Secondary Logo

Journal Logo

Does Lactate Affect the Association of Early Hyperglycemia and Multiple Organ Failure in Severely Injured Blunt Trauma Patients?

Richards, Justin E. MD*,†,‡; Scalea, Thomas M. MD, FACS, MCCM§,‡; Mazzeffi, Michael A. MD, MPH*; Rock, Peter MD, MBA, FCCM*; Galvagno, Samuel M. Jr DO, PhD, FCCM*,†,‡

doi: 10.1213/ANE.0000000000002626
Trauma: Original Clinical Research Report
Free
SDC

BACKGROUND: Early hyperglycemia is associated with multiple organ failure (MOF) after traumatic injury; however, few studies have considered the contribution of depth of clinical shock. We hypothesize that when considered simultaneously, glucose and lactate are associated with MOF in severely injured blunt trauma patients.

METHODS: We performed a retrospective investigation at a single tertiary care trauma center. Inclusion criteria were patient age ≥18 years, injury severity score (ISS) >15, blunt mechanism of injury, and an intensive care unit length of stay >48 hours. Patients with a history of diabetes or who did not survive the initial 48 hours were excluded. Demographics, injury severity, and physiologic data were recorded. Blood glucose and lactate values were collected from admission through the initial 24 hours of hospitalization. Multiple metrics of glucose and lactate were calculated: the first glucose (Glucadm, mg/dL) and lactate (Lacadm, mmol/L) at hospital admission, the mean initial 24-hour glucose (Gluc24hMean, mg/dL) and lactate (Lac24hMean, mmol/L), and the time-weighted initial 24-hour glucose (Gluc24hTW) and lactate (Lac24hTW). These metrics were divided into quartiles. The primary outcome was MOF. Separate Cox proportional hazard models were generated to assess the association of each individual glucose and lactate metric on MOF, after controlling for ISS, admission shock index, and disposition to the operating room after hospital admission. We assessed the interaction between glucose and lactate metrics in the multivariable models. Results are reported as hazard ratios (HRs) for an increase in the quartile level of glucose and lactate measurements, with 95% confidence intervals (CIs).

RESULTS: A total of 507 severely injured blunt trauma patients were evaluated. MOF occurred in 46 of 507 (9.1%) patients and was associated with a greater median ISS (33.5, interquartile range [IQR]: 22–41 vs 27, IQR: 21–34; P < .001) and a greater median admission shock index (0.82, IQR: 0.68–1.1 vs 0.73, IQR: 0.60–0.91; P = .02). Patients who were transferred to the operating room after the initial trauma resuscitation were also more likely to develop MOF (20 of 119, 14.4% vs 26 of 369, 7.1%; P = .01). Three separate Cox proportional regression models demonstrated the following HR for an increase in the individual glucose metric quartile and MOF, while controlling for confounding variables: Glucadm HR: 1.35, 95% CI, 1.02–1.80; Gluc24hMean HR: 1.63, 95% CI, 1.14–2.32; Gluc24hTW HR: 1.14, 95% CI, 0.86–1.50. Three separate Cox proportional hazards models also demonstrated the following HR for each individual lactate metric quartile while controlling for the same confounders, with MOF again representing the dependent variable: Lacadm HR: 1.94, 95% CI, 1.38–2.96; Lac24hMean HR: 1.68, 95% CI, 1.22–2.31; Lac24hTW HR: 1.49, 95% CI, 1.10–2.02. When metrics of both glucose and lactate were entered into the same model only lactate remained significantly associated with MOF: Lacadm HR: 1.86, 95% CI, 1.29–2.69, Lac24hMean HR: 1.54, 95% CI, 1.11–2.12, and Lac24hTW HR: 1.48, 95% CI, 1.08–2.01. There was no significant interaction between lactate and glucose variables in relation to the primary outcome.

CONCLUSIONS: When glucose and lactate are considered simultaneously, only lactate remained significantly associated with MOF in severely injured blunt trauma patients.

From the *Department of Anesthesiology, University of Maryland School of Medicine, Baltimore, Maryland

Division of Trauma Anesthesiology

Program in Trauma, R Adams Cowley Shock Trauma Center, Baltimore, Maryland

§Department of Surgery, University of Maryland School of Medicine, Baltimore, Maryland.

Published ahead of print December 26, 2017.

Accepted for publication September 27, 2017.

Funding: None.

The authors declare no conflicts of interest.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website.

Reprints will not be available from the authors.

Address correspondence to Justin E. Richards, MD, Division of Trauma Anesthesiology and Program in Trauma, R Adams Cowley Shock Trauma Center, 22 S Greene St, T1R77, Baltimore, MD 21201. Address e-mail to justin.richards@som.umaryland.edu.

Stress-induced hyperglycemia is a well-established risk factor for multiple organ failure (MOF) and mortality in critically ill trauma patients.1,2 While early hyperglycemia in critical illness is associated with worse outcomes, the parameters of appropriate glucose control have been intensely debated.3–5 Several studies in the trauma literature have suggested that targeted control of serum blood glucose and normalization of hyperglycemia is associated with improved clinical outcomes.2,6 However, results from the Normoglycemia in Intensive Care Evaluation–Survival Using Glucose Algorithm Regulation (NICE-SUGAR) tempered enthusiasm for tight glucose control for critically ill patients, even though a nonsignificant association for improved survival was observed in the subgroup of trauma patients treated with aggressive insulin therapy.7 Results from several trauma studies have generated discussion regarding the use of blood glucose as a readily available marker of injury severity,8–11 although many of these studies have not fully accounted for the depth of clinical shock.

Presently, lactate remains one of the most valuable markers of adequacy of hemodynamic resuscitation and is a marker for the degree of shock in the trauma population.12,13 Lactate is a chemical biomarker that is the byproduct of anaerobic metabolism and represents the depth of shock and end-organ hypoperfusion. Lactate levels rise not only from hypoxia or anaerobic glycolysis but also from accelerated aerobic glycolysis during the stress response.14 Similar to blood glucose, lactate is associated with the development of MOF15 and mortality in critically ill trauma patients.12,13

Despite strong evidence that serum blood glucose and lactate are separate independent risk factors for MOF in critically ill patients, few studies have evaluated both markers simultaneously to assess the association with MOF. Recent evidence from a large retrospective investigation from multiple mixed population intensive care units (ICUs) demonstrated that when considered together, lactate attenuated the relationship of hyperglycemia and mortality and that serum glucose no longer remained a significant predictor of in-hospital death.16 Considering the evidence that both lactate and early hyperglycemia are separately associated with increased morbidity in the trauma population, we performed the present investigation to confirm the correlation between lactate and glucose in critically ill trauma patients and to explore the association of these 2 laboratory values with patient outcomes. We hypothesized that lactate and glucose are positively correlated and significantly associated with the development of MOF in severely injured blunt trauma patients.

Back to Top | Article Outline

METHODS

Patients and Data

Following institutional review board (IRB) approval (University of Maryland, Baltimore, MD: IRB #00066309), we performed a retrospective cohort study at a single academic, tertiary care trauma center. Due to the retrospective nature of the investigation, informed patient consent was waived by the IRB. Patients were identified for the study using our institution’s trauma registry. This registry is maintained by dedicated trauma nursing personnel and includes demographic data, injury mechanism, medical comorbidities, physiologic data, disposition after hospital admission, duration of ICU, and hospital length of stay (LOS). In addition, the registry captures laboratory data during the first 24 hours of hospitalization. Inclusion criteria for the study were admission between January 1, 2010 and December 31, 2014, age ≥18 years, an injury severity score (ISS) >15, blunt traumatic mechanism of injury, and ICU admission. Patients with a penetrating mechanism of injury were excluded. Considering recent evidence that has demonstrated worse outcomes when traumatic brain injury occurs concomitantly with hyperglycemia,17,18 we only included patients with an abbreviated injury scale score of the head <3, similar to previous investigations.1,2 Furthermore, we also limited inclusion to patients who were without a known preinjury history of diabetes mellitus, consistent with prior studies examining the association of hyperglycemia in trauma populations.6,8,11,19 We do not routinely collect hemoglobin A1c values on all trauma admissions and therefore were unable to document patients with undiagnosed diabetes mellitus.

The following variables were collected from the registry for all patients: ISS, the revised trauma score, the Glasgow Coma Scale (GCS) score, disposition after admission (ie, to operating room or ICU), and the shock index (SI). The admission SI was calculated as the heart rate (beats per minute) divided by the systolic blood pressure (mm Hg) from hospital admission, and has been described as a marker of injury severity in the trauma literature and is associated with both transfusion requirements and worse outcomes.20,21

Serum glucose (mg/dL) and lactate (mmol/L) values collected during the first 24 hours of admission were also recorded. The source of glucose values was documented from either whole blood samples or bedside glucometer measurements; all lactate values were obtained from whole blood samples. Hypoglycemia was defined as a blood glucose level <60 mg/dL. Multiple metrics of glucose and lactate were calculated and recorded: the first glucose (Glucadm) and lactate (Lacadm) at hospital admission, the mean 24-hour after hospital admission glucose (Gluc24hMean) and lactate (Lac24hrMean), and the time-weighted 24-hour after hospital admission glucose (Gluc24hTW) and lactate (Lac24hTW). Time-weighted measurements are utilized to avoid surveillance bias. Surveillance bias is a nonrandom collection of data and is central to the idea that the more frequently an event is surveyed, the more likely it will be that the event is identified.22 Time-weighted measurements also correct for irregular sampling intervals. These measurements are assessed by first calculating the time difference of 2 consecutive data points. The mean laboratory value of consecutive data points is then multiplied by the total amount of time between these 2 consecutive measurements. This value is divided by the total amount of elapsed time during which laboratory values were obtained. In theory, time-weighted measurements represent a more accurate description of varying data points gathered over irregular time periods, have been specifically used to avoid surveillance bias, and have demonstrated valuable applications in retrospective analysis of glucose and lactate data.16,23 In summary, time-weighted measurements provide a more quantifiable estimate of consecutive values over a defined period of time that is not captured by simply calculating the mean value. To capture time-weighted values, patients were required to have recorded at least 2 glucose values and 2 lactate values during the 24-hour period after hospital admission.

The primary outcome of interest for this study was MOF, defined by the Denver MOF score.24 This definition of MOF has been validated and compared against other MOF scores and demonstrated greater specificity and relative simplicity.25,26 Organ failure was assigned based on daily evaluation of cardiac, hepatic, renal, and pulmonary organ system dysfunction (ie, grades 0, 1, 2, or 3), with higher scores indicating more severe dysfunction. The daily summation of the highest scores for each organ system was calculated from the medical record and MOF was documented if the daily summation of the individual organ systems was >3 and occurred more than 48 hours after hospital admission. Therefore, patients who survived <48 hours were excluded from the final analysis. Time to MOF was assessed in days from admission until the first documented summation of individual organ system scores >3.

Back to Top | Article Outline

Statistical Analysis

Descriptive statistics were utilized to evaluate the final study population. Variables of demographics, injury characteristics, and admission physiologic data were compared with the primary outcome of interest. Categorical variables were analyzed with the χ2 statistic and represented by frequency (n) and percent (%). Continuous variables were analyzed for distribution and evaluated with analysis of variance testing for normally distributed data and are presented as the mean and standard deviation (SD). Nonparametrically distributed data were analyzed with the Kruskal-Wallis test and presented as the median and interquartile range (IQR).

The 6 metrics of glucose (Glucadm, Gluc24hMean, Gluc24hTW) and lactate (Lacadm, Lac24hMean, Lac24hTW) were described as continuous variables. Linear correlation between similar metrics of glucose and lactate were assessed with Spearman’s correlation coefficient. Glucose and lactate metrics were then divided separately into quartiles and the association between each metric and MOF was assessed using individual Cox proportional hazards models. Covariates were assessed by univariate analysis if the P value <0.1 and were considered to represent clinical significance. Patients who died or who were discharged from the hospital without the primary outcome event were censored, thus adjusting for the effect of early death on subsequent MOF. Separate proportional hazards models were described for each calculated glucose and lactate metric in the following sequence, with MOF as the dependent variable: glucose + covariates, lactate + covariates, glucose + lactate + covariates, and glucose + lactate + covariates + interaction variable between glucose and lactate. Furthermore, log–log survival plots were examined to assess whether the proportional hazards assumption was met for each model. Collinearity was assessed with the variance inflation factor.

Results are reported as hazard ratios (HR) for each increase in the quartile level of glucose and lactate measurements, with 95% confidence intervals (CIs). A P value of <.05 was considered statistically significant. An a priori sample size calculation was also performed. Based on previous data and assuming that any metric of lactate measurement would confer a HR of 1.5 for MOF, with a median lactate of 4.0 mmol/L (IQR: 3.0–5.0 mmol/L) in patients with MOF compared to a median lactate of 2.0 mmol/L (IQR: 1.0–3.0 mmol/L) in those without MOF, and that the correlation of lactate and glucose demonstrated a correlation coefficient of 0.33, a population sample size of 64 patients with MOF would be required to achieve a power of 80% at an α of .05. Anticipating a 15% incidence of MOF, we would ultimately require a final population of 426 patients. All analyses were performed with Stata software version 12.1 (Stata Corp, College Station, TX).

Back to Top | Article Outline

RESULTS

Five hundred and eighty-nine patients fulfilled inclusion criteria. The Figure outlines the flow diagram of inclusion and exclusion criteria for the final study population, which consisted of 507 severely injured blunt trauma patients. The population was predominantly male (308 of 507, 60.7%), with a mean age of 45.1 (SD: 17.6) years. The median ISS was 27 (IQR: 21–34); and the median GCS score was 15 (IQR: 14–15). The overall median ICU LOS was 8.47 (IQR: 4.8–16.0) days, and 24 of 507 (4.7%) patients died before hospital discharge.

Figure.

Figure.

Table 1.

Table 1.

MOF occurred in 46 of 507 (9.1%) patients. The median time to MOF was 6.5 days (IQR: 3–12 days). Patients with a greater median ISS (33.5, IQR: 22–41 vs 27, IQR: 21–34; P < .001) and a greater median admission SI (0.82, IQR: 0.68–1.1 vs 0.73, IQR: 0.60–0.91; P = .02) were more likely to develop MOF. Patients who were transferred to the operating room after hospital admission were also more likely to develop MOF (20 of 119, 14.4% vs 26/369, 7.1%; P = .01). There were no significant differences in age, gender, mechanism of injury, revised trauma score, or admission GCS and the development of MOF (Table 1). MOF was significantly associated with mortality (11 of 46, 23.9% vs 13/461, 2.8%; P < .001).

Back to Top | Article Outline

Glucose

Table 2.

Table 2.

A total of 2324 glucose values were collected with a mean of 4.6 (SD: 1.9) values per patient in the initial 24 hours of hospital admission. The median Glucadm was 147 mg/dL (IQR: 122–181 mg/dL), the median Gluc24hMean was 136 mg/dL (IQR: 123.3–152.3 mg/dL), and the median Gluc24hTW was 133.8 mg/dL (IQR: 118.6–140.5 mg/dL). A total of 3 occurrences of hypoglycemia were documented in 3 separate patients, and was not associated with MOF (P > .05) or mortality (P > .05). The cutoff values for the quartiles of each glucose metric are presented in Supplemental Digital Content, Appendix 1, http://links.lww.com/AA/C130. Separate Cox proportional hazards models demonstrated the following HRs, indicating the risk of MOF for an increase in quartile level for each glucose metric (Table 2). For example, if true, these results suggest that a severely injured blunt trauma patient with an admission glucose of 140 mg/dL (quartile 2) has a 35% increased risk of MOF compared to a similar blunt trauma patient with an admission glucose of 100 mg/dL (quartile 1).

Back to Top | Article Outline

Lactate

Table 3.

Table 3.

A total of 2319 lactate values were collected from the study population and each patient had a mean of 4.6 (SD: 1.9) values collected within the initial 24 hours of hospital admission. The median Lacadm was 3.4 mmol/L (IQR: 2.4–4.8 mmol/L), the median Lac24hMean was 2.7 mmol/L (IQR: 2.0–3.5 mmol/L), and the median Lac24hTW was 2.6 mmol/L (IQR: 1.9–3.3 mmol/L). The cutoff values for the quartiles of each lactate metric are presented in Supplemental Digital Content, Appendix 2, http://links.lww.com/AA/C130. Separate Cox proportional hazards models demonstrated the following HRs for each increase in quartile level of lactate metrics, with MOF again representing the dependent variable (Table 3). If true, this suggests that a severely injured blunt trauma patient with an admission lactate of 3.0 mmol/L (quartile 2) has a 94% increase in risk of MOF compared to a similar blunt trauma patient with an admission lactate of 1.0 mmol/L (quartile 1).

Back to Top | Article Outline

Contribution of Both Glucose and Lactate

Table 4.

Table 4.

Admission glucose was positively and significantly correlated with admission lactate (rs= 0.49, 95% CI, 0.41–0.56; P < .001). Similarly, Gluc24hMean and Lac24hMean were positively and significantly correlated (rs= 0.42, 95% CI, 0.35–0.50; P < .001), as well as Gluc24hTW and Lac24hTW (rs= 0.33, 95% CI, 0.26–0.42; P < .001). Cox proportional hazards models were evaluated to include the contribution of both glucose and lactate with the subsequent development of MOF. These models reflected that when simultaneously considering glucose and lactate, only the metrics of lactate demonstrated a significant association with MOF (Table 4). For example, utilizing this model and considering both glucose and lactate measurements, a severely injured blunt trauma patient with a 24-hour time-weighted lactate of 3.0 mmol/L (quartile 3) would have a 48% increased risk of MOF compared to a similar trauma patient with a 24-hour time-weighted lactate of 2.0 mmol/L (quartile 2). Interaction terms between each metric of glucose and lactate were generated and included in the hazards model; however, none of these interaction terms demonstrated any evidence of statistically significant interaction. Evaluation of the proportionality assumption assessed that for each final hazards model, the test of proportional assumption was confirmed. There was no evidence of collinearity in any of the final models.

Back to Top | Article Outline

DISCUSSION

Our present study evaluated the contribution of multiple metrics of both glucose and lactate to MOF in severely injured blunt trauma patients. Results from this investigation are similar to a recent study from Kaukonen et al,16 which demonstrated that when considered simultaneously, lactate modified the relationship of stress–hyperglycemia and clinical outcomes in critically ill patients. Such data have prompted a considerable discussion of prior literature regarding hyperglycemia and critical illness.

Since the landmark findings in a study by van den Berghe et al3 few topics in the critical care literature have generated as much attention and controversy as stress–hyperglycemia. Numerous subsequent investigations documented the deleterious effects of hyperglycemia, such as greater risk of wound infections,27 prolonged ICU LOS,28 increased risk of MOF,1 and mortality.29 Perhaps even more interesting is the fact that stress–hyperglycemia portends a worse clinical prognosis in patients without a history of diabetes. Kerby et al30 noted that posttraumatic hyperglycemia occurred in nearly 25% of trauma patients and that nondiabetic, hyperglycemic patients had the greatest adjusted risk of mortality. It appears evident that blood glucose is a valuable marker of injury severity and may help identify patients at increased risk of resource utilization and poor clinical outcomes.

Lactate represents a byproduct of anaerobic metabolism and its accumulation in patients who have sustained severe trauma corresponds to an increasing degree of hemorrhagic shock. Serum lactate reflects the magnitude of injury severity and adequacy of end-organ perfusion such that the ability to correct elevated lactate to normal levels is associated with improved survival in the critically injured trauma population.12,13 Lactate is also a valuable triage tool to predict massive transfusion requirements.31 Despite the large amount of evidence in favor of lactate as a marker of hemodynamic resuscitation, there exists disparate opinion regarding how serum lactate should be utilized and interpreted in clinical care. Numerous studies in the trauma literature have noted that a single lactate value at admission is predictive of adverse outcomes.31,32 However, other investigations have demonstrated that multiple lactate measurements and the change in lactate over a period of time are valuable indices of resuscitation.12,13

The results of our investigation confirm that glucose and lactate are significantly and positively correlated after traumatic injury, similar to prior investigations by Duane et al.10,33 Furthermore, we identified that when considered simultaneously, lactate remained associated with worse clinical outcomes while glucose no longer demonstrated a significant relationship, which was consistent with prior literature. Green et al34 suggested in a retrospective cohort of patients in the emergency department that while hyperglycemia was associated with increased mortality, this association disappeared when lactate was considered in a multivariable model. A potential explanation for these findings is that lactate and glucose share a common metabolic pathway as a function of carbohydrate metabolism. Increased levels of glucose may generate lactate through glycolysis and, conversely, lactate may generate glucose via the Cori cycle. It is possible that in periods of increased stress response, such as traumatic injury, hyperglycemia is a reflection of not only a catecholamine surge and altered gluconeogenesis but also increased lactate production. While numerous investigations have separately documented the deleterious association of hyperglycemia and lactate, our data suggest that when considered in the same prediction model lactate retains a significant association with MOF, while glucose does not, in severely injured blunt trauma patients.

The clinical significance of this study lies within the utility and ability to interpret readily available laboratory markers of injury severity. For example, it is a common practice in our state-wide trauma system to obtain prehospital blood glucose measurements. It is possible that these values may then assist in the triage of severely injured nondiabetic, blunt trauma patients who may require intense resource utilization and are at increased risk for MOF. While the role of prehospital lactate has been described previously,32 it is not prevalent in all trauma systems and is presently the focus of an ongoing investigation at our institution. Furthermore, initial prehospital glucose measurements may also prompt the receiving facility to obtain lactate values on hospital admission, such that these values may then be followed throughout the initial trauma resuscitation to identify patients at risk of subsequent MOF. Considering more than one-quarter of the patients in our cohort required acute operative intervention after admission, it may also be argued for the perioperative trauma physician and acute care anesthesiologist that in severely injured blunt trauma patients attention to elevated lactate values and hemodynamic resuscitation is preferred over aggressive insulin therapy targeting glucose normalization. However, it is critical to acknowledge that our study was not designed to answer such specific questions and that this idea is speculative and should be the focus of further investigations.

Strengths of this study include a large cohort of blunt trauma patients from a mature and well-developed trauma system. We chose MOF a priori as the primary outcome because this is a well-documented occurrence in critically ill trauma patients and is one of the leading causes of subsequent mortality after severe injury.15,25,26 However, there are certain limitations to this retrospective investigation, which contains inherent bias. Due to the retrospective nature, the documentation of MOF was dependent on available laboratory data. We evaluated the admission and mean values of glucose and lactate as these are easily interpreted and have demonstrated significance with regard to clinical outcomes.1,9 We also evaluated time-weighted values to minimize selection bias; however, the real-time bedside clinical utility of time-weighted measurements is not without limitations.29 Our population was restricted to severely injured, blunt trauma patients, and therefore the results may not translate to all critically ill patients. Patients with previously diagnosed diabetes were excluded to specifically identify those patients with stress-induced hyperglycemia. Because hemoglobin A1C values are not routinely obtained at our institution, it is possible that patients with undiagnosed diabetes were included in the study cohort, as is a limitation in previous studies on hyperglycemia in trauma patients.1,2,6,8,10,19 We were also unable to collect certain data points, such as dose of insulin administered in response to hyperglycemia, and therefore we are unable to provide a recommendation on the optimal level of glycemic control in this critically ill trauma population. This may explain why Glucose24hTW was not associated with the primary outcome. Previous literature in the trauma population has noted that while admission glucose is predictive of MOF and mortality, this relationship ceased to exist at 24 hours after admission, presumably due to improved glucose control.2,33 Of note, the depth of shock was considered in only one of these studies, which suggested that lactate was superior to glucose as a predictor of mortality at 24 hours after admission.33 Prior investigations have also noted that insulin therapy should not commence until after the initial resuscitation is complete.35 However, it is of importance to mention that our study was not designed to address questions regarding the aggressiveness of insulin therapy or glycemic control.

In addition, volume and type of crystalloid, colloid, and transfused blood products could not be obtained from our trauma database. At our institution, it is standard practice to transfuse severely injured patients with a balanced ratio of red blood cells, fresh frozen plasma, and platelets.36 It may be considered that lactate represents a surrogate marker of hemorrhagic shock and therefore the degree of resuscitation and required volume transfusion. Nonetheless, it is important to recognize that the intent of our investigation was not to identify all independent risk factors for MOF in critically ill trauma patients, but rather to explore the simultaneous contribution and association of glucose and lactate to posttraumatic MOF. It is also crucial to mention that we do not intend to minimize the role of glucose or stress–hyperglycemia in the critically ill trauma patient. Our data suggest that effects of early hyperglycemia (ie, within 24 hours of hospital admission) on MOF may be affected by early clinical shock. We did not evaluate glucose levels after the initial 24 hours of hospital admission. Persistent hyperglycemia remains a well-established marker in critically ill patients and acute changes in glucose, as well as prolonged glucose variability, are all associated with worse clinical outcomes.29

Our present investigation provides further evidence that glucose and lactate demonstrate a collinear relationship and represent valuable markers of injury severity that may serve to aid in the initial triage and prognostic implications for severely injured blunt trauma patients. While certain metrics of glucose during the initial 24 hours after admission were separately associated with MOF, this relationship no longer remained significant when similar measures of lactate were considered in the same statistical model. To our knowledge, the present study is one of the few investigations to evaluate the association of both early glucose and lactate with MOF. These findings contribute to the growing body of literature in critically ill patients that suggest the relationship of stress–hyperglycemia and MOF is potentially influenced by the degree of clinical shock. While such conclusions are hypothesis generating, future prospective studies are necessary to further delineate the role of hyperglycemia and ongoing hemodynamic resuscitation in severely injured blunt trauma patients.

Back to Top | Article Outline

DISCLOSURES

Name: Justin E. Richards, MD.

Contribution: This author helped with study concept and design, data collection, data analysis, and manuscript preparation.

Name: Thomas M. Scalea, MD, FACS, MCCM.

Contribution: This author helped analyze the data, and prepare and review the manuscript.

Name: Michael A. Mazzeffi, MD, MPH.

Contribution: This author helped analyze the data, and prepare and review the manuscript.

Name: Peter Rock, MD, MBA, FCCM.

Contribution: This author helped prepare and review the manuscript.

Name: Samuel M. Galvagno Jr, DO, PhD, FCCM.

Contribution: This author helped design the study, analyze the data, prepare and review the manuscript.

This manuscript was handled by: Richard P. Dutton, MD.

Back to Top | Article Outline

REFERENCES

1. Sperry JL, Frankel HL, Vanek SL, et al. Early hyperglycemia predicts multiple organ failure and mortality but not infection. J Trauma. 2007;63:487–493.
2. Sperry JL, Frankel HL, Nathens AB, et al.; Inflammation and the Host Response to Injury Investigators. Characterization of persistent hyperglycemia: what does it mean postinjury? J Trauma. 2009;66:1076–1082.
3. van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in critically ill patients. N Engl J Med. 2001;345:1359–1367.
4. Wiener RS, Wiener DC, Larson RJ. Benefits and risks of tight glucose control in critically ill adults: a meta-analysis. JAMA. 2008;300:933–944.
5. Jacobi J, Bircher N, Krinsley J, et al. Guidelines for the use of an insulin infusion for the management of hyperglycemia in critically ill patients. Crit Care Med. 2012;40:3251–3276.
6. Scalea TM, Bochicchio GV, Bochicchio KM, Johnson SB, Joshi M, Pyle A. Tight glycemic control in critically injured trauma patients. Ann Surg. 2007;246:605–10.
7. Finfer S, Chittock DR, Su SY, et al. Intensive versus conventional glucose control in critically ill patients. N Engl J Med. 2009;360:1283–1297.
8. Laird AM, Miller PR, Kilgo PD, Meredith JW, Chang MC. Relationship of early hyperglycemia to mortality in trauma patients. J Trauma. 2004;56:1058–1062.
9. Yendamuri S, Fulda GJ, Tinkoff GH. Admission hyperglycemia as a prognostic indicator in trauma. J Trauma. 2003;55:33–38.
10. Duane TM, Dechert T, Dalesio N, et al. Is blood sugar the next lactate? Am Surg. 2006;72:613–617.
11. Paladino L, Subramanian RA, Nabors S, Bhardwaj S, Sinert R. Triage hyperglycemia as a prognostic indicator of major trauma. J Trauma. 2010;69:41–45.
12. Abramson D, Scalea TM, Hitchcock R, Trooskin SZ, Henry SM, Greenspan J. Lactate clearance and survival following injury. J Trauma. 1993;35:584–588.
13. Régnier MA, Raux M, Le Manach Y, et al. Prognostic significance of blood lactate and lactate clearance in trauma patients. Anesthesiology. 2012;117:1276–1288.
14. Kraut JA, Madias NE. Lactic acidosis. N Engl J Med. 2014;371:2309–2319.
15. Minei JP, Cuschieri J, Sperry J, et al.; Inflammation and the Host Response to Injury Collaborative Research Program. The changing pattern and implications of multiple organ failure after blunt injury with hemorrhagic shock. Crit Care Med. 2012;40:1129–1135.
16. Kaukonen KM, Bailey M, Egi M, et al. Stress hyperlactatemia modifies the relationship between stress hyperglycemia and outcome: a retrospective observational study. Crit Care Med. 2014;42:1379–1385.
17. Bhattacharjee S, Layek A, Maitra S, Sen S, Pal S, Gozi NK. Perioperative glycemic status of adult traumatic brain injury patients undergoing craniotomy: a prospective observational study. J Neurosurg Anesthesiol. 2014;26:313–319.
18. Pecha T, Sharma D, Hoffman NG, Sookplung P, Curry P, Vavilala MS. Hyperglycemia during craniotomy for adult traumatic brain injury. Anesth Analg. 2011;113:336–342.
19. Sung J, Bochicchio GV, Joshi M, Bochicchio K, Tracy K, Scalea TM. Admission hyperglycemia is predictive of outcome in critically ill trauma patients. J Trauma. 2005;59:80–83.
20. Cannon CM, Braxton CC, Kling-Smith M, Mahnken JD, Carlton E, Moncure M. Utility of the shock index in predicting mortality in traumatically injured patients. J Trauma. 2009;67:1426–1430.
21. Mutschler M, Nienaber U, Münzberg M, et al.; TraumaRegister DGU. The Shock Index revisited - a fast guide to transfusion requirement? A retrospective analysis on 21,853 patients derived from the TraumaRegister DGU. Crit Care. 2013;17:R172.
22. Haut ER, Pronovost PJ. Surveillance bias in outcomes reporting. JAMA. 2011;305:2462–2463.
23. Vogelzang M, van der Horst IC, Nijsten MW. Hyperglycaemic index as a tool to assess glucose control: a retrospective study. Crit Care. 2004;8:R122–R127.
24. Sauaia A, Moore EE, Johnson JL, Ciesla DJ, Biffl WL, Banerjee A. Validation of postinjury multiple organ failure scores. Shock. 2009;31:438–447.
25. Dewar DC, Tarrant SM, King KL, Balogh ZJ. Changes in the epidemiology and prediction of multiple-organ failure after injury. J Trauma Acute Care Surg. 2013;74:774–779.
26. Hutchings L, Watkinson P MD, Young JD, Willett K. Defining multiple organ failure after major trauma: A comparison of the Denver, Sequential Organ Failure Assessment and Marshall scoring systems. J Trauma Acute Care Surg. 2017;82:534–541.
27. Jackson RS, Amdur RL, White JC, Macsata RA. Hyperglycemia is associated with increased risk of morbidity and mortality after colectomy for cancer. J Am Coll Surg. 2012;214:68–80.
28. Frisch A, Chandra P, Smiley D, et al. Prevalence and clinical outcome of hyperglycemia in the perioperative period in noncardiac surgery. Diabetes Care. 2010;33:1783–1788.
29. Badawi O, Waite MD, Fuhrman SA, Zuckerman IH. Association between intensive care unit-acquired dysglycemia and in-hospital mortality. Crit Care Med. 2012;40:3180–3188.
30. Kerby JD, Griffin RL, MacLennan P, Rue LW 3rd. Stress-induced hyperglycemia, not diabetic hyperglycemia, is associated with higher mortality in trauma. Ann Surg. 2012;256:446–452.
31. Vandromme MJ, Griffin RL, Weinberg JA, Rue LW 3rd, Kerby JD. Lactate is a better predictor than systolic blood pressure for determining blood requirement and mortality: could prehospital measures improve trauma triage? J Am Coll Surg. 2010;210:861–869.
32. Guyette F, Suffoletto B, Castillo JL, Quintero J, Callaway C, Puyana JC. Prehospital serum lactate as a predictor of outcomes in trauma patients: a retrospective observational study. J Trauma. 2011;70:782–786.
33. Duane TM, Ivatury RR, Dechert T, et al. Blood glucose levels at 24 hours after trauma fails to predict outcomes. J Trauma. 2008;64:1184–1187.
34. Green JP, Berger T, Garg N, et al. Hyperlactatemia affects the association of hyperglycemia with mortality in nondiabetic adults with sepsis. Acad Emerg Med. 2012;19:1268–1275.
35. Mowery NT, Dortch MJ, Dossett LA, et al. Insulin resistance despite tight glucose control is associated with mortality in critically ill surgical patients. J Intensive Care Med. 2009;24:242–251.
36. Holcomb JB, Tilley BC, Baraniuk S, et al.; PROPPR Study Group. Transfusion of plasma, platelets, and red blood cells in a 1:1:1 vs a 1:1:2 ratio and mortality in patients with severe trauma: the PROPPR randomized clinical trial. JAMA. 2015;313:471–482.

Supplemental Digital Content

Back to Top | Article Outline
Copyright © 2017 International Anesthesia Research Society