Abstract
Background:
Thiamine supplementation is recommended for patients with alcohol use disorder (AUD). We hypothesize that critically-ill AUD patients commonly are not given thiamine supplementation.
Objective:
To describe thiamine supplementation incidence in AUD patients with different critical illnesses [alcohol withdrawal, septic shock, traumatic brain injury (TBI), and diabetic ketoacidosis (DKA)] in the USA.
Study design:
Retrospective observational study.
Setting:
Cerner HealthFacts database.
Patients:
Adult ICU patients diagnosed with AUD and admitted with alcohol withdrawal and/or septic shock, TBI, and/or DKA and admitted between 2010–2017.
Measurements:
Incidence and predicted probability of thiamine supplementation in alcohol withdrawal and other critical illnesses.
Results:
The study included 14,998 patients with AUD. Mean age was 52.2±12.6, 77% were male, and in-hospital mortality was 9%. Overall, 7,689 (51%) patients received thiamine supplementation. The incidence of thiamine supplementation for alcohol withdrawal, septic shock, TBI, and DKA, thiamine administration was 59%, 26%, 41% and 24%, respectively. The majority of those receiving thiamine (3957, 52%) received it within 12 hours of ED presentation. The predominant route of thiamine administration was enteral (3119, 41%).
Limitations:
Specific dosing and duration were not completely captured.
Conclusion:
Thiamine supplementation was not provided to almost half of all AUD patients, raising a quality of care issue for this cohort. Supplementation was numerically less frequent in AUD patients with septic shock, DKA, and TBI as compared to those with alcohol withdrawal. These data will be important for the design of quality improvement studies in critically ill patients with AUD.
Keywords: Alcoholism, Alcohol Use Disorder, Alcohol Abuse, Thiamine Deficiency, Thiamine, Critical Illness
Introduction:
Thiamine deficiency can lead to Wernicke-Korsakoff syndrome (WKS) [i.e. spectrum of Wernicke’s encephalopathy and Korsakoff syndrome] and beriberi. Untreated thiamine deficiency has been reported as the cause of death in 20% of patients and irreversible neurological damage in 85% (1–4). The main treatment for WKS is thiamine replacement, which is generally well tolerated and safe (5,6). Alcohol use is the most common risk factor for thiamine deficiency in North America, Europe and Australia (3,6–8), and patients with septic shock, traumatic brain injury (TBI), and diabetic ketoacidosis (DKA) have also been shown to be at risk (9–13). Thus, it is possible that critically-ill patients with alcohol use disorder (AUD) are especially at risk of thiamine deficiency and could benefit from thiamine supplementation.
Many guidelines exist recommending treatment with thiamine in AUD and alcohol withdrawal to prevent WKS (2,5,6,14). However, limited data are available on how many critically-ill patients with AUD actually receive thiamine supplementation. Two previous studies have given different estimates, both of which were relatively low. In one single-center pilot study evaluating the practice pattern of thiamine supplementation in AUD patients presenting with septic shock, only 64% of patients received thiamine (15). The majority (84%) of these patients received the first dose of thiamine in the Intensive Care Unit (ICU), suggesting a potential missed opportunity to administer thiamine earlier in the Emergency Department (ED). A separate study (9) showed that only 21% of AUD patients presenting with TBI received thiamine during their hospital stay. The particularly low rates of thiamine repletion reported in TBI raises the possibility that critical illness (apart from just alcohol withdrawal) may distract clinician attention away from the underlying AUD and need for the provision of thiamine.
In our study, we hypothesized that thiamine supplementation in critically-ill AUD patients in general is low.
Methods:
Study design and data source:
This was a retrospective observational study using the Cerner HealthFacts database, a large, multicenter, and nationally representative dataset in the United States. This database is Health Insurance Portability and Accountability Act compliant and stores de-identified electronic health record data generated from Cerner and non-Cerner participating contributing facilities to facilitate analysis and reporting (16). Encounter-level data (information about encounter type, medications, diagnoses, laboratory orders, and results), hospital characteristics (setting, location, and size), patient characteristics (demographics, comorbidities, and disposition), and year of treatment were abstracted. This study was categorized as exempt by the Institutional Review Board at Beth Israel Deaconess Medical Center given the deidentified nature of the database used for the study.
Patient Population:
We included adult patients (>17 years at the time of admission) who were directly admitted to the ICU from the ED between 2010 and 2017 with ICD-9 and/or ICD-10 codes for AUD or AUD attributable conditions (17) and at least one ICD-9 and/or ICD-10 code for one of the following conditions: alcohol withdrawal, septic shock, TBI, and/or DKA (ICD-9 and ICD-10 codes used can be found in Appendix Table 1). Previous studies have shown that administrative codes (i.e. ICD-9 and ICD-10 codes) have high positive predictive value (PPV) for the diagnosis of AUD (18,19) and we have validated this approach in our own pilot study (15) which found that use of ICD-9 and ICD-10 codes had a PPV of 98.1% when assessed by manual chart review. Patients who had an ICU encounter without a corresponding ED encounter were considered to be direct admissions and were excluded as it was not known if these patients received thiamine at another hospital. Only the index ICU encounter was considered and ICU encounters from subsequent admissions were excluded. Patients were considered to have received thiamine if they received any dose of thiamine via any route of administration i.e. oral/feeding tube or intravenous based on medication codes (Appendix Table 1). Thiamine is sometimes administered jointly with other vitamins in a pre-mixed solution prepared by a pharmacist. However, since this solution is compounded in the pharmacy, individual components, including thiamine, are separately identifiable. Other exclusion criteria were: a) patients for whom the time of ICU admission was not within 48 hours of the time of ED admission or hospital admission; b) patients who had an ED admission time more than 48 hours after hospital admission; c) patients who had a documented time of thiamine administration either more than 48 hours before ED/hospital admission or more than 48 hours after time of hospital discharge. These patients were excluded to prevent bias in either direction.
Patients were grouped by the presence of alcohol withdrawal, septic shock, TBI, or DKA alone or by alcohol withdrawal in combination with each of the other diagnoses, or alcohol withdrawal with more than one other diagnosis (8 categories total) (Figure 1, Appendix Table 2).
Figure 1: Flow Diagram.
AW= alcohol withdrawal, SS= septic shock, TBI= traumatic brain injury, DKA= diabetic ketoacidosis.
Statistical Methods:
Descriptive statistics were used to characterize the study sample. Continuous data were reported as means with standard deviations or medians with interquartile ranges, based on the distribution of the data, and categorical data were reported as counts with percentages.
The outcome of the study was the incidence of thiamine supplementation in different illness categories. The predicted probability of thiamine supplementation in each group, along with 95% confidence intervals, was calculated using generalized estimating equations (GEE), specified with the binomial family of distributions and the logit link function. An exchangeable covariance structure was used to account for clustering of patients within hospitals. The model controlled for age, gender, race, comorbidities, year, number of beds in the hospital, whether the hospital was a teaching or non-teaching hospital, rural/urban status, and geographic census region. Mortality was presented stratified by illness group and thiamine supplementation status.
We described the timing of thiamine administration in each illness group by classifying the time-interval between time of ED admission and the time of thiamine supplementation into 12 hour intervals and determining the counts and percentages of patients in each interval and illness group. Furthermore, the route of thiamine administration, classified into four categories including intravenous, intramuscular, enteral and unknown, is presented.
All analyses were performed using Stata version 15.1 (Statacorp, College Station, TX), and all graphics were generated using the ggplot2 package from the R statistical environment.
Role of the Funding Source:
The study was funded by the National Institute of Health (1R03 AA026093). The funding source had no impact on the study’s design, conduct, and reporting. Additionally, Dr. Donnino is supported, in part, by other NIH funding (5R01DK112886–05 and 5K24HL127101–08).
Results:
Study cohort characteristics:
Over the 8-year study period, 14,998 patients from 133 hospitals met the inclusion criteria (Figure 1). The mean age was 52.2±12.6 of which 77% were male. A majority of the patients (78%; 11,730/14,998) were identified as having alcohol withdrawal, with 68% (10,248) having alcohol withdrawal without any of the other illnesses of interest. For a complete description of patient characteristics see Table 1 and Appendix Table 2.
Table 1:
Demographic and Hospital-level Factors for Alcohol Use Disorder Patients with Exclusive* Critical Illness
| Total N=14,998* (100%) | Alcohol Withdrawal N=10,248 (68%) | Septic Shock N=1,920 (13%) | Traumatic Brain Injury N=539 (4%) | Diabetic Ketoacidosis N=809 (5%) | |
|---|---|---|---|---|---|
| Age (years), mean (SD) | 52 (13) | 51 (12) | 58 (12) | 54 (15) | 47 (13) |
| Male n (%) | 11,590 (77%) | 8,044 (79%) | 1,323 (69%) | 417 (78%) | 629 (78%) |
| Race n (%) | |||||
| Caucasian | 11,020 (74%) | 7,723 (75%) | 1,328 (69%) | 404 (75%) | 481 (60%) |
| African American | 1,762 (12%) | 1,137 (11%) | 245 (13%) | 59 (11%) | 160 (20%) |
| Other | 1,708 (11%) | 1,094 (11%) | 247 (13%) | 54 (10%) | 130 (16%) |
| Past Medical History n (%) | |||||
| Congestive Heart Failure | 1,872 (13%) | 1,066 (10%) | 479 (25%) | 34 (6%) | 50 (6%) |
| Liver Disease | 6,298 (42%) | 3,894 (38%) | 1,388 (72%) | 149 (28%) | 231 (29%) |
| Diabetes | 3,136 (21%) | 1,232 (12%) | 462 (24%) | 71 (13%) | 809 (100%) |
| Chronic Pulmonary Disease | 3,030 (20%) | 1,996 (20%) | 546 (28%) | 84 (16%) | 75 (9%) |
| Lymphoma | 49 (<1%) | 22 (<1%) | 17 (1%) | 3 (1%) | 3 (<1%) |
| Renal Disease | 1,169 (8%) | 483 (5%) | 434 (23%) | 22 (4%) | 74 (9%) |
| Metastatic Disease | 96 (1%) | 48 (1%) | 37 (2%) | 2 (<1%) | 1 (<1%) |
| Type of Teaching Hospital | |||||
| Non-Teaching | 5,402 (36%) | 3,827 (37%) | 700 (37%) | 88 (16%) | 327 (40%) |
| Teaching | 9,161 (61%) | 6,123 (60%) | 1,154 (60%) | 447 (83%) | 458 (57%) |
| Rural/Urban Status n (%) | |||||
| Rural | 3,004 (20%) | 1,946 (19%) | 424 (22%) | 133 (25%) | 157 (19%) |
| Urban | 11,994 (80%) | 8,302 (81%) | 1,496 (78%) | 406 (75%) | 652 (81%) |
| Hospital Beds n (%) | |||||
| <5 | 577 (4%) | 394 (4%) | 64 (3%) | 12 (2%) | 25 (3%) |
| 6–99 | 1,431 (10%) | 1,177 (12%) | 83 (4%) | 8 (2%) | 82 (10%) |
| 100–199 | 2,832 (19%) | 2,036 (20%) | 352 (18%) | 47 (9%) | 165 (20%) |
| 200–299 | 2,857 (19%) | 2,028 (20%) | 379 (20%) | 50 (9%) | 177 (22%) |
| 300–499 | 3,619 (24%) | 2,387 (23%) | 426 (22%) | 239 (44%) | 175 (22%) |
| 500+ | 3,682 (25%) | 2,226 (22%) | 616 (32%) | 183 (34%) | 185 (23%) |
| Geographic Location n (%) | |||||
| Northeast | 3,868 (26%) | 2,920 (29%) | 329 (17%) | 157 (29%) | 113 (14%) |
| South | 4,223 (28%) | 2,800 (27%) | 618 (32%) | 144 (27%) | 262 (32%) |
| Midwest | 1,905 (13%) | 1,373 (13%) | 166 (9%) | 101 (19%) | 113 (14%) |
| West | 5,002 (33%) | 3,155 (31%) | 807 (42%) | 137 (25%) | 321 (40%) |
‘Exclusive’ refers to 4 mutually exclusive conditions (alcohol withdrawal, septic shock, traumatic brain injury and diabetic ketoacidosis) presented in this table, which totals 13,516 patients. The remaining 1,482 patients (not shown in this table, but presented in appendix table 2) have combinations of disease conditions (eg. alcohol withdrawal + septic shock, alcohol withdrawal + DKA etc). The ‘Total’ column refers to the full cohort of 13,516 + 1,482 = 14,998 patients.
Incidence of thiamine supplementation:
Of the total cohort, 7,689 (51%) patients received thiamine. The proportions of patients receiving thiamine administration for a single included illness were: alcohol withdrawal 59% (6,038), septic shock 26% (491), TBI 41% (220), and DKA 24% (195). After multivariate adjustment, the predicted probabilities of thiamine supplementation are: alcohol withdrawal (53% [95% CI: 47% - 59%]), septic shock (26% [95% CI: 21% - 30%]), TBI (30% [95% CI: 23% - 37%]) and DKA (27% [95% CI: 22% - 33%]) (Table 2). The observed incidence and predicted probability with 95% confidence interval of thiamine supplementation respectively in AUD patients with alcohol withdrawal and another critical illness were: for septic shock (50%; 49% [95% CI: 43% - 55%]), for TBI (53%; 48% [95% CI: 40% - 56%]), and for DKA (55%; 58% [95% CI: 50% - 66%]) (Appendix Table 3).
Table 2:
Outcomes in Alcohol Use Disorder Patients with Exclusive Critical Illness:
| Observed Thiamine Supplementation (percent, 95% CI) | Predicted Thiamine Supplementation (percent, 95% CI) | Observed Mortality (percent, 95% CI) | |
|---|---|---|---|
| All* | 51.3 (50.5 – 52.1) | 47.4 (42.1 – 52.6) | 9.3 (8.8 – 9.8) |
| Alcohol withdrawal | 58.9 (58.0 – 59.9) | 53.3 (47.4 – 59.2) | 2.8 (2.5 – 3.1) |
| Septic Shock | 25.6 (23.7 – 27.6) | 25.6 (21.0 – 30.1) | 42.3 (40.1 – 44.5) |
| Traumatic Brain Injury | 40.8 (36.6 – 45.1) | 30.3 (23.4 – 37.2) | 9.1 (6.8 – 11.8) |
| Diabetic Ketoacidosis | 24.1 (21.2 – 27.2) | 27.2 (21.5 – 32.9) | 2.3 (1.4 – 3.6) |
The ‘All’ row here refers to the entire cohort of 14,998 patients.
Thiamine was administered within 12 hours of ED presentation in 52% (3,957) of those administered thiamine. For a complete description of thiamine supplementation timing, see Table 3 and Appendix Table 4. The most common route of thiamine supplementation was enteral: 41% (3,119). See Table 4 for a description of the routes of thiamine supplementation.
Table 3:
Timing of First Thiamine Dose in Alcohol Use Disorder Patients with Exclusive Critical Illness
| Timing of thiamine administration | All*
(n, %) |
Alcohol Withdrawal (n, %) | Septic Shock (n, %) | Traumatic Brain Injury (n, %) | Diabetic Ketoacidosis (n, %) |
|---|---|---|---|---|---|
| ≤ 12 hours | 3957 (51.5) | 3258 (54.0) | 196 (39.9) | 87 (39.6) | 86 (44.1) |
| 12 to ≤24 hours† | 1449 (18.9) | 1098 (18.2) | 113 (23.0) | 56 (25.5) | 44 (22.6) |
| 24 to ≤36 hours | 505 (6.6) | 389 (6.4) | 26 (5.3) | 21 (9.6) | 18 (9.2) |
| 36 to ≤48 hours | 436 (5.7) | 317 (5.3) | 40 (8.2) | 15 (6.8) | 14 (7.2) |
| 48 to ≤54 hours | 113 (1.5) | 82 (1.4) | 5 (1.0) | 5 (2.3) | 7 (3.6) |
| >54 hours | 1229 (16.0) | 894 (14.8) | 111 (22.6) | 36 (16.4) | 26 (13.3) |
| Total | 7689 (100.0) | 6038 (100.0) | 491 (100.0) | 220 (100.0) | 195 (100.0) |
The ‘All’ represents the entire cohort of 14,998 patients i.e. patients from the four exclusive critical illness groups and also patients with more than one critical illness group who received thiamine.
Each hour grouping does not include the lower boundary of the hour grouping
Table 4:
Route of First Dose of Thiamine Administered in Alcohol Use Disorder Patients
| Route | Count† (%) |
|---|---|
| Intravenous | 3075 (40.0) |
| Intramuscular | 443 (5.8) |
| Enteral* | 3119 (40.6) |
| Unknown | 1052 (13.7) |
| Total | 7689 (100.0) |
includes administration by mouth, nasogastric tube, orogastric tube, jejunostomy tube, gastrostomy tube.
Count represents unique patients in the dataset as only the route of the first dose of thiamine for each patient is enumerated.
Mortality:
Overall mortality in the study cohort was 9% (1,395). Mortality was 3% (283) in those with alcohol withdrawal only, 42% (812) if septic shock only, 9% (49) if TBI only, and 2% (19) if DKA only (Table 2). Mortality data for AUD patients with more than one critical illness are provided in Appendix Table 3. The total mortality in patients receiving thiamine was 6% (456), compared with 13% (939) in those not receiving thiamine. See Appendix table 5 for mortality in each illness group stratified by thiamine supplementation status.
Discussion:
In this study of a large cohort of critically ill patients with AUD in the United States, approximately half the patients admitted to the intensive care unit with AUD did not receive thiamine. Patients with AUD who were admitted for septic shock, TBI or DKA but without a diagnosis of alcohol withdrawal had both a numerically lower incidence and a lower predicted probability of receiving thiamine than those who were admitted with alcohol withdrawal.
Alcohol use is a well recognized and important risk factor for thiamine deficiency. AUD is associated with reduction in dietary intake of thiamine and also inhibits intestinal absorption of thiamine, both of which lead to thiamine deficiency (3,20). If this thiamine deficiency is left uncorrected it may contribute to significant morbidity and mortality (1–4). Various guidelines have recognized the importance of prophylactic use of thiamine in preventing Wernicke’s encephalopathy (2,5,6,14). The European Federation of Neurological Societies (EFNS) has recommended prophylactic parenteral administration of thiamine in all subjects at high risk of thiamine deficiency managed in the emergency department (5). The American Society of Addiction Medicine (ASAM) in their alcohol withdrawal management guidelines recommended provision of thiamine for hospitalized patients admitted for alcohol withdrawal or who develop alcohol withdrawal during their hospital course to prevent Wernicke’s encephalopathy (14). Though optimal thiamine administration timing is not established, recommendations are to administer thiamine as soon as possible (6) and preferably intravenously (5) to prevent irreversible brain injury. In addition, thiamine deficiency is seen in septic shock, TBI and DKA (3,6,7,9–13) and thiamine deficiency in these illnesses in patients with underlying AUD might be more pronounced due to increased metabolic need in these conditions. Cancer, prior bariatric surgery, malabsorption due to gastrointestinal disease, parenteral nutrition, starvation and malnutrition are also risk factors for thiamine deficiency, although these patient populations were not the focus of the present study (5). While not the focus of guidelines, an additional concern in critically ill subjects with thiamine deficiency would include cardiovascular and gastrointestinal beriberi which has a presentation similar to other underlying critically ill states such as sepsis and will result in death if unrecognized (21–25).
Limited single-center studies have explored thiamine supplementation in critically ill AUD patients and have shown varying prevalence (9,15,26). In a single Scottish center, thiamine supplementation was observed in 21% of AUD patients admitted to the neurosurgical unit with TBI (9). Conversely, 82% of a cohort with AUD from a single center in the US in whom addiction psychiatry services were involved received thiamine. Our previous work on the practice pattern (26) of thiamine administration in AUD patients with septic shock showed a thiamine supplementation incidence of 64% (15). In that study, there was a significant mortality benefit with thiamine treatment in AUD patients presenting with septic shock (44% mortality in treated vs 79% mortality in untreated, p-value 0.02). Additionally, this study uncovered a potential missed opportunity to administer thiamine in these patients in the ED as only 15% received their first dose of thiamine in the ED. We speculated that septic shock shifted the clinician’s focus away from AUD, explaining to a certain degree the lower than expected incidence of thiamine supplementation. This study, however, was a single center study with a small sample size limiting generalizability. The hypothesis proposed in this study laid the foundation to our current study in a nationwide cohort.
In our current study, just 59% of AUD patients diagnosed only with alcohol withdrawal received thiamine supplementation. This group was most expected to receive thiamine as a result of a more obvious alcohol-related critical illness (14). Of the AUD patients diagnosed only with septic shock, TBI or DKA, 26%, 41% and 24%, respectively, received thiamine, and the adjusted predicted probabilities were similarly low. The finding that AUD patients without overt alcohol withdrawal are numerically less likely to receive thiamine supports our earlier hypothesis that, in the face of other acute critical illnesses, clinicians may be more focused on those critical illnesses than AUD and this lack of focus can lead to failure to provide thiamine supplementation. This hypothesis is further supported by higher incidence of thiamine supplementation in patients with septic shock, TBI and DKA if they were also diagnosed with alcohol withdrawal. Of the patients with AUD who received thiamine, only 40% received it via the recommended intravenous route (5) and only 52% received the first dose within 12 hours of presentation. This suggests that even in patients who receive thiamine, there may be room for improvement in route and timing of administration (6).
In the early 20th century, dietary deficiency was identified as a cause for beriberi and Wernicke-Korsakoff syndrome, and by the mid 20th century, thiamine specifically was identified and started being used in treatment of these diseases (4,5,27). Subsequently, guidelines to supplement thiamine to treat AUD and alcohol withdrawal patients were implemented (2,5,6,14). With the knowledge of the importance of prophylactic use of thiamine in AUD, one could expect a high incidence of thiamine supplementation in the 21st century. However, our paper highlights a serious quality assurance issue in AUD with critical illnesses in the US. Automatic flags and suggestions to administer thiamine in the electronic health record with AUD diagnosis code, having an option of thiamine administration in ED/ICU order-set while admitting AUD patients and pharmacy flags for AUD patients are some of the measures that could help improve thiamine supplementation (28,29).
Our study has some limitations. Firstly, there is a potential to misclassify AUD using ICD-9 and ICD-10 codes, leading to overestimation of thiamine underutilization in AUD. However, ICD-9 and ICD-10 codes have been shown to have high PPV to identify AUD (15,18,19), mitigating this bias. Secondly, specific dosing and duration of thiamine supplementation were beyond the scope of the study and were not completely captured in the database. We, however, included patients who received even a single dose of thiamine by both intravenous and/or enteral route and anticipate the true number of patients receiving the recommended high dose intravenous thiamine (5,6,14) to be lower than our reported numbers. Finally, the large sample size, and the multicenter and retrospective nature of our study prevented us from verifying the data by manual chart review.
Conclusion:
In our study, approximately half of patients with AUD and critical illness did not receive thiamine supplementation. Patients with AUD admitted with septic shock, TBI or DKA and without alcohol withdrawal were found to have numerically lower utilization of thiamine in their treatment than patients with alcohol withdrawal. This information highlights a potential area for quality improvement.
Supplementary Material
Acknowledgments
Funding Source: National Institute of Health grant 1R03 AA026093.
Footnotes
Data Sharing Statement:
The study protocol and statistical analyses codes are available on request. The data cannot be shared publicly as proprietary rights belong to Cerner. Please contact mdonnino@bidmc.harvard.edu for details.
Bibliography
- 1.Sinha S, Kataria A, Kolla BP, Thusius N, Loukianova LL. Wernicke Encephalopathy-Clinical Pearls. Mayo Clin Proc. 2019. Jun;94(6):1065–72. [DOI] [PubMed] [Google Scholar]
- 2.Latt N, Dore G. Thiamine in the treatment of Wernicke encephalopathy in patients with alcohol use disorders. Intern Med J. 2014. Sep;44(9):911–5. [DOI] [PubMed] [Google Scholar]
- 3.Polegato BF, Pereira AG, Azevedo PS, Costa NA, Zornoff LAM, Paiva SAR, et al. Role of thiamin in health and disease. Nutr Clin Pract. 2019. Aug;34(4):558–64. [DOI] [PubMed] [Google Scholar]
- 4.Isenberg-Grzeda E, Kutner HE, Nicolson SE. Wernicke-Korsakoff-syndrome: under-recognized and under-treated. Psychosomatics. 2012. Dec;53(6):507–16. [DOI] [PubMed] [Google Scholar]
- 5.Galvin R, Bråthen G, Ivashynka A, Hillbom M, Tanasescu R, Leone MA, et al. EFNS guidelines for diagnosis, therapy and prevention of Wernicke encephalopathy. Eur J Neurol. 2010. Dec;17(12):1408–18. [DOI] [PubMed] [Google Scholar]
- 6.Thomson AD, Cook CCH, Touquet R, Henry JA, Royal College of Physicians, London. The Royal College of Physicians report on alcohol: guidelines for managing Wernicke’s encephalopathy in the accident and Emergency Department. Alcohol Alcohol. 2002. Dec;37(6):513–21. [DOI] [PubMed] [Google Scholar]
- 7.Crook MA, Sriram K. Thiamine deficiency: The importance of recognition and prompt management. Nutrition. 2014. Jul;30(7–8):953–4. [DOI] [PubMed] [Google Scholar]
- 8.The Department of Nutrition for Health and Development of the World Health Organization (WHO). Thiamine deficiencyand its prevention and control in major emergencies [Internet]. [cited 2021 Jul 24]. Available from: https://www.who.int/publications/i/item/WHO-NHD-99.13
- 9.Ferguson RK, Soryal IN, Pentland B. Thiamine deficiency in head injury: a missed insult? Alcohol Alcohol. 1997. Aug;32(4):493–500. [DOI] [PubMed] [Google Scholar]
- 10.Donnino MW, Andersen LW, Chase M, Berg KM, Tidswell M, Giberson T, et al. Randomized, Double-Blind, Placebo-Controlled Trial of Thiamine as a Metabolic Resuscitator in Septic Shock: A Pilot Study. Crit Care Med. 2016. Feb;44(2):360–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Moskowitz A, Graver A, Giberson T, Berg K, Liu X, Uber A, et al. The relationship between lactate and thiamine levels in patients with diabetic ketoacidosis. J Crit Care. 2014. Feb;29(1):182.e5–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.McConachie I, Haskew A. Thiamine status after major trauma. Intensive Care Med. 1988;14(6):628–31. [DOI] [PubMed] [Google Scholar]
- 13.Attaluri P, Castillo A, Edriss H, Nugent K. Thiamine deficiency: an important consideration in critically ill patients. Am J Med Sci. 2018. Jun 21;356(4):382–90. [DOI] [PubMed] [Google Scholar]
- 14.The ASAM clinical practice guideline on alcohol withdrawal management. J Addict Med. 2020;14(3S Suppl 1):1–72. [DOI] [PubMed] [Google Scholar]
- 15.Holmberg MJ, Moskowitz A, Patel PV, Grossestreuer AV, Uber A, Stankovic N, et al. Thiamine in septic shock patients with alcohol use disorders: An observational pilot study. J Crit Care. 2018. Feb;43:61–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Cerner Health Facts | SC CTSI [Internet]. [cited 2021 Mar 11]. Available from: https://sc-ctsi.org/resources/cerner-health-facts
- 17.Alcohol-Related ICD Codes [Internet]. [cited 2020 Jan 31]. Available from: https://nccd.cdc.gov/DPH_ARDI/Info/ICDCodes.aspx
- 18.Kim HM, Smith EG, Stano CM, Ganoczy D, Zivin K, Walters H, et al. Validation of key behaviourally based mental health diagnoses in administrative data: suicide attempt, alcohol abuse, illicit drug abuse and tobacco use. BMC Health Serv Res. 2012. Jan 23;12:18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Quan H, Li B, Saunders LD, Parsons GA, Nilsson CI, Alibhai A, et al. Assessing validity of ICD-9-CM and ICD-10 administrative data in recording clinical conditions in a unique dually coded database. Health Serv Res. 2008. Aug;43(4):1424–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Subramanya SB, Subramanian VS, Said HM. Chronic alcohol consumption and intestinal thiamin absorption: effects on physiological and molecular parameters of the uptake process. Am J Physiol Gastrointest Liver Physiol. 2010. Jul;299(1):G23–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Moskowitz A, Donnino MW. Thiamine (vitamin B1) in septic shock: a targeted therapy. J Thorac Dis. 2020. Feb;12(Suppl 1):S78–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Saya RP, Baikunje S, Prakash PS, Subramanyam K, Patil V. Clinical correlates and outcome of shoshin beriberi. N Am J Med Sci. 2012. Oct;4(10):503–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Dabar G, Harmouche C, Habr B, Riachi M, Jaber B. Shoshin Beriberi in Critically-Ill patients: case series. Nutr J. 2015. May 17;14:51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.van Stigt RM, van der Wal G, Kamphuis S, Braber A. Beriberi in the ICU: remarkable shock reversal with thiamine. Netharlands Journal of Critical Care. 2019. Mar;2. [Google Scholar]
- 25.Lei Y, Zheng M-H, Huang W, Zhang J, Lu Y. Wet beriberi with multiple organ failure remarkably reversed by thiamine administration: A case report and literature review. Medicine (Baltimore). 2018. Mar;97(9):e0010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Isenberg-Grzeda E, Chabon B, Nicolson SE. Prescribing thiamine to inpatients with alcohol use disorders: how well are we doing? J Addict Med. 2014. Feb;8(1):1–5. [DOI] [PubMed] [Google Scholar]
- 27.Lanska DJ. Chapter 30: historical aspects of the major neurological vitamin deficiency disorders: the water-soluble B vitamins. Handb Clin Neurol. 2010;95:445–76. [DOI] [PubMed] [Google Scholar]
- 28.Day GS, Ladak S, Del Campo CM. Improving thiamine prescribing at an academic hospital network using the computerized provider order entry system: a cohort study. CMAJ Open. 2020. Jun;8(2):E383–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Wai JM, Aloezos C, Mowrey WB, Baron SW, Cregin R, Forman HL. Using clinical decision support through the electronic medical record to increase prescribing of high-dose parenteral thiamine in hospitalized patients with alcohol use disorder. J Subst Abuse Treat. 2019. Apr;99:117–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.