If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Vitamin D deficiency on admission to the emergency department is a mortality predictor for patients with septic shock treated with early protocol-driven resuscitation bundle therapy
Corresponding author at: Won Young Kim, MD, PhD, Department of Emergency Medicine, University of Ulsan College of Medicine, Asan Medical Center, 88 Olympic-ro 43-Gil, Songpa-gu, Seoul 138-736, Republic of Korea.
Vitamin D is an important immune modulator and is associated with susceptibility to infection. However, past studies have reported inconsistent results regarding the association between vitamin D deficiency and mortality in patients with sepsis, and early-stage data regarding septic shock are limited. This study aimed to determine the relationship between vitamin D deficiency on admission to the emergency department (ED) and mortality in patients with septic shock.
Methods
We analyzed prospectively collected data on adult patients with septic shock who were treated with protocol-driven resuscitation bundle therapy in the ED between September 2019 and February 2021. Septic shock was defined by the sepsis-3 definition and vitamin D deficiency was defined as a 25-hydroxyvitamin D <20 ng/ml. The primary outcome was 30-day mortality.
Results
A total of 302 patients were included, 236 (78.1%) patients had vitamin D deficiency; it was significantly higher in non-survivors than in survivors (89.3% vs. 73.9%, P = 0.004). Mortality was higher in vitamin D deficient patients than in non-deficient patients (31.8% vs. 13.6%, P = 0.004). In multivariate analysis, vitamin D deficiency (odds ratio [OR], 2.43; 95 % confidence interval [CI], 1.03–5.74), hyperlactatemia (OR, 3.65; 95 % CI, 1.95–6.83), Sequential Organ Failure Assessment scores (OR, 1.22; 95% CI, 1.09–1.36), and albumin levels (OR, 0.39; 95% CI, 0.21–0.73) were significantly associated with 30-day mortality.
Conclusions
Vitamin D deficiency was prevalent in patients with septic shock visiting the ED and was associated with mortality.
Sepsis is a life-threatening medical condition that results in significant morbidity and mortality. Each year, it affects over 750,000 patients in the United States and causes approximately 250,000 deaths.
Septic shock, occurring in approximately 15% of sepsis cases, is a medical emergency that accounts for 10% of intensive care unit (ICU) admissions and has a mortality rate of > 50%.
The current guidelines for the management of sepsis and septic shock recommend urgent assessment and early application of protocol-driven resuscitation care bundle therapy. The management includes fluid resuscitation, performing blood cultures, and the administration of broad-spectrum antibiotics.
Sepsis resuscitation generally focuses on macro-circulatory failures, such as decreased mean arterial pressure and cardiac output. However, a growing body of evidence suggests that the inability of the cell to consume oxygen may play a crucial role in the pathogenesis of sepsis. And that evidence entitled metabolic resuscitations for mitochondria such as the combination of vitamin C, thiamine, and corticosteroid. Vitamin D deficiency is prevalent in patients with critical illnesses
In addition to its role in regulating calcium and phosphate metabolisms, vitamin D plays a key role in immunity, endothelial functioning, and the optimal functioning of antimicrobial activity. Previous studies have evaluated vitamin D deficiency as a diagnostic or predictive marker for infections.
Vitamin D Supplementation in Influenza and COVID-19 Infections Comment on:“Evidence that Vitamin D Supplementation Could Reduce Risk of Influenza and COVID-19 Infections and Deaths” Nutrients 2020, 12 (4), 988.
Although vitamin D deficiency is prevalent in patients with sepsis, evidence of the association between vitamin D deficiency and sepsis mortality is limited and inconsistent. Considering the role of vitamin D in maintaining the innate and adaptive immune systems
, we hypothesized that vitamin D deficiency is a surrogate marker of disease severity and may be a modifiable risk factor that could be rapidly treated via supplementation. We believe this investigation will result in high yields as, currently, no proven metabolic therapy is available for septic shock treatment. However, the prevalence of vitamin D deficiency and the association between vitamin D levels and mortality needs to be clarified.
Thus, this study aimed to determine the prevalence of vitamin D deficiency and investigate the relationship between vitamin D deficiency on admission to the emergency department (ED) and mortality in patients with septic shock treated with protocol-driven resuscitation bundle therapy.
Methods
Study design and population
This observational, single-center study using prospectively collected septic shock registry data was conducted in the ED of a tertiary referral academic center in Seoul, South Korea, which provides treatment to approximately 120,000 patients per year. This study included all adult patients aged ≥18 years with septic shock who were treated with protocol-driven resuscitation care bundle therapy and had data on 25-hydroxyvitamin D [25(OH)D] levels taken at the time of admission to the ED from September 2019 and February 2021. The requirement of informed consent was waived, and the study design was approved by the appropriate ethics review board of our hospital.
Refractory hypotension was defined as persistent hypotension with systolic blood pressure <90 mmHg or mean arterial pressure <70 mmHg after adequate intravenous fluid challenge (20–30 mL/kg or at least ≧1 L of crystalloid solution administered over 30 min) or as the need for vasopressors after fluid resuscitation. Patients with septic shock were defined as those with refractory hypotension, hyperlactatemia (≧2 mmol/L), and suspected or confirmed infections by the sepsis-3 definition.
Patients who visited the ED multiple times during the study period were included only once. Transfer cases with do-not-resuscitate orders were excluded to eliminate the confounding effect on mortality as much as possible. (Supplementary S1)
Data collection
The septic shock registry retrieved data on demographic and clinical data on age, sex, previous medical history, initial vital signs, severity, laboratory test values on admission, and interventions. Sequential Organ Failure Assessment (SOFA) scores
were calculated based on physiological and laboratory data collected in the ED. Mental status was assessed according to the alert/responsive to voice/responsive to pain/unresponsive scale, and patients with Glasgow coma scale scores <15 were considered to have altered mental statuses.
Serum 25(OH)D levels (reference value: 10.0–29.0 ng/mL) were analyzed using 3 mL of whole blood collected in a serum separating tube with radioimmunoassay (RIA) using Freedom EVO® (Tecan Schweiz AG, Switzerland) and Dream Gamma-10 with 25-OH-Vitamin D total-RIA-CT (DIAsource). Analyses of 25(OH)D levels were performed immediately after blood sample collection on weekdays and within 48 hours on weekends, until which the blood sample was stored in a refrigerator at 2°C–8°C. When the analysis was expected to be delayed for >48 hours, the blood samples were frozen at -20°C. Vitamin D deficiency was defined as 25(OH)D level <20 ng/ml (50 nmol/L) according to the Endocrine Society guidelines.
All patients were followed up for >30 days as the primary study outcome was 30-day mortality.
Statistical analysis
Continuous variables are presented as mean ± standard deviation due to their normal distribution. Categorical variables are presented as numbers and percentages. To compare the difference among the variables, Student's t-test or the Wilcoxon rank-sum test was used for continuous variables and the chi-square test or Fisher's exact test was used for categorical variables. Univariate and multivariate logistic regression analyses were performed to assess the relationships between patient characteristics or laboratory data and the probability of 30-day mortality. Multivariate logistic regression analyses were performed to determine the predictors of 30-day mortality and the values were reported as odds ratios (ORs) of 30-day mortality. The logistic model of the goodness of fit was evaluated using the Hosmer–Lemeshow test. For all tests, P values were two-tailed and values <0.05 were considered statistically significant. All analyses were performed using IBM SPSS Statics for Windows, version 21.0 (IBM Corp., Armonk, NY, USA).
Results
Study population
During the study period, 337 patients with septic shock who had data on serum vitamin D levels were enrolled. After excluding 8 patients with multiple ED visits and 27 patients who were transferred to other hospitals with do-not-resuscitate orders, 302 patients were included. The mean patient age was 66.5 years, and 59.6% of the patients were men. The prevalence of vitamin D deficiency was 78.1%, and the overall 30-day mortality rate was 27.8%.
Baseline characteristics
The baseline characteristics, including demographic, clinical, and laboratory data, of patients in the 30-day survivor and non-survivor groups, are presented in Table 1. There was no significant difference between the two groups with regard to age, sex, and medical history, except for malignancy (61.5% vs. 77.4%, P = 0.009). The body temperature was lower (37.19 vs. 37.74 °C, P = 0.002) and changes in mental status were more common (27.4 % vs. 14.2%, P = 0.007) in the non-survivor group than in the survivor group. There were statistically significant differences between the two groups in the initial laboratory data including hemoglobin levels (10.74 vs. 9.91 g/dl, P = 0.006), platelet levels (169.8 vs. 127.7 × 103/µL, P = 0.003), prothrombin time (PT) (1.37 vs. 1.91 international normalized ratios (INR), P < 0.001), albumin levels (2.65 vs. 2.25 g/dL, P < 0.001), creatinine levels (1.82 vs. 2.38 mg/dL, P = 0.011), and lactate levels (3.35 vs. 6.01 mmol/L, P < 0.001). The mean SOFA score was significantly higher in the non-survivor group than that in the survivor group (10.16 vs. 7.41, P < 0.001).
Table 1Baseline characteristics between 30-day survivors and 30-day non-survivors.
Total (n = 302)
Survivor (n = 218)
Non-survivor (n = 84)
P value
Age (years)
66.50 ± 11.51
66.72 ± 11.54
65.92 ± 11.48
0.585
Male (%)
180 (59.6)
128 (58.7)
52 (61.9)
0.613
Comorbidities (%)
Hypertension
107 (35.4)
78 (35.8)
29 (34.5)
0.838
Diabetes mellitus
84 (27.8)
65 (29.8)
19 (22.6)
0.211
Chronic renal disease
31 (10.3)
23 (10.6)
8 (9.5)
0.792
Coronary artery disease
35 (11.6)
21 (9.6)
14 (16.7)
0.087
Malignancy
199 (65.9)
134 (61.5)
65 (77.4)
0.009
Initial vital sign
SBP (mmHg)
87.12 ± 21.19
87.11 ± 21.39
87.13 ± 20.79
0.995
DBP (mmHg)
56.49 ± 13.23
56.30 ± 12.86
56.99 ± 14.19
0.685
Heart rate (bpm)
109.7 ± 23.65
108.5 ± 23.15
112.7 ± 24.79
0.173
Body temperature (°C)
37.58 ± 1.42
37.74 ± 1.37
37.19 ± 1.46
0.002
Change of mental status (%)
54 (17.9)
31 (14.2)
23 (27.4)
0.007
Vitamin D (ng/mL)
14.44 ± 9.35
15.59 ± 10.08
11.43 ± 6.23
<0.001
Vitamin D Deficiency (%)
236 (78.1)
161 (73.9)
75 (89.3)
0.004
Laboratory data
WBC (x103/μL)
10.90 ± 9.08
10.67 ± 8.07
11.50 ± 11.34
0.541
Hemoglobin (g/dL)
10.51 ± 2.35
10.74 ± 2.34
9.91 ± 2.30
0.006
Platelet (x103/μL)
158.1 ± 109.8
169.8 ± 105.7
127.7 ± 115.0
0.003
PT (INR)
1.52 ± 0.80
1.37 ± 0.48
1.91 ± 1.24
<0.001
Albumin (g/dL)
2.54 ± 0.59
2.65 ± 0.57
2.25 ± 0.53
<0.001
Creatinine (mg/dL)
1.98 ± 1.70
1.82 ± 1.61
2.38 ± 1.85
0.011
CRP (mg/dL)
15.60 ± 10.70
15.45 ± 10.81
15.98 ± 10.43
0.701
Procalcitonin (ng/dL)
27.23 ± 43.23
25.80 ± 40.34
30.94 ± 50.05
0.364
Lactate (mmol/L)
4.09 ± 3.11
3.35 ± 2.53
6.01 ± 3.64
<0.001
SOFA score
8.18 ± 3.24
7.41 ± 2.78
10.16 ± 3.54
<0.001
ICU admission (%)
101 (33.4)
69 (31.7)
32 (38.1)
0.288
Values are expressed as the mean ± standard deviation, the median [interquartile range], or the number (%).
SBP, systolic blood pressure; DBP, diastolic blood pressure; WBC, white blood cells; PT, prothrombin time; INR, international normalized ratio; CRP, c-reactive protein; SOFA, sequential organ failure assessment; ICU, intensive care unit.
The average 25(OH)D level was significantly higher in the survivor group than in the non-survivor group (15.59 ± 10.08 vs. 11.43 ± 6.23 ng/mL, P < 0.001). The prevalence of vitamin D deficiency as a categorical variable was significantly higher in the non-survivor group than in the survivor group (89.3% vs. 73.9%, P = 0.004).
Characteristics of patients with Vitamin D deficiency
The baseline characteristics, including demographic, clinical, and laboratory data, of patients in the vitamin D deficient and non-deficient groups, are summarized in Table 2. Overall, there were no statistically significant differences between the two groups except for age (70.77 vs. 65.31 years, P = 0.001). Among the laboratory test values, only PT (1.36 vs. 1.56 INR, P = 0.010) and serum lactate levels (4.31 vs. 3.34 mmol/L, P = 0.025) were significantly different between the two groups.
Table 2Comparison of baseline characteristics of patients according to vitamin D deficiency and non-deficiency group.
Total (n = 302)
25(OH)D Non-deficiency group (n = 66)
25(OH)D Deficiency group (n = 236)
P value
Age (years)
66.50 ± 11.51
70.77 ± 10.80
65.31 ± 11.44
0.001
Male (%)
180 (59.6)
39 (59.1)
141 (59.7)
0.924
Comorbidities (%)
Hypertension
107 (35.4)
23 (34.8)
84 (35.6)
0.911
Diabetes mellitus
84 (27.8)
17 (25.8)
67 (28.4)
0.67 3
Chronic renal disease
31 (10.3)
6 (9.1)
25 (10.6)
0.722
Coronary artery disease
35 (11.6)
11 (16.7)
24 (10.2)
0.145
Malignancy
199 (65.9)
38 (57.6)
161 (68.2)
0.107
Initial vital sign
SBP (mmHg)
87.12 ± 21.19
89.92 ± 22.52
86.33 ± 2079
0.224
DBP (mmHg)
56.49 ± 13.23
58.09 ± 14.10
56.04 ± 12.97
0.267
Heart rate (bpm)
109.7 ± 23.65
109.9 ± 22.06
109.6 ± 24.12
0.936
Body temperature (°C)
37.58 ± 1.42
37.95 ± 1.42
37.48 ± 1.40
0.017
Change of mental status (%)
54 (17.9)
9 (13.6)
45 (19.1)
0.309
Laboratory data
WBC (x103/μL)
10.90 ± 9.08
9.57 ± 8.37
11.28 ± 9.26 ±
0.178
Hemoglobin (g/dL)
10.51 ± 2.35
10.26 ± 2.25
10.58 ± 2.38
0.317
Platelet (x103/μL)
158.1 ± 109.8
165.9 ± 106.4
155.9 ± 110.8
0.513
PT (INR)
1.52 ± 0.80
1.36 ± 0.42
1.56 ± 0.87
0.010
Albumin (g/dL)
2.54 ± 0.59
2.60 ± 5.11
2.52 ± 0.61
0.291
Creatinine (mg/dL)
1.98 ± 1.70
1.83 ± 1.74
2.02 ± 1.68
0.438
CRP (mg/dL)
15.60 ± 10.70
16.10 ± 10.86
15.46 ± 10.67
0.668
Procalcitonin (ng/dL)
27.23 ± 43.23
31.07 ± 45.98
26.17 ± 42.49
0.426
Lactate (mmol/L)
4.09 ± 3.11
3.34 ± 2.76
4.31 ± 3.18
0.025
30-day mortality (%)
84 (27.8)
9 (13.6)
75 (31.8)
0.004
Values are expressed as the mean ± standard deviation, the median [interquartile range], or number (%).
The variation in the prevalence of vitamin D deficiency and mean 25(OH)D level according to months is shown in Fig. 1. The prevalence of vitamin D deficiency was the lowest (71.7%) in autumn, whereas the prevalence of vitamin D deficiency was above 80.0% in spring, summer, and winter. However, there was no significant difference between the seasons (P = 0.209).
Fig. 1Seasonal variation of the prevalence of vitamin D deficiency and mean 25(OH) D values.
The prevalence of vitamin D deficiency was the lowest (71.7%) in the autumn of September, October, and November, whereas the prevalence of vitamin D deficiency was above 80.0% in other spring, summer, and winter. However, there were no significant seasonal differences (P = 0.209).
Risk factors for 30-day mortality in patients with septic shock
The results of univariate and multivariate logistic regression analyses to explore and identify the risk factors for 30-day mortality in patients with septic shock are presented in Table 3. Variables with P < 0.10 in univariate analysis, such as malignancy, body temperature, hemoglobin levels, platelet counts, creatinine levels, albumin levels, hyperlactatemia, mental changes, SOFA scores, and vitamin D deficiency were considered as candidate variables in multivariable analyses. Variables that overlap in the representation of the same organ function, or have less clinical importance were excluded, malignancy, hemoglobin levels, albumin levels, hyperlactatemia, SOFA score, and vitamin D deficiency were analyzed as adjusted variables in multivariate analyses finally.
Table 3Univariate and multivariate logistic regression analysis for the association of mortality in septic shock.
In multivariate analysis, the following factors were independently associated with 30-day mortality: vitamin D deficiency (OR, 2.43; 95% confidence interval [CI], 1.03–5.74, P = 0.042), albumin levels (OR, 0.39; 95% CI, 0.21–0.73, P = 0.003), hyperlactatemia (OR, 3.65; 95% CI, 1.95–6.83, P < 0.001), and SOFA scores (OR, 1.22; 95% CI, 1.09–1.36, P < 0.001).
Discussion
This study found that vitamin D deficiency was prevalent (78.1%) in patients with septic shock who were treated with protocol-driven resuscitation bundle therapy in the ED and that the prevalence of vitamin D deficiency was higher in non-survivors than in survivors (89.3% vs. 73.9%, P = 0.004). After adjusting for significant variables such as malignancy, hemoglobin levels, albumin levels, hyperlactatemia, and SOFA scores, vitamin D deficiency was found to be an independent predictor of 30-day mortality (OR, 2.43; 95% CI, 1.03–5.74).
Vitamin D deficiency is common in critically ill patients and is associated with mortality.
Increases in pre-hospitalization serum 25 (OH) D concentrations are associated with improved 30-day mortality after hospital admission: a cohort study.
Prevalence of vitamin D deficiency in critically ill patients and its influence on outcome: experience from a tertiary care centre in North India (an observational study).
Vitamin D deficiency-a potential risk factor for sepsis development, correlation with inflammatory markers, SOFA score and higher early mortality risk in sepsis.
The prevalence of vitamin D deficiency (78.1%) in our study was higher than that reported in other studies. The higher prevalence might be due to the application of different definitions. Septic shock defined according to the sepsis-3 definition may enroll more severe patients than that defined according to the sepsis-2 definition. The prevalence of vitamin D deficiency in patients with septic shock defined according to the sepsis-3 definition has been reported for the first time in our study. The sepsis-3 definition was used to make the study population more homogenous in terms of severity and disease entity because sepsis is a heterogenous clinical syndrome
The mean age was significantly lower (65.31 vs. 70.77 years, P = 0.001) in the vitamin D-deficient group than the vitamin D non-deficient group in this study.
In addition, most studies showed notable associations of all severity scoring criteria,
Vitamin D deficiency-a potential risk factor for sepsis development, correlation with inflammatory markers, SOFA score and higher early mortality risk in sepsis.
Therefore, as predicted, higher severity scores and a higher number of comorbidities were associated with vitamin D deficiency. This is supported by our results that higher serum lactate levels and a higher prevalence of active cancer were found in vitamin D-deficient patients than in vitamin D non-deficient patients.
We found that mortality was higher in vitamin D-deficient patients (31.8%) than in non-deficient patients (13.6%). Additionally, vitamin D deficiency was independently associated with increased 30-day mortality. In line with our study, previous studies have reported that vitamin D deficiency was correlated with mortality in patients with sepsis. Amrein et al.
reported that vitamin D deficiency was associated with in-hospital mortality of patients with sepsis and a hazard ratio of 2.05. Another study including 135 patients admitted to the ICU
reported that the mortality rates were higher among patients with vitamin D deficiency than in those without vitamin D deficiency, with an adjusted relative risk of 2.2. Although there are inconsistencies in the conclusions on the association between vitamin D deficiency and mortality, the hypothesis that they are associated is gaining more strength.
However, there are limited data and inconsistent results regarding vitamin D deficiency in patients with septic shock presenting to the ED. Ginde et al.
showed that vitamin D levels can predict 30-day mortality with an area under a receiver operating characteristic curve of 0.701. The present study adds to the current understanding regarding the association between vitamin D deficiency and mortality in patients with septic shock and can be a starting point for more extensive research based on the ED. Moreover, it can provide a basis for early treatment for patients with septic shock who need prompt intervention.
The incidence of sepsis and serum vitamin D levels decrease in the winter season
because both outdoor activities and exposure to ultraviolet B rays needed to synthesize vitamin D are lower in winter. Therefore, the prevalence of vitamin D deficiency was generally higher in winter and lower in summer, similar to the results of this study; however, our study did not show the general distribution and any significant difference between seasons. Marked decreases in outdoor activities due to the COVID-19 pandemic during the study period could have resulted in increased vitamin D deficiencies. Especially, the number of COVID-19 confirmed cases increased dramatically last autumn in Korea, so restrictions on social and outdoor activities had been stronger. This could have contributed to the study results.
With consistent findings on the prevalence of vitamin D deficiency in patients with sepsis or septic shock, many previous reports, along with our study, have provided answers to the question of whether vitamin D deficiency is correlated with mortality. Therefore, our findings need to be applied to providing metabolic resuscitation in patients with septic shock. Several studies have reported that vitamin D supplementation lowered the severity of organ dysfunction
Effect of high-dose vitamin D3 on hospital length of stay in critically ill patients with vitamin D deficiency: the VITdAL-ICU randomized clinical trial.
To date, there have been no randomized controlled trials involving patients in the ED. Early recognition and intervention are being increasingly emphasized for septic shock; moreover, the time lost up to intervention after hospitalization may hamper improving outcomes. Therefore, randomized controlled trials of routine large-dose loading of vitamin D supplementation in the ED can be proposed to improve mortality rates in patients with septic shock as a novel metabolic resuscitation concept.
The limitation of this study was that it was a single-center study that included many patients with local patterns of illness. Moreover, geopolitical and racial differences should also be considered. Furthermore, patients with chronic renal failure, chronic liver disease, and malignancy were included. These comorbidities can affect vitamin D metabolism and serum vitamin D levels and add to the heterogeneity of the study population; hence the possibility of their confounding effect cannot be ruled out. Also, there would be the possibility of selection bias from the time scale (middle of the year) because of the effect of vitamin D levels which has seasonal variation.
Conclusions
The prevalence of vitamin D deficiency was high in patients with septic shock visiting the ED, and vitamin D deficiency was associated with 30-day mortality. Further studies are needed to determine whether a routine supplement of vitamin D in case of septic shock is warranted.
Authors Contributions
Y.W.K. and Y-J.K. conceived and designed the analysis; Y.S.S., S.M.K., and S-I.H. collected the data; J-S.K. and S.M.R contributed data or analysis tools; B.C. performed the analysis; B.C. and W.Y.K. wrote and edited the paper.
Vitamin D Supplementation in Influenza and COVID-19 Infections Comment on:“Evidence that Vitamin D Supplementation Could Reduce Risk of Influenza and COVID-19 Infections and Deaths” Nutrients 2020, 12 (4), 988.
Increases in pre-hospitalization serum 25 (OH) D concentrations are associated with improved 30-day mortality after hospital admission: a cohort study.
Prevalence of vitamin D deficiency in critically ill patients and its influence on outcome: experience from a tertiary care centre in North India (an observational study).
Vitamin D deficiency-a potential risk factor for sepsis development, correlation with inflammatory markers, SOFA score and higher early mortality risk in sepsis.
Effect of high-dose vitamin D3 on hospital length of stay in critically ill patients with vitamin D deficiency: the VITdAL-ICU randomized clinical trial.
00434-7&id=gr1.jpg){kind=link}