Article, Toxicology

Quantitative analysis of high plasma lactate concentration in ED patients after alcohol intake

a b s t r a c t

Introduction: plasma lactate concentration is known to increase after Alcohol intake. However, this increase has rarely been analyzed quantitatively in emergency department (ED) settings. Evaluating plasma lactate elevation in ED patients after alcohol intake is important because it can affect patients’ evaluation based on the plasma lactate level.

Methods: This study analyzed venous lactate concentrations of 196 continuous patients presented to our ED after alcohol intake. The control group comprised 219 successive ED patients without alcohol intake. Patients who had conditions that might induce lactate elevation were excluded from both groups.

Results: Venous lactate concentration was significantly higher in the alcohol intake group (2.83 mmol/L; 95% con- fidence interval, 2.69-2.96 mmol/L) than in the control group (1.65 mmol/L; 95% confidence interval, 1.53- 1.77 mmol/L; P b .05). Lactate concentrations exceeding 3 mmol/L and exceeding 4 mmol/L were found, respec- tively, in 41.8% and 12.2% of the alcohol intake group compared with in 8.7% and 2.3% of the control group (P b .05). Lactate concentrations do not correlate with patients’ level of consciousness. Therefore, a higher plasma ethanol level is apparently unrelated to elevated lactate.

Discussion and conclusion: Analyses show that plasma lactate concentration is significantly higher in ED patients after alcohol intake and to a greater degree than previously reported, even in patients without previously known alcohol-relatED diseases. Emergency department physicians must be careful when interpreting the lactate level of the patients with alcohol intake.

Introduction

Plasma lactate concentration as a triage tool or prognosis predicting tool

Plasma lactate concentration is increasingly regarded as important as a prognostic indicator for emergency department (ED) patients. In patients with infection who present to the ED, venous lactate concentra- tion is known to correlate well with the prognosis [1]. Even slightly higher venous lactate concentration (2.0-3.9 mmol/L) can indicate worse prognosis in this population [2]. Venous lactate concentrations tend to be higher than arterial lactate concentrations, but they are equally reliable quantitative data to predict prognosis. Moreover, venous blood tests have the benefit that they are less invasive [3]. In addition to being useful for patients with infection, plasma lactate concentration testing is expected to be useful at a point of care to help determine a triage level at the entrance to an ED [4,5]. For drug overdose

? Financial support and conflicts of interests: None for all authors.

* Corresponding author at: 7-3-1, Hongo, Bunkyo-ku, Tokyo, 1138655 Japan. Tel.: +81 3 3815 5411.

E-mail address: sonot-icu@h.u-tokyo.ac.jp (T. Sonoo).

patients, plasma lactate concentration is reported at the ED as an inde- pendent prognosis predictor [6].

Alcohol induced elevation of plasma lactate level

Plasma lactate concentration is frequently elevated in ED patients after alcohol intake. Alcohol intake, especially chronic intake, is known to be able to increase plasma lactate concentration slightly by several physiological mechanisms such as elevated NADH/NAD ratio, followed by Ethanol metabolism by alcohol dehydrogenase mainly in the liver, or decreased liver function for conversion of lactate into glucose (gluco- neogenesis) because of chronic alcoholic liver injury [7]. Only 2 reports have described inebriated ED patients’ plasma lactate concentration, but they were conducted more than 20 years ago [8,9]. One report describes that alcoholic liver injury induced by chronic alcoholic intake might contribute to lactate elevation [8]. The other report explains that plasma lactate elevation in many cases is attributable to conditions other than alcohol intake [9]. However, young and healthy patients who present to the ED with simple alcohol intoxication without signs of baseline liver injury or other clinical conditions also frequently present with high plasma lactate concentration. This higher concentration might

http://dx.doi.org/10.1016/j.ajem.2016.01.021

Table 1

Detailed exclusion criteria for the patients

Exclusion criteria Details

Age b 20 y

Severe conditions Any conditions that required or caused intensive care unit admission

Specific diseases Suspected seizure

Cardiac arrest (anytime before or during hospital arrival)

Organ ischemia (acute coronary syndrome, arterial or venous thrombosis, aortic dissection, cerebral infarction)

Hypoxia Oxygen saturation b85% anytime before or during hospital arrival

low blood pressure Systolic blood pressure b80 mm Hg anytime before or during hospital arrival

Hypoglycemia Blood sugar b70 mg/dL

Intoxication Any kind of medication (including metformin), carbon monoxide intoxication other than ethanol

Anemia Hemoglobin b7.0 g/dL

Liver dysfunction Any of the following: total bilirubin N 2.0 mg/dL, AST N 70 IU/dL, ALT N 90 IU/dL, ?-GTP N 100 IU/dL, ALP N 480 IU/dL

Renal dysfunction Acute renal failure with creatinine N 2.0 mg/dL

Abbreviations: AST, aspartate aminotransferase; ALT, alanine aminotransferase; ALP alkaline phosphatase; ?-GTP, ?-glutamyl transpeptidase.

affect patients’ triage, evaluation, or clinical decision at the ED. Analyses of plasma lactate concentration after alcohol intake in healthy volun- teers [10] or analysis of blood chemistry changes that do not include lac- tate after alcohol intake [11] has been reported. However, no report describes a quantitative analysis of plasma lactate concentrations in ED patients after alcohol intake without another condition that might affect the plasma lactate concentration. To settle this clinical question, we conducted a quantitative analysis of plasma lactate concentrations by comparing data of an alcohol-consuming ED patient group to those of a non-alcohol-consuming ED patient group.

Methods

Data of people who had recently consumed alcohol and who were seen by emergency medical staff at our ED were collected from the pa- tient database at our hospital from April 1, 2015, through October 20, 2015. These alcohol-consuming patients were selected by searching the text in the “history of present illness” data field in patient notes and the database using specific words such as ethanol, alcohol, drunk, beer, sho-chu, and wine. Then each case was judged. Recent alcohol in- take was defined as alcohol intake during the prior 6 hours at the time of presentation to the ED. This cohort was designated as the “alcohol group.” A “control group” included people who presented to our ED without recent alcohol intake. Data were collected from August 5, 2015, through October 6, 2015. At our ED, venous blood gas data,

including lactate concentrations, were checked for all patients having any abnormality in vital signs (judged according to the criteria shown in Table 1) or any change in consciousness level compared with their baseline (judged using Japan Coma Scale). First venous gas data of both groups obtained after arrival were analyzed. In addition, objective information was collected from each patient. This information included the patient’s age, sex, final diagnosis, disposition, vital signs (on arrival), blood test results (including venous blood gas; on arrival), and medical history. Patients were also asked about the history of the present illness, especially the presence or absence of alcohol intake, injury, seizure, syncope, and drug overdose.

Although the patient data were collected retrospectively for this study, the data are objective except for reports of “history of present ill- ness.” Regarding “history of present illness,” all patients’ detailed data were stocked in a comprehensive database. electronic health records were searched for specific words. The results were subsequently reviewed for individual cases. Consequently, there might be only slight biases related to collection of the patients’ data and division of patients into groups. In addition, patients who presented with potential condi- tions that can affect plasma lactate concentrations were excluded. The conditions included severe diseases that caused intensive care unit ad- mission, suspected seizure, conditions that caused tissue hypoperfusion, drug intoxication other than ethanol, and chronic live disease. Further- more, people younger than 20 years, the legal drinking age in Japan, were excluded from this study. The patient inclusion and exclusion

Fig. 1. Patient collection algorithm.

Fig. 2. Venous lactate concentration comparison. 2-1, Venous lactate concentrations of all patients analyzed in each group. 2-2, Venous lactate concentrations of patients who underwent liver function tests. 2-3, Venous lactate concentrations of simple alcohol intoxication patients.

algorithm is presented in Fig. 1. Detailed exclusion criteria are shown in Table 1. The retrospective data collection in our ED was approved by the Ethics Committee of The University of Tokyo Department of Medicine.

Results

The numbers of the patients from whom venous lactate concentra- tions were obtained in the Alcohol and Control group were, respective- ly, 242 and 291. Excluded patients based on the defined criteria (Table 1) were, respectively, 46 and 72. Therefore, between-group com- parison of venous lactate concentrations was performed for 196 alcohol group patients and 219 control group patients. Liver function tests were conducted for 94 and 190 patients in the respective groups.

Venous lactate concentrations were significantly higher in the alcohol group (2.83 mmol/L; 95% confidence interval [CI], 2.69-2.96 mmol/L) than in the control group (1.65 mmol/L; 95% CI, 1.53-1.77 mmol/L; Fig. 2-1). Because the liver function test was not performed in all

subjects, analysis of the subgroup of patients who had undergone the liver function test was performed. In the alcohol group, 16 patients were excluded because of liver function abnormality, and lactate concen- trations were higher (4.38 mmol/L; 95% CI, 2.92-5.84 mmol/L) in the ex- cluded patients. However, the correlation between plasma bilirubin or transaminase concentration and venous lactate concentration was poor. The difference attributed to alcohol intake was also significant in this subgroup: 2.80 mmol/L (95% CI, 2.60-3.01 mmol/L) in the alcohol group and 1.66 mmol/L (95% CI, 1.53-1.79 mmol/L) in the control group (Fig. 2-2).

To estimate the effect of pure alcohol intake, simple acute alcohol in- toxication was defined as follows: alcohol group patients who were not admitted to the hospital and were discharged with self-ambulation, with no injury, no pain complaint or shortness of breath, no history of cardiac disease or malignancy, and no condition, as presented in Table 1. This subgroup of simple acute alcohol intoxication consists of 65 cases. For them, venous lactate concentration was significantly

Fig. 3. Share of patients with higher lactate concentrations in each group.

Table 2

Detailed information for individual patients with venous lactate concentration greater than or equal to 4.0 mmol/L

	Case no.	Age (y)/sex	Lactate level (mmol/L)	Medical history	Final diagnosis
Alcohol group	1	38/F	4.0	None	Mandibular injury
	2	70/M	4.0	Thymus tumor, BPH	Syncope
	3	51/M	4.0	None	Acute alcohol intoxication
	4	44/M	4.1	None	Acute alcohol intoxication
	5	76/F	4.1	HTN, DM, DL	Knee bruise
	6	35/M	4.1	Duodenal ulcer	Acute alcohol intoxication
	7	23/M	4.1	None	Syncope
	8	25/F	4.2	None	nonspecific abdominal pain
	9	58/F	4.2	Hepatitis C infection	Nonspecific chest discomfort
	10	61/M	4.3	HTN	Orbital rupture
	11	31/M	4.4	HTN, DL	Acute alcohol intoxication
	12	67/F	4.4	HTN, DL	Head bruise
	13	51/M	4.5	None	Facial bruise
	14	23/M	4.5	None	Acute alcohol intoxication
	15	56/M	4.5	DM, HTN, DL, alcoholism	Nonspecific chest pain
	16	51/M	4.6	HTN	Head bruise
	17	74/M	4.7	AAA, cardiac valvular disease	Nonspecific neck pain
	18	56/M	4.8	Duodenal ulcer, HTN	Head bruise
	19	20/M	5.0	None	Acute alcohol intoxication
	20	40/F	5.4	None	Acute alcohol intoxication
	21	25/M	5.4	None	Acute alcohol intoxication
	22	29/M	5.5	None	Acute alcohol intoxication
	23	24/M	5.9	None	Mandibular bruise
	24	51/M	6.3	None	Syncope, multiple bruises
Control group	1	86/M	4.7	DM, HTN	Head bruise, multiple bruises
	2	22/M	4.8	None	Transient palpitation
	3	35/M	5.3	None	Heatstroke
	4	74/M	5.6	CI, laryngeal cancer	Multiple bruises
	5	65/M	6.8	DL, cataract, glaucoma	Nonspecific chest discomfort

Abbreviations: AAA, Abdominal aortic aneurysm; F, female; M, male; BPH, benign prostate hyperplasia; CI, cerebral infarction; HTN, hypertension; DL, dyslipidemia; DM, diabetes mellitus.

higher than that for the control group (3.04 mmol/L; 95% CI, 2.80- 3.27 mmol/L; Fig. 2-3).

Proportions of patients with venous lactate concentration greater than or equal to 3 mmol/L and greater than or equal to 4 mmol/L were also significantly higher in the alcohol group than in the control group: 41.8% vs 8.7% for >= 3 mmol/L and 12.2% vs 2.3% for >= 4 mmol/L, respectively (Fig. 3). Information about patients with greatly (>=4 mmol/L) elevated venous lactate concentration is presented in Table 2.

However, alcohol group patients’ venous lactate concentrations showed no correlation with their level of consciousness (Fig. 4).

Assuming that serum ethanol concentrations are, to some extent, correlated with the patients’ level of consciousness, this result implies that serum ethanol concentrations have little influence on the plasma lactate elevation.

Discussion

For this analysis, venous lactate concentrations are estimated to have increased 1.7-fold after alcohol intake. This marked elevation seems attributable to alcohol intake. An earlier study conducted by

Fig. 4. Venous lactate concentration according to consciousness level in alcohol group.

Fulop et al [8] found a similar share of patients with lactate concentra- tion greater than or equal to 3 mmol/L to that reported herein, but this study population includes many patients with chronic alcoholism. MacDonald et al [9] reported that only 7 of 60 patients after alcohol intake had arterial lactate concentrations greater than or equal to

2.4 mmol/L, which corresponds to a venous lactate concentration of about 3.0 mmol/L [3]. Furthermore, Oshima et al [10] conducted a study of healthy volunteers to prove 1.5- to 2.0-fold elevation of venous lactate concentrations after alcohol intake, but the absolute values for venous lactate concentrations were reported at most as 2 mmol/L. Com- pared with those earlier studies, both absolute values and increase rates of venous lactate concentrations have a tendency to be higher in this study population, which includes large numbers of patients with no condition that might contribute to plasma lactate elevation.

Although venous lactate concentrations of patients without venous blood gas test were unknown, the conclusion will not change that acute alcohol intake leading to any change in vital signs or any change in level of consciousness raises the plasma lactate concentration. Be- cause liver function tests were not performed in some patients, patients with chronic liver injury might be present in each group, but analysis limited to patients with liver function tests shows the same tendency to support the conclusion. In addition, the proportion of patients, al- though they came to the ED after alcohol intake, with liver dysfunction attributable to chronic alcohol intake was relatively low compared with previous reports [7], which might reflect Japanese drinking habit. Al- though results of our study suggest that plasma ethanol concentrations might not correlate with lactate concentrations, direct comparisons be- tween serum ethanol concentrations and lactate concentrations must be done in future studies. Because alcohol intake itself can engender sig- nificant increases in plasma lactate concentrations, ED physicians must be careful when they interpret plasma lactate concentrations in alcohol- intoxicated ED patients. Plasma lactate concentrations should be discounted in this population as a valuable tool in triage or a point-of- care test to predict prognosis.

Limitations

This study has several limitations. First, cases were collected contin- uously but were analyzed retrospectively. However, as described

previously, all EDs were included in the database. The information was derived by standardized measures. Therefore, few biases exist in re- lation to patient collection and groups. Furthermore, most information is objective. Second, liver function tests were not conducted for all pa- tients. Finally, direct Correlation analyses must be done between plasma ethanol concentration and lactate concentration.

Conclusion

Among ED patients, plasma lactate concentrations are significantly higher in patients who have recently consumed alcohol, even in pa- tients without conditions known to increase lactate concentrations. Emergency department physicians must be careful in interpreting the lactate concentrations of such patients.

References

AJEM