serve as a reasonable marker of the supply of vitamin D to the body. With current evidence, it is unclear what levels of circulating 25OHD correspond to specific tissue effects of vitamin D.
The question of which cutoff value to use is more problematic. Turan et al.1 used a variety of cutoff values to define hypovitaminosis D and found that vitamin D deficiency persisted across different definitions. Their use of different cutoff values highlights the controversy regarding threshold values for defining vitamin D deficiency. This controversy exists in large part because studies about vitamin D focus on different clinical end points (e.g., skeletal disease, muscle function, calcium absorption, parathyroid hormone secretions, cancer). Given the alterations in endocrine function that can occur with surgery or critical illness, it is also reasonable to question whether there should be different thresholds for vitamin D deficiency in these patient groups; studies to date have not addressed this issue.
The Institute of Medicine Food and Nutrition Board (FNB) concluded that bone health was the only satisfactory outcome criterion for recommendations regarding calcium and vitamin D intakes (as well as the adequacy of 25OHD levels as a marker of vitamin D exposure).8,9 The FNB’s reasoning was that bone health was the only outcome indicator for which there was sufficient evidence for both causality and a dose–response with vitamin D. The FNB felt that there was a lack of convincing data to link vitamin D status to nonskeletal outcomes such as cardiovascular disease, death, immune functions, and quality of life. In 2010, the cut points for serum 25OHD concentrations for bone health outcomes were established by the FNB committee. Serum 25OHD levels <12 ng/mL (<30 nmol/L) are considered vitamin D deficiency, whereas serum 25OHD levels of 12 to 20 ng/mL (30–50 nmol/L) are considered inadequate for optimal bone and overall health (i.e., vitamin D insufficiency).
The Endocrine Society2 and others have relied upon levels at which calcium absorption decreases or parathyroid hormone increases to set threshold values for 25OHD. These studies suggest that levels of 30 ng/mL or greater are adequate.3 Thus, the Endocrine Society guidelines2 for evaluation, prevention, and treatment of vitamin D deficiency define vitamin D deficiency as a 25OHD level <20 ng/mL (50 nmol/L) and vitamin D insufficiency as 25OHD levels between 21 and 29 ng/mL (53–73 nmol/L). Although there is conflict between the FNB and Endocrine Society guidelines,9,10 it is important to note that the exact levels where deficiency occurs are limited by available data. The unfortunate result can be different diagnostic thresholds. In addition, most data arise from studies performed in relatively healthy individuals, and the exact needs of hospitalized patients with sepsis, surgery, or other critical illnesses remain undefined.
There are many causes of vitamin D deficiency (Table 1).3 Because vitamin D is a fat-soluble compound, absorption of adequate amounts from the diet requires intact biliary secretion and intestinal absorption. Thus, patients with fat intolerance, malabsorption, malnutrition, low incomes (leading to poor diets), or milk intolerance are at risk for vitamin D deficiency (Table 1). After absorption from the intestine (via the lymphatics as a component of chylomicrons) or synthesis in the skin, vitamin D circulates to the liver, where in the endoplasmic reticulum it is 25-hydroxylated to 25-hydroxyvitamin D (25OHD or calcidiol) and then to the kidneys where it is 1-hydroxylated to 1,25-dihydroxyvitamin D (also known as calcitriol) (Fig. 1). Thus, patients with hepatic disease are at risk for vitamin D deficiency (Table 1). Other major causes include decreased skin synthesis of vitamin D (caused by decreased sunlight exposure from use of sunscreens, being house-bound or nursing home-bound, or from living in high latitudes), increased catabolism (drugs such as phenytoin, glucocorticoids, and antiretroviral drugs; hyperthyroidism; primary hyperparathyroidism; granulomatous diseases), nephrotic syndrome (25OHD lost in urine), and vitamin D resistance syndromes. Patients with dementia are at increased risk because of limited sun exposure and poor diets. Obesity is also associated with vitamin D deficiency because of the sequestration of vitamin D in fat tissue.
HYPOVITAMINOSIS D: A CAUSE OR A MARKER OF DISEASE?
Reverse causation bias9,11 makes the interpretation of many observational studies (such as the study by Turan et al.1) problematic. Low 25OHD levels could be the result of disease rather than a cause of disease. When poor health curtails outdoor activities and sunlight exposure, adversely affects diet, or requires specific treatments that impair vitamin D absorption or metabolism, low 25OHD levels may act as a surrogate marker of poor health rather than a marker of impaired cellular functions from deficiency of vitamin D. Moreover, the data presented by Powe et al.12 and others suggest that common genetic polymorphisms in DBP strongly associate with vitamin D concentrations irrespective of the actual levels of DBP. The result was that when levels of both 25OHD and DBP were reduced, the bioavailable vitamin D level could be normal. In other words, reduced 25OHD levels do not uniformly indicate vitamin D deficiency. We can be more confident in our diagnosis of vitamin D deficiency when reduced concentrations of 25OHD are accompanied by clinical manifestations such as hypocalcemia (especially decreased ionized calcium concentrations), increased parathyroid hormone levels, decreased calcitriol levels, decreased calcium absorption, and reduced bone mineral density.
Unlike the use of hemoglobin A1c to monitor diabetic glucose control, there is no measurement that can assess the adequacy of 25OHD concentrations over time. Any single measurement of 25OHD may not reflect the true state of hormonal balance. Measurements of 25OHD may be rendered abnormal as a result of hemodilution, interstitial extravasation (i.e., capillary leak with increased vascular permeability), decreased synthesis of binding proteins, and renal wasting of 25OHD.13 It is also important to note that acute fluctuations in 25OHD concentrations may be seen in critically ill patients (as the result of volume changes and other factors).13 Thus, it can be misleading to make conclusions about the significance of a single 25OHD measurement in such patients.
Levels of 25OHD can decrease as a result of decreased DBP levels. Several investigators have reported reduced DBP in patients with systemic inflammation and tissue injury, such as critically ill, trauma, and surgical patients.14–18 For example, in a study by Dahl et al.,14 levels of DBP measured shortly after admission were significantly lower in multiple trauma patients than in control patients (median 143 mg/L [range 84–215] compared with 334 mg/L [range 265–390]). The authors concluded that the decrease in DBP was related to the increase in DBP–actin complexes formed after actin’s release from injured cells. Van den Berghe et al.16 reported vitamin D levels in 22 surgical patients admitted to the intensive care unit (ICU) (with an anticipated ICU stay of >10 days) and compared results with 22 healthy controls who were well matched by age, sex, and body mass index. DBP levels were significantly lower in the patients with surgical critical illness (mean 248 vs 346 mg/L). Reid et al.18 examined the relationship between acute changes in the systemic inflammatory response and plasma concentrations of vitamin D in 33 patients recovering from knee arthroplasty. Most patients had reduced 25OHD levels and increased parathyroid hormone levels at baseline consistent with vitamin D deficiency. These patients demonstrated a large increase in C-reactive protein concentrations, consistent with inflammation that persisted for at least 5 days after surgery. There was a 40% decrease in 25OHD levels within 12 hours after surgery, accompanied by a 15% reduction in albumin and 10% reduction in DBP. Twenty-five OHD levels did not return to normal 5 days after surgery and remained depressed at a 3-month follow-up. DBP returned to baseline by 5 days after surgery. Patients received a median of 2.5 L of fluid over the first 24 hours after surgery. The investigators hypothesized that most of the decrease in 25OHD levels resulted from the inflammatory response with minor effects from hemodilution and decreased DBP levels. The importance of the studies in this paragraph is that they clearly show that acute changes in 25OHD concentrations after surgery may not reflect vitamin D intake or skin synthesis. Also, it is not clear that the acute changes in 25OHD concentrations that accompany inflammation and tissue injury actually indicate a cellular deficiency in vitamin D.
Similar to the high prevalence of vitamin D deficiency reported by Turan et al.,1 numerous studies of normal populations, as well as studies of patients with acute and chronic disease, indicate a high prevalence of vitamin D deficiency. The exact incidence appears to vary with the population studied and risk factors for vitamin D deficiency (Table 1). Variations in the incidence may also result from differences in assays used to measure 25OHD. The FNB and Endocrine Society estimate that 20% to 100% of US and European elderly men and women living in the community are deficient in vitamin D.2,3 In a recent review, Hossein-nezhad and Holick4 estimate that 20% to 80% of the US, Canadian, and European adult population is deficient in vitamin D! This wide range underscores the difficulty we have in assessing the medical importance of any study that purports to identify a patient population that is claimed to be “particularly at risk” for hypovitaminosis D. The prevalence of 25OHD levels <20 ng/mL (50 nmol/L) is estimated to be approximately 32% in the US population. More than 70% of non-Hispanic African American and >40% of Hispanic/Mexican individuals have levels <20 ng/mL (50 nmol/L). Similarly, one-third of children visiting a Dutch pediatric outpatient department had 25OHD concentrations <12 ng/mL (30 nmol/L).19
The prevalence of hypovitaminosis D was evaluated in adult individuals (N = 15,390) enrolled in the Third National Health and Nutrition Examination Survey.20 Vitamin D deficiency was defined as a 25OHD level <28 ng/mL (70 nmol/L) and severe deficiency as a level <10 ng/mL (25 nmol/L). Mean 25OHD levels were lower in females versus males, elderly (≥60 years of age versus <60), and Hispanics and African Americans versus Caucasians. Overall, vitamin D deficiency occurred in 40% of men and 51% of women; severe deficiency occurred in 1.1% of men and 2.7% of women. Powe et al.12 recently reported vitamin D results from the Healthy Aging in Neighborhoods of Diversity across Life Span (HANDLS) study of African Americans and Caucasians (N = 2085). African Americans had reduced levels of 25OHD (15.6 ± 0.2 vs 25.8 ± 0.4 ng/mL) and DBP (168 ± 3 vs 337 ± 5 μg/mL) compared with Caucasians. Most African American participants (77% using a cutoff value of 20 ng/mL or 50 nmol/L; 96% using a cutoff value of 30 ng/mL or 75 nmol/L) would be classified as deficient in vitamin D.
Thomas et al.21 evaluated vitamin D status of 290 consecutive patients on a general medical ward in Boston, Massachusetts. Fifty-seven percent (N = 164) had vitamin D deficiency (25OHD ≤15 ng/mL or 38 nmol/L). Twenty-two percent (N = 65) had severe vitamin D deficiency (25OHD <8 ng/mL or 20 nmol/L). Kiebzak et al.22 measured 25OHD levels in 100 patients admitted to a general hospital rehabilitation unit. Eleven percent had a 25OHD level <8 ng/mL (20 nmol/L), 75% had levels <24 ng/mL (60 nmol/L), and 94% had levels <32 ng/mL (80 nmol/L). Low 25OHD levels were correlated with lower functional independence scores, reduced grip strength, and increased length of stay.
Jeng et al.23 measured 25OHD and DBP in critically ill patients with (N = 24) and without sepsis (N = 25) and in healthy controls (N = 21). Mean 25OHD levels were 16.0 ± 8.5 ng/mL (40 ± 21 nmol/L) and 16.2 ± 7.2 ng/mL (41 ± 18 nmol/L) in sepsis and nonseptic critically ill patients; levels were 26.0 ± 7.6 ng/mL (65 ± 19 nmol/L) in the healthy control patients. The prevalence of vitamin D insufficiency (25OHD < 30 ng/mL [75 nmol/L]) was 100% in the sepsis patients, 92% in the nonseptic patients, and 66.5% in the healthy control patients. Van den Berghe et al.16 reported vitamin D levels in 22 surgical patients admitted to the ICU and compared the results with those of 22 healthy matched controls. Mean 25OHD levels were 10.9 ± 4.2 ng/mL (27 ± 11 nmol/L) in ICU patients and 20.1 ± 8.9 ng/mL (50 ± 22 nmol/L) in control patients. The reduced 25OHD levels in ICU patients were associated with significantly reduced ionized calcium (1.14 ± 0.09 vs 1.20 ± 0.09 mmol/L, P < 0.008) and calcitriol concentrations. Lee et al.24 assessed 25OHD levels in 42 ICU patients referred to the Department of Endocrinology (only 2 of these referrals were for hypocalcemia). The serum concentrations of 25OHD were 16 ± 9 ng/mL (41 ± 22 nmol/L). Only 3 patients (7%) had levels of 25OHD considered sufficient (>24 ng/mL or >60 nmol/L).
Quraishi et al.5 retrospectively evaluated 25OHD levels and hospital-acquired infections in 770 patients receiving gastric bypass surgery. The overall prevalence of preoperative hypovitaminosis D (25OHD < 30 ng/mL or <75 nmol/L) was 58%, and the overall rate of hospital-acquired infections was 5.3% (surgical site infections 2.6%). Comparing patients with 25OHD levels <30 ng/mL (75 nmol/L) with patients with levels >30 ng/mL (75 nmol/L) resulted in an adjusted odds ratio (OR) of 3.0 (95% confidence interval [CI], 1.3–6.9) for hospital-acquired infections and an adjusted OR of 3.9 (95% CI, 1.1–14.2) for surgical-site infections. There was an inverse relationship between 25OHD levels and the likelihood of nosocomial or surgical-site infections, with rates increasing below a threshold 25OHD level of 50 ng/mL (125 nmol/L).
Low levels of 25OHD have been associated with negative health outcomes in epidemiologic studies.10 These outcomes include skeletal (i.e., decreased bone mineral density, rickets, osteomalacia, fractures) and muscle disorders.2 Muscle weakness is a prominent feature of severe vitamin D deficiency. Several studies have demonstrated a reduction in falls with vitamin D repletion.2 In several studies, low levels of vitamin D were associated with increased risk of mortality25,26; an increased incidence of cardiovascular events, cancer, and infections; and a longer ICU stay.11 Levels of 25OHD were positively associated with levels of cathelicidin (LL-37, an antimicrobial peptide produced by macrophages and neutrophils).23
VITAMIN D LEVELS AND NONSKELETAL HEALTH OUTCOMES
In a meta-analysis of randomized controlled trials, Autier and Gandini26 evaluated the effect of vitamin D supplementation upon all-cause mortality and reported a significant decrease in risk of mortality (OR, 0.93; 95% CI, 0.87–0.99). The primary end points of most of the studies that were included in the meta-analysis were fractures or mineral bone density. Vitamin D supplements were associated with a 1.4- to 5.2-fold increase in 25OHD levels compared with control patients. Melamed et al.25 evaluated the relationship between 25OHD status and mortality (all-cause mortality, cancer mortality, or cardiovascular mortality) using 13,331 noninstitutionalized adults from the Third National Health and Nutrition Examination Survey. Study subjects had 25OHD measured upon entry into the study (1988–1994) and had a mean follow-up of 8.7 years. Using multivariate analysis, the lowest 25OHD quartile (25OHD < 17.8 ng/mL or <45 nmol/L) had a 26% increased rate of all-cause mortality compared with the highest quartile (>32.1 ng/mL or >80 nmol/L). Zittermann et al.27 performed a meta-analysis of prospective cohort studies (N = 62,548) to evaluate the association between 25OHD levels and mortality. For the highest compared with the lowest categories of 25OHD, the estimated relative risk for mortality was 0.71 (95% CI, 0.50–0.91). The median baseline 25OHD value was approximately 11 ng/mL (28 nmol/L). Increases in 25OHD levels of 5 (12.5 nmol/L), 10 (25 nmol/L), and 20 ng/mL (50 nmol/L) were associated with significant decreases in mortality (relative risk of 0.86, 0.77, and 0.69, respectively).
Autier et al.11 recently performed a systematic review of the association between 25OHD levels and nonskeletal health outcomes. They reviewed both prospective observational and interventional studies in adults older than 18 years. Prospective observational studies reported inverse associations between levels of 25OHD and cardiovascular disease, serum lipid concentrations, inflammation, glucose metabolism, weight gain, infectious diseases, multiple sclerosis, mood disorders, cognitive function, physical function, and all-cause mortality. However, in marked contrast, interventional studies failed to show an effect of vitamin D supplementation in reducing disease occurrence. The authors interpret these findings as consistent with low 25OHD arising from ill health rather than being a cause of ill health. Observational research is not sufficient to support the premise that treating a biomarker such as 25OHD would result in improved health. Such claims require evidence from prospective randomized controlled trials.
VITAMIN D SUPPLEMENTATION
We recommend that all patients at risk for vitamin D deficiency (Table 1) be screened by the use of 25OHD levels. In addition to the conditions listed in Table 1, patients with bone disease (i.e., rickets, osteomalacia, moderate-to-severe osteoporosis), nontraumatic fractures, and frequent falls should be screened for vitamin D deficiency. We recommend that patients with documented vitamin D deficiency be treated with vitamin D supplements.
The Endocrine Society2 recommends that adults aged 19 to 70 years at risk for 25OHD deficiency receive 600 IU/d of vitamin D and adults older than 70 years receive 800 IU/d of vitamin D. However, they also note that larger doses (i.e., at least 1500–2000 IU/d of supplemental vitamin D) may be required to increase the blood level of 25OHD >30 ng/mL (75 nmol/L). For adults with documented vitamin D deficiency, recommendations include 50,000 IU of vitamin D weekly or 6000 IU/d for 8 weeks, followed by maintenance therapy of 1500 to 2000 IU/d. Larger maintenance doses (i.e., 3000–6000 IU/d) are recommended for patients with malabsorption or obesity and for those taking medications affecting vitamin D metabolism. The goal is to obtain a 25OHD level >30 ng/mL (70 nmol/L). Although the guidelines fail to mention specifically critically ill and surgical patients, we consider many of these to require the greater doses, and with that proviso, we support the supplementation doses recommended by the Endocrine Society. Nevertheless, we recognize that in a meta-analysis of prospective randomized trials of the effects of vitamin D supplements upon health outcomes (primarily skeletal outcomes), Autier and Gandini26 failed to demonstrate a dose–response to vitamin D supplements (300–2000 IU/d); however, all supplement groups had an increase in 25OHD levels over baseline levels.
Very few studies have evaluated the use of vitamin D replacement or supplementation during acute illness. Of note, most studies in acutely ill and surgical patients have used lower doses of vitamin D and have not increased levels >30 ng/mL (75 nmol/L). Most supplement studies evaluated “healthy” patients over relatively long periods. Most of these studies used low (likely inadequate) doses of vitamin D, failed to individualize dosing based on patient characteristics or medical conditions, and (most importantly) did not monitor 25OHD concentrations during repletion with the goal of obtaining a specified level. One example is the study by Hsia et al.28 who randomized 36,282 postmenopausal women to calcium and vitamin D (400 IU/d) versus placebo in a fracture trial. Patients were followed for a mean of 7 years. This was not an acute population study, 25OHD levels were not reported, and there was a lack of monitoring of levels during the study. The study reported no differences in myocardial infarction, stroke, or death between study groups. One wonders why, given the results that were available before this study was performed, that the investigators or the study sponsors thought that this approach to vitamin D supplementation might improve cardiovascular outcomes.
In summary, the study by Turan et al.1 rediscovers the large prevalence of reduced vitamin D concentrations in surgical populations. The study also provides additional data associating reduced 25OHD levels with poorer outcomes. However, at least 2 questions remain. First, should there be greater surveillance of 25OHD levels? Second, would outcomes be improved if vitamin D supplementation were provided to patients with reduced 25OHD levels? We are reminded that estrogen-replacement therapy was associated with improved cardiovascular outcomes in many observational studies, but when tested in an adequately powered randomized clinical trial (the Women’s Health Initiative), hormone-replacement therapy worsened cardiovascular outcomes. Vitamin D levels may be low for many reasons and may serve as a marker associated with “illness” (Table 1). A cause-and-effect relationship between low 25OHD levels and poor outcomes has not been established. The relevance of vitamin D deficiency and repletion to outcomes remain unknown and is best answered with an interventional study (i.e., prospective randomized double-blind clinical trial). We suggest that such a trial (1) select a relatively homogenous population of patients with vitamin D deficiency that may benefit from vitamin D repletion (e.g., elderly patients requiring surgery for hip fractures), (2) administer enough supplemental vitamin D to obtain blood concentrations of 25OHD associated with improved outcomes (i.e., >30 ng/mL or >75 nmol/L), and (3) monitor 25OHD levels during treatment to guide replacement and maintenance therapy (because doses must be individualized for each study subject). In the meantime, while we await definitive evidence from controlled trials, we recommend screening patients who are at increased risk for vitamin D deficiency. We recommend vitamin D supplementation (using the regimens supported by the Endocrine Society) for patients undergoing major surgery or having critical illness with documented 25OHD levels <30 ng/ml or <75 nmol/L.2
Name: Gary P. Zaloga, MD.
Contribution: This author helped write the manuscript.
Attestation: Gary P. Zaloga approved the final manuscript.
Conflicts of Interest: Gary P. Zaloga is an employee of Baxter Healthcare. Baxter manufactures and/or distributes solutions containing calcium and phosphorus for parenteral nutrition and intravenous hydration and multivitamin preparations.
Name: John F. Butterworth, IV, MD.
Contribution: This author helped write the manuscript.
Attestation: John F. Butterworth, IV, approved the final manuscript.
Conflicts of Interest: The author has no conflicts of interest to declare.
This manuscript was handled by: Sorin J. Brull, MD, FCARCSI (Hon).
1. Turan A, Hesler BD, You J, Saager L, Grady M, Komatsu R, Kurz A, Sessler DI. The association of serum vitamin D concentration with serious complications after noncardiac surgery. Anesth Analg.
2. Holick MF, Binkley NC, Bischoff-Ferrari HA, Gordon CM, Hanley DA, Heaney RP, Murad MH, Weaver CMEndocrine Society. . Evaluation, treatment, and prevention of vitamin D deficiency: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2011;96:1911–30
3. Holick MF. Vitamin D deficiency. N Engl J Med. 2007;357:266–81
4. Hossein-nezhad A, Holick MF. Vitamin D for health: a global perspective. Mayo Clin Proc. 2013;88:720–55
5. Quraishi SA, Bittner EA, Blum L, Hutter MM, Camargo CA. Association between preoperative 25-hydroxyvitamin D level and hospital-acquired infections following roux-en-Y gastric bypass surgery. JAMA Surg. 2014;149:112–8
6. Wang Y, Liu XJ, Robitaille L, Eintracht S, MacNamara E, Hoffer LJ. Effects of vitamin C and vitamin D administration on mood and distress in acutely hospitalized patients. Am J Clin Nutr. 2013;98:705–11
7. Granado Lorencio F, Blanco-Navarro I, Pérez-Sacrsitán B. Critical evaluation of assays for vitamin D status. Curr Opin Clin Nutr Metab Care. 2013;16:734–40
8. Zheng W, Teegarden DZempleni J, Suttie JW, Gregory JF III, Stover PJ. Vitamin D (chapter 2). Handbook of Vitamins. 20145th ed Boca Raton CRC Press:51–88
9. Ross AC, Manson JE, Abrams SA, Aloia JF, Brannon PM, Clinton SK, Durazo-Arvizu RA, Gallagher JC, Gallo RL, Jones G, Kovacs CS, Mayne ST, Rosen CJ, Shapses SA. The 2011 report on dietary reference intakes for calcium and vitamin D from the Institute of Medicine: what clinicians need to know. J Clin Endocrinol Metab. 2011;96:53–8
10. Rosen CJ, Abrams SA, Aloia JF, Brannon PM, Clinton SK, Durazo-Arvizu RA, Gallagher JC, Gallo RL, Jones G, Kovacs CS, Manson JE, Mayne ST, Ross AC, Shapses SA, Taylor CL. IOM committee members respond to Endocrine Society vitamin D guideline. J Clin Endocrinol Metab. 2012;97:1146–52
11. Autier P, Boniol M, Pizot C, Mullie P. Vitamin D status and ill health: a systematic review. Lancet Diabetes Endocrinol. 2014;2:76–89
12. Powe CE, Evans MK, Wenger J, Zonderman AB, Berg AH, Nalls M, Tamez H, Zhang D, Bhan I, Karumanchi SA, Powe NR, Thadhani R. Vitamin D-binding protein and vitamin D status of black Americans and white Americans. N Engl J Med. 2013;369:1991–2000
13. Quraishi SA, Camargo CA Jr. Vitamin D in acute stress and critical illness. Curr Opin Clin Nutr Metab Care. 2012;15:625–34
14. Dahl B, Schiødt FV, Kiaer T, Ott P, Bondesen S, Tygstrup N. Serum Gc-globulin in the early course of multiple trauma. Crit Care Med. 1998;26:285–9
15. Dahl B, Schiødt FV, Ott P, Wians F, Lee WM, Balko J, O’Keefe GE. Plasma concentration of Gc-globulin is associated with organ dysfunction and sepsis after injury. Crit Care Med. 2003;31:152–6
16. Van den Berghe G, Van Roosbroeck D, Vanhove P, Wouters PJ, De Pourcq L, Bouillon R. Bone turnover in prolonged critical illness: effect of vitamin D. J Clin Endocrinol Metab. 2003;88:4623–32
17. Krishnan A, Ochola J, Mundy J, Jones M, Kruger P, Duncan E, Venkatesh B. Acute fluid shifts influence the assessment of serum vitamin D status in critically ill patients. Crit Care. 2010;14:R216
18. Reid D, Toole BJ, Knox S, Talwar D, Harten J, O’Reilly DS, Blackwell S, Kinsella J, McMillan DC, Wallace AM. The relation between acute changes in the systemic inflammatory response and plasma 25-hydroxyvitamin D concentrations after elective knee arthroplasty. Am J Clin Nutr. 2011;93:1006–11
19. Huibers MH, Visser DH, Deckers MM, van Schoor NM, van Furth AM, Wolf BH. Vitamin D deficiency among native Dutch and first- and second-generation non-Western immigrants. Eur J Pediatr. 2014;173:583–8
20. Zadshir A, Tareen N, Pan D, Norris K, Martins D. The prevalence of hypovitaminosis D among US adults: data from the NHANES III. Ethn Dis. 2005;15(suppl 5):97–101
21. Thomas MK, Lloyd-Jones DM, Thadhani RI, Shaw AC, Deraska DJ, Kitch BT, Vamvakas EC, Dick IM, Prince RL, Finkelstein JS. Hypovitaminosis D in medical inpatients. N Engl J Med. 1998;338:777–83
22. Kiebzak GM, Moore NL, Margolis S, Hollis B, Kevorkian CG. Vitamin D status of patients admitted to a hospital rehabilitation unit: relationship to function and progress. Am J Phys Med Rehabil. 2007;86:435–45
23. Jeng L, Yamshchikov AV, Judd SE, Blumberg HM, Martin GS, Ziegler TR, Tangpricha V. Alterations in vitamin D status and anti-microbial peptide levels in patients in the intensive care unit with sepsis. J Transl Med. 2009;7:28
24. Lee P, Eisman JA, Center JR. Vitamin D deficiency in critically ill patients. N Engl J Med. 2009;360:1912–4
25. Melamed ML, Michos ED, Post W, Astor B. 25-hydroxyvitamin D levels and the risk of mortality in the general population. Arch Intern Med. 2008;168:1629–37
26. Autier P, Gandini S. Vitamin D supplementation and total mortality: a meta-analysis of randomized controlled trials. Arch Intern Med. 2007;167:1730–7
27. Zittermann A, Iodice S, Pilz S, Grant WB, Bagnardi V, Gandini S. Vitamin D deficiency and mortality risk in the general population: a meta-analysis of prospective cohort studies. Am J Clin Nutr. 2012;95:91–100
© 2014 International Anesthesia Research Society
28. Hsia J, Heiss G, Ren H, Allison M, Dolan NC, Greenland P, Heckbert SR, Johnson KC, Manson JE, Sidney S, Trevisan MWomen’s Health Initiative Investigators. . Calcium/vitamin D supplementation and cardiovascular events. Circulation. 2007;115:846–54