Usefulness of serum lactate as a predictor of successful discontinuation of continuous atropine infusion in patients with severe acute organophosphate poisoning

Study setting and population

This was a retrospective and observational study of consecutive adult patients diagnosed with acute OP insecticide poisoning who presented to the emergency department (ED) between March 2011 and December 2016. The ED in this study is located in a suburban tertiary-care hospital, has more than 44,000 annual visits, and is staffed 24 hours per day by board-certified emergency physicians.

OP poisoning was confirmed by patient or guardian statements, and verification of the agent was determined by an emergency physician who transcribed the bottle label into the patient records. Clinical manifestations consistent with OP poisoning and PChE levels were also used to confirm OP poisoning. OP-poisoned patients were treated with decontamination and administration of pralidoxime and atropine. The protocol for pralidoxime administration at our hospital is to infuse 1 to 2 g over 20 to 30 minutes followed by 0.5 to 1.0 g/hr for at least 48 hours according to the severity of clinical conditions and the time course of PChE activity. For atropine, an initial IV bolus of 1 to 10 mg depending on the severity of symptoms and continuous infusion was administered to patients with more than moderate poisoning according to the Namba classification [11].

The study exclusion criteria were (1) <19 years old, (2) nonoral poisoning, (3) non-use of continuous atropine infusion owing to mild poisoning, (4) poisoning by any additional agent (with the exception of alcohol), (5) cardiac arrest on ED arrival, (6) transfer to another hospital after ED admission, (7) additional medical conditions such as chronic liver disease or malignant tumors, which can cause changes in PChE (PChE is primarily synthesized in the liver), and (8) insufficient data including unknown subtype of OP. Since the study was performed retrospectively and observationally, we did not obtain informed consent from participants. Approval of this study was obtained from the institutional review board (CR316122).

Data collection

Data were collected by retrospectively reviewing electronic medical records. Data collection was conducted by 2 emergency physicians who were blinded to the study objectives and hypothesis. Data abstractors were trained before data collection to reduce possible bias during data collection. We used explicit case report forms, and chart abstractors and study coordinators met periodically to resolve any disputes and to review coding rules. The study coordinators monitored the performance of the abstractors. Because the study was performed retrospectively and observationally, patient records and/or information were anonymized before analysis.

We collected demographic data (including age and sex), type of OP, Namba classification [11], ingested amount of OP, elapsed time from ingestion to arrival at the ED, use of charcoal and gastric lavage, cause of poisoning, information regarding preexisting comorbidities, vital signs (systolic blood pressure, diastolic blood pressure, pulse rate, respiratory rate, and body temperature), initial mental status (Glasgow Coma Scale), and ST changes on an electrocardiogram. We investigated the use of vasoactive drugs such as norepinephrine or dopamine and mechanical ventilation as well as the presence of pneumonia during the trial discontinuation of atropine infusion. We evaluated complications, including pneumonia, endotracheal intubation, respiratory failure, decreased mental status, acute kidney injury [12], and hypotension during hospitalization. Total admission and ICU admission days and in-hospital mortality were also evaluated.

We estimated the amount of OP ingested as follows: “a little” or “a spoonful” was taken to be 5 mL, “a mouthful” was presumed to be 25 mL, “a small cup” was presumed to be 100 mL, and “a bottle” was presumed to be 300 mL [13]. The chemical structure of OP is an important parameter in the rate of aging of human AChE; the half-life of the aging process of AChE with dimethyl OP is 3.7 hours, whereas that with diethyl OP is 33 hours [14,15]. Therefore, we divided the subjects into 2 groups according to the chemical formulation of OP ingested: the dimethyl subtype and the diethyl subtype [16]. The unclassified subtype included subjects in which the ingested OP did not match either the dimethyl or diethyl classification. The severity of poisoning was classified as latent, mild, moderate, or severe using the Namba classification [11].

We evaluated arterial blood gas, serum lactate, and PChE levels on ED arrival (initial serum lactate or PChE) and within 12 hours before discontinuation of atropine infusion (serum lactate or PChE before trial discontinuation). If multiple measures of arterial blood gas, serum lactate, and PChE were taken within 12 hours before discontinuation of atropine infusion, the most recent value before the trial was selected. PChE levels (reference range, 7,000 to 19,000 U/L) were obtained using Dimension Vista (Siemens Healthcare Diagnostics, Newark, DE, USA).

At our facility, indications for atropine IV infusion discontinuation are clearance of respiratory secretions and cessation of bronchoconstriction [5]. Atropine infusion discontinuation attempts were divided according to recurrence of cholinergic symptoms and signs. A successful trial was defined as no recurrence of cholinergic symptoms after discontinuing atropine infusion. A failed trial was defined as the need to restart the atropine infusion due to recurrence of cholinergic symptoms including bronchorrhea or bronchospasm.

Study endpoint

The primary outcomes of this study were the investigation of differences between successful and failed trials in terms of serum lactate and PChE levels checked within 12 hours before discontinuation of atropine infusion and investigation of the predictive value of serum lactate and PChE levels for successful discontinuation of atropine infusion in patients with severe OP poisoning.

Statistical analysis

Categorical variables are presented as frequencies and percentages, and continuous variables are presented as means and standard deviations or as medians and interquartile ranges. Normality was assessed using the Shapiro-Wilk test. The chi-square test or Fisher exact test was used to compare categorical variables, while the 2-sample t-test or the Mann-Whitney U-test was used to compare continuous variables. Variables with P-values of <0.05 by univariate analysis were entered into the multiple logistic regression analysis to factors predicting successful discontinuation of atropine infusion in patients with severe OP poisoning, and the results are expressed as odds ratio with 95% confidence interval (CI). The area under the curve (AUC) for the predictive ability of variables was determined using receiver operating characteristic curves. P-values of <0.05 were considered statistically significant, and analyses were performed using IBM SPSS Statistics ver. 23 (IBM Corp., Armonk, NY, USA) and MedCalc Statistical Software ver. 17.5.3 (MedCalc Software, Ostend, Belgium).

Patient characteristics

A total of 87 consecutive adult patients with acute OP poisoning were identified during the study period. Twenty-five patients were excluded based on the following criteria: (1) <19 years old (1 patient); (2) non-oral poisoning (1 patient); (3) non-use of continuous atropine infusion due to mild poisoning (2 patients); (4) poisoning by any additional agent (with the exception of alcohol) (8 patients); (5) cardiac arrest on ED arrival (4 patients); (6) transfer to another hospital after ED admission (1 patient); (7) additional conditions such as chronic liver disease or malignant tumors (2 patients); and (8) insufficient data including unknown subtype (6 patients). After exclusion, 62 patients with severe acute OP poisoning were enrolled in this study (Fig. 1).

Fig. 1.

Study flow chart. OP, organophosphate; ED, emergency department.

A total of 95 atropine discontinuation attempts were made in 62 patients. Baseline characteristics and laboratory findings are shown in Table 1. Of the 62 analyzed patients, 47 (75.8%) were male. The mean age was 60 years (range, 18 to 91 years). The dimethyl subtype was the most common type of OP poisoning (30 patients, 48.8%). The dimethyl subtype included dichlorvos (10 patients), phosphamidon (14 patients), methidathion (4 patients), and fenthion (2 patients). The diethyl subtype included chlorpyrifos (13 patients), parathion (1 patient), and diazinon (2 patients). The median ingested amount was 100 mL and intentionality was present in 52 patients (83.9%). The median initial Glasgow Coma Scale was 8, and initial PChE and serum lactate levels were 322 U/L and 4.38 mmol/L, respectively. Endotracheal intubation due to respiratory failure or unprotected airway due to decreased mental status (53 patients, 85.5%) were the most common complications during hospitalization. Pneumonia and hypotension were found in 23 patients (37.1%) and 19 patients (30.6%), respectively. The median total admission and ICU admission days were 15.5 and 9.0 days, respectively. Pneumonia and multiple organ failure led to in-hospital mortality in 8 patients (12.9%).

Table 1.

Baseline characteristics of patients with severe acute organophosphate poisoning

Comparisons of PChE and serum lactate levels according to success or failure of atropine infusion discontinuation

Thirty-seven patients (59.7%) experienced success on the first attempt to discontinue IV continuous atropine (Fig. 1). Comparisons of laboratory findings according to success or failure are shown in Table 2. There were no significant differences in the median interval from blood sampling to discontinuing IV atropine between successful and failed trials for either serum lactate sampling (5 vs. 5 hours, P=0.201) or PChE sampling (5 vs. 4.5 hours, P=0.861). Although PChE levels before trial discontinuation were higher in successful trial patients than in failed trial patients, there was no significant difference between groups (1,713 vs. 504 U/L, P=0.075). Subgroup analysis revealed that in the dimethyl OP subtype, successful trial patients had significantly higher levels of PChE before the trial than failed trial patients (2,439 vs. 490 U/L, P=0.030). There was also no significant difference between PChE level before trial discontinuation and initial PChE level (557 vs. 53 U/L, P=0.305). Subgroup analysis did not reveal any significant differences between PChE level before trial discontinuation and initial PChE level for any OP subtype. Serum lactate levels before trial discontinuation differed significantly between successful and failed trial groups (1.09 vs. 1.46 mmol/L, P<0.001). In subgroup analysis, patients with successful trials had significantly lower serum lactate levels before trial discontinuation regardless of OP subtype (dimethyl subtype: 1.08 vs. 1.41 mmol/L, P=0.032; diethyl subtype: 1.065 vs. 2.46 mmol/L, P=0.009). Significantly more vasoactive drugs were used during trial atropine discontinuation in failed trials than in successful trials (3.2% vs. 15.2%, P=0.047).

Table 2.

Comparison of laboratory findings according to success or failure in the discontinuation trial of continuous intravenous atropine

Predictors of successful discontinuation of atropine in acute OP poisoning

Serum lactate levels before trial discontinuation and use of vasoactive drugs were analyzed using multiple logistic regression to identify predictors related to successful discontinuation of atropine infusion in severe acute OP poisoning. Serum lactate (odds ratio, 0.516; 95% CI, 0.333 to 0.801, P=0.003) was found to be the only significant predictor for all OP cases (Table 3). The AUC for prediction of successful atropine discontinuation using serum lactate levels before trial discontinuation was 0.742 (95% CI, 0.638 to 0.846) (Fig. 2). The cutoff value for prediction of successful atropine discontinuation using serum lactate levels before trial discontinuation was 1.13 mmol/L (sensitivity, 84.9%; specificity, 58.1%; positive predictive value, 51.9%; negative predictive value, 87.8%). Subgroup analysis demonstrated that the serum lactate level before trial discontinuation was a significant predictor regardless of OP subtype (Table 3).

Table 3.

Predictors of successful discontinuation of continuous atropine infusion in patients with severe acute organophosphate poisoning as determined by multivariate logistic regression analysis

Fig. 2.

Receiver operating characteristic curves of serum lactate before the trial in severe acute organophosphate poisoning. AUC, area under the receiver operating characteristics curve; CI, confidence interval.

In the dimethyl and diethyl subtype, the AUC for prediction of successful atropine discontinuation using the serum lactate level before trial discontinuation was 0.690 (95% CI, 0.527 to 0.854) and 0.824 (95% CI, 0.658 to 0.991), respectively.