Comments Abstract
Rationale: Acute kidney injury (AKI) is common in patients with sepsis and has been associated with high mortality rates. The provision of thiamine to patients with sepsis may reduce the incidence and severity of sepsis-related AKI and thereby prevent renal failure requiring renal replacement therapy (RRT).
Objectives: To test the hypothesis that thiamine supplementation mitigates kidney injury in septic shock.
Methods: This was a secondary analysis of a single-center, randomized, double-blind trial comparing thiamine to placebo in patients with septic shock. Renal function, need for RRT, timing of hemodialysis catheter placement, and timing of RRT initiation were abstracted. The baseline creatinine and worst creatinine values between 3 and 24 hours, 24 and 48 hours, and 48 and 72 hours were likewise abstracted.
Results: There were 70 patients eligible for analysis after excluding 10 patients in whom hemodialysis was initiated before study drug administration. Baseline serum creatinine in the thiamine group was 1.2 mg/dl (interquartile range, 0.8–2.5) as compared with 1.8 mg/dl (interquartile range, 1.3–2.7) in the placebo group (P = 0.3). After initiation of the study drug, more patients in the placebo group than in the thiamine group were started on RRT (eight [21%] vs. one [3%]; P = 0.04). In the repeated measures analysis adjusting for the baseline creatinine level, the worst creatinine levels were higher in the placebo group than in the thiamine group (P = 0.05).
Conclusions: In this post hoc analysis of a randomized controlled trial, patients with septic shock randomized to receive thiamine had lower serum creatinine levels and a lower rate of progression to RRT than patients randomized to placebo. These findings should be considered hypothesis generating and can be used as a foundation for further, prospective investigation in this area.
Acute kidney injury (AKI) is common in patients with sepsis and has been associated with higher mortality rates (1, 2). The pathophysiological understanding of kidney injury in sepsis has traditionally focused on renal hypoperfusion from cytokine-mediated vasodilation, ultimately resulting in acute tubular necrosis and renal failure (3, 4). However, recent studies have challenged this paradigm, illustrating that sepsis-associated kidney injury often occurs in the face of adequate perfusion (5, 6). These findings suggest that alternative pathophysiologic mechanisms, such as redistribution of renal blood flow, venous congestion, microcirculatory failure, cytopathic hypoxia, and the release of proinflammatory cytokines, may have a role in developing AKI (7). Recent studies also suggest that G1 cell cycle arrest of renal tubular epithelial cells might play an important part in the development of sepsis-induced AKI (8, 9). Unfortunately, there are no presently approved pharmacologic agents for either the prevention or treatment of sepsis-related AKI.
One potentially modifiable etiology of renal injury in sepsis may be mitochondrial dysfunction in which cells are unable to extract and/or use oxygen for aerobic metabolism even if adequate oxygen delivery is present (10). Beriberi secondary to thiamine deficiency is a well-known cause of vasodilatory shock and inadequate cellular oxygen extraction in the face of high cardiac output/adequate perfusion, which has previously been associated with renal failure (11–14).
Thiamine is a key factor in aerobic metabolism, working as a cofactor for pyruvate dehydrogenase (15, 16). In the absence of thiamine, pyruvate is unable to enter the Krebs cycle, and pyruvate is converted to lactate rather than acetyl-coenzyme A. Thiamine deficiency, therefore, causes a shift in metabolism to the anaerobic pathway, resulting in elevated serum lactate levels, cellular apoptosis, organ injury (including renal failure), and possibly death (8, 17, 18). As thiamine deficiency appears to be relatively common in critical illness, and has been previously associated with lactic acidosis and hypotension, thiamine supplementation has emerged as an attractive pharmacologic means of enhancing mitochondrial function in sepsis (19, 20). To date, however, the effect of thiamine supplementation on the prevention or treatment of sepsis-related AKI has not been studied.
In the present study, we hypothesized that the provision of thiamine to patients with septic shock may reduce the incidence and severity of sepsis-related AKI and thereby prevent renal failure requiring renal replacement therapy. To test this hypothesis, we performed a post hoc analysis of a prospective randomized trial of thiamine in septic shock.
This was a secondary analysis of a randomized, double-blind trial comparing the administration of intravenous thiamine to placebo in patients with septic shock (20). Adult patients presenting with sepsis (defined as the presence of two or more systemic inflammatory response syndrome criteria with documented or suspected infection), lactate greater than 3 mmol/L, and hypotension (systolic pressure < 90 mm Hg) after a minimum of a 2-L fluid bolus followed by vasopressor dependence were included in the study. Patients were randomized in a 1:1 ratio to thiamine (200 mg in 50 ml 5% dextrose) or placebo (50 ml 5% dextrose) twice daily for 7 days. Thiamine levels were measured in plasma via liquid chromatography/tandem mass spectrometry by Quest Diagnostics (Nichols Institute, Chantilly, VA). Absolute thiamine deficiency was determined using a previously established standard laboratory reference range from Quest Diagnostics; specifically, absolute thiamine deficiency was defined as a level less than or equal to 7 nmol/L. Study protocols, exclusion criteria, and main results have been previously published by our group (20).
Clinical records of patients enrolled in the trial at the coordinating site were reviewed. Premorbid medical comorbidities were abstracted by a physician blinded to study group assignment through review of the past medical history section of the clinical records. Renal function, need for renal replacement therapy (RRT), indication for RRT, and timing of RRT initiation were abstracted. Patients who received study drug (thiamine or placebo) after the decision to initiate RRT was made or who received RRT before the current admission were excluded from the analysis. The baseline creatinine and worst creatinine values between 3 and 24 hours, 24 and 48 hours, and 48 and 72 hours were likewise abstracted.
Descriptive data are provided according to treatment group; continuous data as means with SDs or medians with interquartile range (IQR), depending on the distribution of the data. Categorical data are presented as counts and percentages. Between-group comparisons were made with Fisher exact test for categorical data and t tests or Wilcoxon rank sum tests for continuous data, as appropriate. Repeated measures of worst creatinine levels were compared between the groups using repeated measures analysis. Creatinine levels were log-transformed before this analysis. We included time, group (placebo vs. thiamine), and the baseline creatinine level in the model. An unstructured variance-covariance structure was used. If a patient received RRT, creatinine levels were imputed by carrying forward the last known value before initiation of RRT. No imputation was performed for patients who died. All hypothesis tests were two-sided, with a significance level of P < 0.05. Statistical analyses were conducted with the use of SAS software, version 9.4 (SAS Institute Inc., Cary, NC).
Eighty patients were enrolled in the original trial at the coordinating center. RRT was initiated before study drug administration in 10 patients (7 [18%] in the thiamine group and 3 [7%] in the placebo group, P = 0.18), leaving 70 patients for the primary analysis in the current study (Figure 1). Baseline patient characteristics are presented in Table 1.
Figure 1. Flow diagram detailing study population and primary outcome. RRT = renal replacement therapy.
[More] [Minimize]| Characteristic | Thiamine (n = 31) | Placebo (n = 39) | P Value |
|---|---|---|---|
| Demographics | |||
| Age, yr, mean (SD) | 68 (16) | 66 (17) | 0.6 |
| Sex, female, n (%) | 13 (42) | 17 (44) | 0.9 |
| Race, white, n (%) | 26 (84) | 36 (92) | 0.04* |
| BMI,† kg/m2, mean (SD) | 30 (10) | 29 (7) | 0.8 |
| Comorbidities, n (%) | |||
| Coronary artery disease | 4 (13) | 9 (23) | 0.3 |
| Congestive heart failure | 5 (16) | 11 (28) | 0.2 |
| Hypertension | 15 (48) | 18 (46) | 0.9 |
| Chronic pulmonary disease | 8 (26) | 12 (31) | 0.7 |
| Diabetes | 11 (35) | 6 (15) | 0.05* |
| Insulin dependent | 3 (10) | 5 (13) | |
| Renal disease | 3 (10) | 6 (15) | 0.5 |
| Charlson Comorbidity Index, median (IQR) | 2 (1–3) | 2 (1–5) | 0.3 |
| Laboratory values at enrollment, median (IQR) | |||
| White blood count, ×103‡ | 13.9 (7.3–21.8) | 13.1 (4.6–19.5) | 0.6 |
| Hemoglobin, g/dl‡ | 10.3 (8.5–13.3) | 10.7 (9.3–12.6) | 0.5 |
| Creatinine, mg/dl | 1.2 (0.8–2.5) | 1.8 (1.3–2.7) | 0.3 |
| GFR, ml/min/1.7 m2 | 62 (29.6–91.8) | 42.1 (27.3–54.7) | 0.3 |
| KDIGO stage 4 or 5, n (%) | 8 (25.8) | 12 (30.8) | 0.7 |
| Glucose, mg/dl§ | 146 (95–190) | 146 (128–193) | 0.5 |
| Lactate, mmol/L | 4.1 (2.6–4.6) | 3.8 (3.0–5.5) | 0.9 |
| Potassium, mEq/L§ | 4.3 (3.8–4.8) | 4.4 (3.9–4.9) | 0.7 |
| Bicarbonate, mEq/L§ | 22 (17–23.0) | 19 (17–22) | 0.5 |
| Blood urea nitrogen, mg/dl† | 33.5 (20.5–55.5) | 31 (21–47.5) | 0.9 |
| Thiamine deficient, n (%) | 12 (38.7) | 11 (28.2) | 0.4 |
| Mechanical ventilation and severity of illness | |||
| Mechanical ventilation at time of enrollment, n (%) | 23 (74) | 25 (64) | 0.4 |
| APACHE II score at enrollment, mean (SD) | 24.9 (9.3) | 25.7 (9.5) | 0.7 |
| SOFA score at enrollment, mean (SD) | 9.3 (3.3) | 9.9 (3.7) | 0.4 |
After initiation of the study drug, more patients in the placebo group than in the thiamine group were started on RRT (eight [21%] vs. one [3%]; P = 0.04). The primary indication for RRT initiation was acidosis in six cases (66.7%), including in the one patient who had been randomized to the thiamine group. Of the remaining cases, the indication for RRT initiation was volume overload in one case (11.1%) and uremia in two cases (22.2%). Median time from study drug administration to RRT initiation was 26 hours (interquartile range, 11–78 hr). In the repeated measures analysis adjusting for the baseline creatinine level, the worst creatinine levels were higher in the placebo group than in the thiamine group (P = 0.05; Figure 2).
Of the 70 patients included in this study, in-hospital mortality was 37.1%, with rates of 32.2% and 41.0% in the thiamine and placebo groups, respectively (P = 0.45). Among the patients who received RRT, 6 (66.7%) expired as compared with 20 (32.8%) who did not receive RRT (P = 0.05). In an exploratory analysis of thiamine-deficient patients, 2 of 11 patients in the placebo group ultimately required RRT, as compared with 0 of 12 patients in the thiamine group (P = 0.22).
In this post hoc analysis of a prospective randomized trial, we found that patients receiving thiamine had less progression to renal failure requiring renal replacement therapy than those in the placebo arm. In addition, we found that patients receiving thiamine had lower creatinine levels than patients in the placebo group.
Sepsis-associated kidney injury is common. Recent data from the Protocolized Care for Early Septic Shock (ProCESS) trial suggest that the incidence of AKI in patients with septic shock is in excess of 30% (1). Although the definition of AKI differs somewhat between studies, other investigators have demonstrated even higher incidence rates (21). Furthermore, the development of renal injury has been shown to have a strong association with poor outcome (1, 2).
Although the previously accepted pathophysiologic explanation for organ failure in sepsis was predicated largely on cytokine-mediated vasodilation and related hypoperfusion, other mechanisms of organ injury in sepsis have recently been proposed (22, 23). In a systematic review of kidney histopathology in septic kidney injury, a minority (22%) had microscopic features of acute tubular necrosis (24). A similar pattern was found in animal models of sepsis-related AKI, where just 184 of 1,059 animals (17.4%) were found to have acute tubular necrosis. Of those animals who suffered acute tubular necrosis, all had low cardiac output and decreased renal blood flow. In contrast, the more common histopathological pattern appeared to be one of tubular cell apoptosis (25).
In the present study, we identified a higher rate of kidney injury and need for renal replacement therapy in patients with septic shock who did not receive thiamine than in those who received thiamine. These data, in combination with the literature reviewed above, suggest that mechanisms other than renal hypoperfusion may have contributed to the increased rate of kidney injury and renal failure in the placebo arm of this randomized trial. As thiamine deficiency has been tied to increased apoptosis in neurons (26), vascular endothelium (27), retinal pericytes (27), and cardiac myocytes (28, 29), we hypothesize that thiamine supplementation may have prevented apoptosis-related cell death in renal tubular cells.
Our study has a number of limitations. Foremost, given the post hoc nature of our study design and small sample size (which increases the statistical fragility and chances of type I error), the results should be considered hypothesis-generating and not evidence of a causal link between thiamine deficiency and sepsis-related kidney injury. In addition, there was an imbalance in baseline renal function and medical comorbidities between patients in the control and intervention arms (Table 1), which may limit internal validity. Although these differences did not reach statistical significance, the lack of significance may be due to the small number of patients in the study and resultant low power to detect a difference. Finally, in patients who received RRT, post-RRT creatinine levels were imputed by carrying forward the last known value before initiation of RRT. Although this should bias toward the null (true renal function had likely worsened and not remained constant leading up to RRT initiation), these missing data are a limitation of our post hoc design.
In this post hoc analysis of a randomized controlled trial, patients with septic shock randomized to receive intravenous thiamine had lower serum creatinine levels and a lower rate of progression to RRT than patients randomized to placebo treatment. Given an emerging understanding of sepsis-related AKI that focuses increasingly on cytopathic hypoxia and tubular apoptosis, our results raise the possibility of a renal-protective role of thiamine supplementation. Due to the small sample size of this study and post hoc design, the results presented here should be considered hypothesis generating and future, prospective animal and human studies are needed to better clarify the role of thiamine in sepsis-related renal dysfunction.
| 1 . | Kellum JA, Chawla LS, Keener C, Singbartl K, Palevsky PM, Pike FL, Yealy DM, Huang DT, Angus DC; ProCESS and ProGReSS-AKI Investigators. The effects of alternative resuscitation strategies on acute kidney injury in patients with septic shock. Am J Respir Crit Care Med 2016;193:281–287. |
| 2 . | Schrier RW, Wang W. Acute renal failure and sepsis. N Engl J Med 2004;351:159–169. |
| 3 . | Landry DW, Oliver JA. The pathogenesis of vasodilatory shock. N Engl J Med 2001;345:588–595. |
| 4 . | Honore PM, Jacobs R, De Waele E, Diltoer M, Spapen HD. Renal blood flow and acute kidney injury in septic shock: an arduous conflict that smolders intrarenally? Kidney Int 2016;90:22–24. |
| 5 . | Legrand M, Dupuis C, Simon C, Gayat E, Mateo J, Lukaszewicz AC, Payen D. Association between systemic hemodynamics and septic acute kidney injury in critically ill patients: a retrospective observational study. Crit Care 2013;17:R278. |
| 6 . | Maiden MJ, Otto S, Brealey JK, Finnis ME, Chapman MJ, Kuchel TR, Nash CH, Edwards J, Bellomo R. Structure and function of the kidney in septic shock: a prospective controlled experimental study. Am J Respir Crit Care Med 2016;194:692–700. |
| 7 . | Lankadeva YR, Kosaka J, Evans RG, Bailey SR, Bellomo R, May CN. Intrarenal and urinary oxygenation during norepinephrine resuscitation in ovine septic acute kidney injury. Kidney Int 2016;90:100–108. |
| 8 . | Zarbock A, Gomez H, Kellum JA. Sepsis-induced acute kidney injury revisited: pathophysiology, prevention and future therapies. Curr Opin Crit Care 2014;20:588–595. |
| 9 . | Kashani K, Al-Khafaji A, Ardiles T, Artigas A, Bagshaw SM, Bell M, Bihorac A, Birkhahn R, Cely CM, Chawla LS, et al. Discovery and validation of cell cycle arrest biomarkers in human acute kidney injury. Crit Care 2013;17:R25. |
| 10 . | Fink MP. Cytopathic hypoxia: mitochondrial dysfunction as mechanism contributing to organ dysfunction in sepsis. Crit Care Clin 2001;17:219–237. |
| 11 . | Stratta P, Canavese C, Triolo G, Segoloni G, Pacitti A, Salomone M, Mangiarotti G, Vercellone A. Acute renal failure in fulminating beriberi. Int J Artif Organs 1986;9:443–444. |
| 12 . | Loma-Osorio P, Peñafiel P, Doltra A, Sionis A, Bosch X. Shoshin beriberi mimicking a high-risk non-ST-segment elevation acute coronary syndrome with cardiogenic shock: when the arteries are not guilty. J Emerg Med 2011;41:e73–e77. |
| 13 . | Anderson SH, Charles TJ, Nicol AD. Thiamine deficiency at a district general hospital: report of five cases. Q J Med 1985;55:15–32. |
| 14 . | Mallat J, Lemyze M, Thevenin D. Do not forget to give thiamine to your septic shock patient! J Thorac Dis 2016;8:1062–1066. |
| 15 . | Frank RA, Leeper FJ, Luisi BF. Structure, mechanism and catalytic duality of thiamine-dependent enzymes. Cell Mol Life Sci 2007;64:892–905. |
| 16 . | Manzanares W, Hardy G. Thiamine supplementation in the critically ill. Curr Opin Clin Nutr Metab Care 2011;14:610–617. |
| 17 . | Donnino MW, Miller J, Goyal N, Loomba M, Sankey SS, Dolcourt B, Sherwin R, Otero R, Wira C. Effective lactate clearance is associated with improved outcome in post-cardiac arrest patients. Resuscitation 2007;75:229–234. |
| 18 . | Costa NA, Gut AL, de Souza Dorna M, Pimentel JA, Cozzolino SM, Azevedo PS, Fernandes AA, Zornoff LA, de Paiva SA, Minicucci MF. Serum thiamine concentration and oxidative stress as predictors of mortality in patients with septic shock. J Crit Care 2014;29:249–252. |
| 19 . | Costa NA, Azevedo PS, Polegato BF, Zornoff LA, Paiva SA, Minicucci MF. Thiamine as a metabolic resuscitator in septic shock: one size does not fit all. J Thorac Dis 2016;8:E471–E472. |
| 20 . | Donnino MW, Andersen LW, Chase M, Berg KM, Tidswell M, Giberson T, Wolfe R, Moskowitz A, Smithline H, Ngo L, et al.; Center for Resuscitation Science Research Group. Randomized, double-blind, placebo-controlled trial of thiamine as a metabolic resuscitator in septic shock: a pilot study. Crit Care Med 2016;44:360–367. |
| 21 . | Rangel-Frausto MS, Pittet D, Costigan M, Hwang T, Davis CS, Wenzel RP. The natural history of the systemic inflammatory response syndrome (SIRS): a prospective study. JAMA 1995;273:117–123. |
| 22 . | Gomez H, Ince C, De Backer D, Pickkers P, Payen D, Hotchkiss J, Kellum JA. A unified theory of sepsis-induced acute kidney injury: inflammation, microcirculatory dysfunction, bioenergetics, and the tubular cell adaptation to injury. Shock 2014;41:3–11. |
| 23 . | Swaminathan S, Rosner MH, Okusa MD. Emerging therapeutic targets of sepsis-associated acute kidney injury. Semin Nephrol 2015;35:38–54. |
| 24 . | Langenberg C, Bagshaw SM, May CN, Bellomo R. The histopathology of septic acute kidney injury: a systematic review. Crit Care 2008;12:R38. |
| 25 . | Kosaka J, Lankadeva YR, May CN, Bellomo R. Histopathology of septic acute kidney injury: a systematic review of experimental data. Crit Care Med 2016;44:e897–e903. |
| 26 . | Wang JJ, Hua Z, Fentress HM, Singleton CK. JNK1 is inactivated during thiamine deficiency-induced apoptosis in human neuroblastoma cells. J Nutr Biochem 2000;11:208–215. |
| 27 . | Beltramo E, Berrone E, Buttiglieri S, Porta M. Thiamine and benfotiamine prevent increased apoptosis in endothelial cells and pericytes cultured in high glucose. Diabetes Metab Res Rev 2004;20:330–336. |
| 28 . | Gioda CR, de Oliveira Barreto T, Prímola-Gomes TN, de Lima DC, Campos PP, Capettini LdosS, Lauton-Santos S, Vasconcelos AC, Coimbra CC, Lemos VS, et al. Cardiac oxidative stress is involved in heart failure induced by thiamine deprivation in rats. Am J Physiol Heart Circ Physiol 2010;298:H2039–H2045. |
| 29 . | de Andrade JA, Gayer CR, Nogueira NP, Paes MC, Bastos VL, Neto JdaC, Alves SC Jr, Coelho RM, da Cunha MG, Gomes RN, et al. The effect of thiamine deficiency on inflammation, oxidative stress and cellular migration in an experimental model of sepsis. J Inflamm (Lond) 2014;11:11. |
Supported by Harvard Catalyst (The Harvard Clinical and Translational Science Center) National Center for Research Resources and the National Center for Advancing Translational Sciences grant UL1 TR001102; National Institute of Alternative and Complementary Medicine grant 5R21AT005119 (M.W.D.), National Heart, Lung, and Blood Institute grant 5K24HL127101 (M.W.D.); American Heart Association grant 15SDG22420010 (M.N.C.); and National Institutes of Health grant 2T32HL007374-37 (A.M.).
The content is solely the responsibility of the authors and does not necessarily represent the official views of Harvard Catalyst, Harvard University and its affiliated academic healthcare centers, or the National Institutes of Health.
Author Contributions: All authors contributed substantially to the design of the work, data acquisition, and interpretation of the result. L.W.A. performed the statistical analyses. A.M., M.K., and M.W.D. drafted the manuscript. All authors reviewed the manuscript and revised it for intellectual content. All approved the manuscript before submission.
Author disclosures are available with the text of this article at www.atsjournals.org.
