Prediction and validation of hemodialysis duration in acute methanol poisoning

The duration of hemodialysis (HD) in methanol poisoning (MP) is dependent on the methanol concentration, the operational parameters used during HD, and the presence and severity of metabolic acidosis. However, methanol assays are not easily available, potentially leading to undue extension or premature termination of treatment. Here we provide a prediction model for the duration of high-efficiency HD in MP. In a retrospective cohort study, we identified 71 episodes of MP in 55 individuals who were treated with alcohol dehydrogenase inhibition and HD. Four patients had residual visual abnormality at discharge and only one patient died. In 46 unique episodes of MP with high-efficiency HD the mean methanol elimination half-life (T1/2) during HD was 108 min in women, significantly different from the 129 min in men. In a training set of 28 patients with MP, using the 90th percentile of gender-specific elimination T1/2 (147 min in men and 141 min in women) and a target methanol concentration of 4 mmol/l allowed all cases to reach a safe methanol of under 6 mmol/l. The prediction model was confirmed in a validation set of 18 patients with MP. High-efficiency HD time in hours can be estimated using 3.390 x(Ln (MCi/4)) for women and 3.534 x (Ln (MCi/4)) for men, where MCi is the initial methanol concentration in mmol/l, provided that metabolic acidosis is corrected. The duration of hemodialysis (HD) in methanol poisoning (MP) is dependent on the methanol concentration, the operational parameters used during HD, and the presence and severity of metabolic acidosis. However, methanol assays are not easily available, potentially leading to undue extension or premature termination of treatment. Here we provide a prediction model for the duration of high-efficiency HD in MP. In a retrospective cohort study, we identified 71 episodes of MP in 55 individuals who were treated with alcohol dehydrogenase inhibition and HD. Four patients had residual visual abnormality at discharge and only one patient died. In 46 unique episodes of MP with high-efficiency HD the mean methanol elimination half-life (T1/2) during HD was 108 min in women, significantly different from the 129 min in men. In a training set of 28 patients with MP, using the 90th percentile of gender-specific elimination T1/2 (147 min in men and 141 min in women) and a target methanol concentration of 4 mmol/l allowed all cases to reach a safe methanol of under 6 mmol/l. The prediction model was confirmed in a validation set of 18 patients with MP. High-efficiency HD time in hours can be estimated using 3.390 x(Ln (MCi/4)) for women and 3.534 x (Ln (MCi/4)) for men, where MCi is the initial methanol concentration in mmol/l, provided that metabolic acidosis is corrected. Methanol poisoning (MP), if untreated, may result in blindness, coma, and death, and today it still remains a xenobiotic of concern in outbreaks.1.Becker C.E. Methanol poisoning.J Emerg Med. 1983; 1: 51-58Abstract Full Text PDF PubMed Scopus (52) Google Scholar, 2.Methanol poisoning.Lancet. 1978; 2: 510-511Google Scholar, 3.Bennett Jr., I.L. Nation T.C. Olley J.F. Pancreatitis in methyl alcohol poisoning.J Lab Clin Med. 1952; 40: 400-405Google Scholar, 4.Gonda A. Gault H. Churchill D. et al.Hemodialysis for methanol intoxication.Am J Med. 1978; 64: 749-758Abstract Full Text PDF PubMed Scopus (95) Google Scholar, 5.Ghannoum M. Hoffman R.S. Mowry J.B. et al.Trends in toxic alcohol exposures in the United States from 2000 to 2013: a focus on the use of antidotes and extracorporeal treatments.Semin Dial. 2014; 27: 395-401Crossref Scopus (36) Google Scholar Hemodialysis (HD) can efficiently remove methanol and its toxic metabolite formic acid while correcting metabolic acidosis and improving patient outcome.4.Gonda A. Gault H. Churchill D. et al.Hemodialysis for methanol intoxication.Am J Med. 1978; 64: 749-758Abstract Full Text PDF PubMed Scopus (95) Google Scholar, 5.Ghannoum M. Hoffman R.S. Mowry J.B. et al.Trends in toxic alcohol exposures in the United States from 2000 to 2013: a focus on the use of antidotes and extracorporeal treatments.Semin Dial. 2014; 27: 395-401Crossref Scopus (36) Google Scholar, 6.Pappas S.C. Silverman M. Treatment of methanol poisoning with ethanol and hemodialysis.Can Med Assoc J. 1982; 126: 1391-1394PubMed Google Scholar, 7.McCoy H.G. Cipolle R.J. Ehlers S.M. et al.Severe methanol poisoning. Application of a pharmacokinetic model for ethanol therapy and hemodialysis.Am J Med. 1979; 67: 804-807Abstract Full Text PDF PubMed Scopus (97) Google Scholar Current guidelines suggest instituting HD when the methanol concentration is >15.6 mmol/l (50 mg/dl) until concentrations are inferior to 6.25 mmol/l (20 mg/dl) and acid-base disturbance has been corrected.8.Barceloux D.G. Bond G.R. Krenzelok E.P. et al.American Academy of Clinical Toxicology practice guidelines on the treatment of methanol poisoning.J Toxicol Clin Toxicol. 2002; 40: 415-446Crossref PubMed Scopus (506) Google Scholar, 9.Roberts D.M. Yates C. Megarbane B. et al.recommendations for the role of extracorporeal treatments in the management of acute methanol poisoning: a systematic review and consensus statement.Crit Care Med. 2014; 43: 461-472Crossref Scopus (93) Google Scholar However, methanol concentrations may not be readily available in an appropriate time frame, which may lead to either undue prolongation or premature termination of HD. Accordingly, Hirsch et al.10.Hirsch D.J. Jindal K.K. Wong P. et al.A simple method to estimate the required dialysis time for cases of alcohol poisoning.Kidney Int. 2001; 60: 2021-2024Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar proposed a formula to predict dialysis duration in toxic alcohol poisoning. Their proposal is based on a natural logarithmic formula requiring the knowledge of total body water using Watson’s formula (based on age, gender, height, and weight),11.Watson P.E. Watson I.D. Batt R.D. Total body water volumes for adult males and females estimated from simple anthropometric measurements.Am J Clin Nutr. 1980; 33: 27-39PubMed Scopus (1233) Google Scholar and the knowledge of dialyzer’s urea clearance at the prescribed blood flow and dialysate flow. The formula has been generated using two cases of MP and subsequently validated in three other cases.12.Youssef G.M. Hirsch D.J. Validation of a method to predict required dialysis time for cases of methanol and ethylene glycol poisoning.Am J Kidney Dis. 2005; 46: 509-511Abstract Full Text Full Text PDF PubMed Scopus (18) Google Scholar However, in the era of high-efficiency dialysis, we hypothesized that estimation of dialysis duration based on the methanol’s elimination half-life (T1/2) would be simpler and more intuitive in predicting HD duration. The aims of the present study were (1) to describe the clinical presentation, outcomes, and methanol toxicokinetics in a large cohort of methanol poisoned patients, (2) to use a training set of cases of MP to determine methanol elimination T1/2 during HD and to propose a simplified model to predict HD time, and (3) to validate the proposed model using a validation cohort. The study flow chart is shown in Figure 1. Between December 1997 and June 2013, among the 73 eligible MP episodes, 71 episodes occurred in 55 patients who required HD (Table 1). Baseline characteristics of subjects are presented in Table 1. Most patients were middle-aged healthy men with a history of psychiatric disorder. MP was intentional in 66 episodes (93%). Characteristics of poisoning episodes are summarized in Table 1. Nearly half of the cohort also coingested other substances.Table 1Clinical characteristics of patients and episodes of methanol poisoningParametersValueSubjects characteristics (n)55 Male gender (n, %)38 (69) Age (years)46 (36,55) Psychiatric disease (n, %)48 (87)Intoxication episode (n)71 Creatinine on admission (μmol/l)75 (68,88) Intentional ingestion (n, %)66 (90) Delay between ingestion and ER consultation (min)230 (90,510) Delay 24–36 h4 Delay 36–96 h2 Delay>96 h1 Delay between ER and HD (min)540 (390,705) Co-ingestion (n, %)31 (44) Ethanol16 Isopropyl3 Salicylate2 Acetaminophen2 Cocaine2 Cannabis2 Others7 ≥2 co-ingestants8Clinical severity Glasgow <8 (n, %)22 (31) Intubation (n, %)21 (30) Requirement for inotropes/vasopressors (n, %)9(13) Visual abnormality at presentation (n, %)9(13)Laboratory Arterial pH7.31 (7.23, 7.38) Anion gap (mmol/l)16 (13,25) Osmol gap (mOsm/kg)74 (37,134) Initial methanol concentrationaTo convert to mg/dl multiply by 3.2. (mmol/l)61.5 (25.1, 110.7) Peak methanol concentrationaTo convert to mg/dl multiply by 3.2. (mmol/l)64 (26, 111.4)Alcohol dehydrogenase inhibition Ethanol (n, %)53 (74) Fomepizole (n, %)9(13) Both (n, %)9(13)Abbreviations: ER, emergency room; HD, hemodialysis.Values are medians (25th, 75th percentiles).a To convert to mg/dl multiply by 3.2. Open table in a new tab Abbreviations: ER, emergency room; HD, hemodialysis. Values are medians (25th, 75th percentiles). In most episodes, there were a mild metabolic acidosis, a slightly elevated anion gap, and an increased osmol gap. When compared with other patients, those who presented with a decreased level of consciousness were significantly more acidotic and had a higher osmol gap (Table 2). Among the nine patients who had visual abnormalities at presentation, four had residual visual symptoms at discharge. Compared with patients without any visual symptoms, those with persistent visual abnormalities had a significantly lower median initial pH (7.14 vs. 7.29 P=0.01), without a significantly higher anion gap (27 vs. 19 mmol/l, P=0.08) or an osmol gap (95 vs. 74 mOsm/kg, P=0.54). Only one patient with extreme acidosis (pH=6.85), high anion gap (40 mmol/l), high osmol gap (184 mOsm/kg), and a methanol concentration of 115 mmol/l was comatose on admission and developed cerebral edema resulting in neurologic death. Six patients experienced complications not directly related to MP such as pneumonia (n=4), pyelonephritis (n=1), and myocardial infarction (n=1). MP resulted in a median hospital stay of 2 days (1, 3) in our center before discharge or transfer to the referring center.Table 2Determinants of decreased level of consciousnessParametersLevel of consciousnessGlasgow≥8 (n=50)Glasgow<8 (n=21)P-valueArterial pH7.32 (7.26, 7.38)7.26 (7.13, 7.33)0.002Anion gap (mmol/l)17 (13, 24)16 (13, 27)0.30Osmol gap (mOsm/kg)52 (33, 108)131 (79, 164)<0.001Time to presentation(min)233 (91, 642)253 (105, 510)0.72Values are medians (25th, 75th percentiles). Open table in a new tab Values are medians (25th, 75th percentiles). Figure 2 shows methanol elimination T1/2 during HD according to gender. In univariate linear regression analysis, T1/2 was positively influenced by the male gender (β=21 min, 95% confidence interval (CI): 9-34 min, R2=0.21, P=0.001) and increasing weight (β=0.55, R2=0.13, P=0.015). In a stepwise multiple regression, gender remained the only significant determinant of methanol elimination T1/2 during HD. Neither the type of alcohol dehydrogenase (ADH) inhibition (ethanol or fomepizole), the initial methanol concentration, nor the HD-related parameters were found to be significantly associated with the changes in methanol elimination T1/2 in this cohort. The characteristics of training set, which consisted of 28 episodes of MP in 28 subjects (nine women), are shown in Table 3. Various half-lives were used to model HD duration to a calculated safe level of methanol<6 mmol/l: the use of gender-specific 90th percentiles and targeting a methanol level of 4 mmol/l by the end of HD session proved to be the most robust method in achieving a safe methanol level for all individuals using the bootstrap method (Table 4). Table 5 shows the delivered HD time, the individual modeling of HD time to a methanol of 6 mmol/l, and the 90th percentile modeling of HD time to 4 mmol/l (Figure 3). In the training set, the delivered HD time was longer compared with the individual modeling HD time by 172 min (95% CI: 107–237) but not statistically longer compared with the 90th percentile model of HD time by 7 min (95% CI: –52–66).Table 3Clinical characteristics and episode of methanol poisoning in training and validation setParametersTraining set (n=28)Validation set (n=18)Age (years)47 (40, 52)49 (41, 56)Creatinine on admission (μmol/l)77 (72, 87)71 (64, 87)Psychiatric disease24 (86)15 (83)Male gender (n, %)19 (68)11 (61)Weight (kg)69 (67, 80)70 (67, 80)Multiple substances15 between ER and HD severity Glasgow <8 (n, Intubation (n, Requirement (n, Visual abnormality presentation (n, Arterial (7.23, Anion gap (mmol/l)17 (13, (13, Osmol Initial methanol concentrationaTo convert to mg/dl multiply by 3.2. Peak methanol concentrationaTo convert to mg/dl multiply by 3.2. (26, methanol concentrationaTo convert to mg/dl multiply by 3.2. dehydrogenase inhibition Ethanol (n, Fomepizole (n, Both (n, in women 90th percentile (95% by in men 90th percentile (95% by confidence ER, emergency room; HD, hemodialysis.Values are median (25th, 75th or mean (95% confidence To convert to mg/dl multiply by 3.2. Open table in a new tab Table of the of target methanol concentration on the calculated residual methanol concentration using 90th percentile of methanol elimination set (n=28)Validation set of residual residual mmol/l (n, 95% by bootstrap of the of subjects with calculated residual methanol concentration using the bootstrap of residual residual mmol/l (n, 95% by bootstrap of the of subjects with calculated residual methanol concentration using the bootstrap of residual residual mmol/l (n, 95% by bootstrap of the of subjects with calculated residual methanol concentration using the bootstrap confidence are median (25th, 75th 95% of the of subjects with calculated residual methanol concentration 6 mmol/l using the bootstrap Open table in a new tab Table individual modeling and 90th percentile modeling of HD set (n=28)Validation set HD time 95% by bootstrap modeling HD modeling of HD time is as time to a calculated residual methanol concentration of using methanol concentration and elimination 95% by bootstrap HD modeling HD time 95% by bootstrap percentile modeling HD percentile modeling of HD time is as the time to a methanol concentration of using methanol concentration and the 90th percentile of gender-specific methanol elimination half-life of the training 95% by bootstrap HD percentile modeling HD time 95% by bootstrap confidence HD, hemodialysis. Values are mean (95% modeling of HD time is as time to a calculated residual methanol concentration of 6 mmol/l using methanol concentration and elimination 90th percentile modeling of HD time is as the time to a methanol concentration of 4 mmol/l using methanol concentration and the 90th percentile of gender-specific methanol elimination half-life of the training Open table in a new tab Abbreviations: confidence ER, emergency room; HD, hemodialysis. Values are median (25th, 75th or mean (95% confidence confidence Values are median (25th, 75th Abbreviations: confidence HD, hemodialysis. Values are mean (95% The validation set consisted of 18 episodes of MP in 18 patients women), which were from the training The validation set was to the training set in of clinical severity of and methanol elimination T1/2 during HD (Table 3). The 90th percentile modeling of HD time using a target methanol concentration of 4 mmol/l allowed to reach a calculated safe methanol concentration of mmol/l in all cases (Figure 3). Table 5 shows the delivered HD time, the individual modeling of HD time to methanol concentration mmol/l, and the 90th percentile modeling of HD The delivered HD time was longer compared with the individual modeling HD time by min (95% CI: and longer compared with the 90th percentile model of HD time by min (95% CI: Figure 4 shows based on the 90th percentile modeling to a methanol of 4 mmol/l, all patients in training and validation set of MP would the HD duration based on the individual modeling to a methanol concentration of 6 mmol/l. Figure 5 a gender-specific to predict HD time, using high-efficiency HD, based on the 90th percentile of elimination T1/2 and targeting a methanol concentration of 4 mmol/l. To predict dialysis time based on initial methanol concentration, we propose using the for women and for men, where MCi is the initial methanol concentration of the analysis, we a which that the 90th percentile of T1/2 and a target methanol concentration of 4 mmol/l a methanol concentration of mmol/l in all while resulting in the HD duration time Table with Table with Table with Table with Table study a simple from a model using the 90th percentile methanol elimination T1/2 (147 min in men and 141 min in women) and targeting a methanol concentration of 4 mmol/l to predict the duration of high-efficiency HD. proved to be in our study cohort and a safe methanol concentration in all study also that our formula may the for of methanol concentrations during HD. Methanol is a with a of and The elimination of methanol without an is of with an elimination of at D. Methanol and ethylene glycol of clinical and Toxicol. 1: PubMed Scopus Google Scholar When or ethanol is the T1/2 is significantly and with the total body elimination from the and the D. et and in methanol Toxicol PubMed Scopus Google Scholar, J. et for the treatment of methanol J Med. 2001; PubMed Scopus Google Scholar The systematic review and the clinical guidelines by the patients, where the mean T1/2 during HD was D.M. Yates C. Megarbane B. et al.recommendations for the role of extracorporeal treatments in the management of acute methanol poisoning: a systematic review and consensus statement.Crit Care Med. 2014; 43: 461-472Crossref Scopus (93) Google Scholar in patients, et D. et hemodialysis is to to methanol and during treatment for methanol poisoning.Kidney Int. 2014; Full Text Full Text PDF PubMed Scopus Google Scholar that the mean methanol elimination T1/2 during HD was min longer compared with the elimination T1/2 in the present is the result of lower blood vs. lower vs. and lower dialysate flow vs. in is it is slightly and has only been validated in three cases of D.J. Jindal K.K. Wong P. et al.A simple method to estimate the required dialysis time for cases of alcohol poisoning.Kidney Int. 2001; 60: 2021-2024Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar, G.M. Hirsch D.J. Validation of a method to predict required dialysis time for cases of methanol and ethylene glycol poisoning.Am J Kidney Dis. 2005; 46: 509-511Abstract Full Text Full Text PDF PubMed Scopus (18) Google Scholar we not validate in our cohort as the on was in clinical and we not estimate total body water using Watson’s To our this is the study to that men had a longer methanol elimination T1/2 is the result of a lower methanol of in is that the in clearance are by in elimination during its to total methanol clearance during HD. In the present study, we not any significant between HD parameters and the methanol elimination is in to was and D. et hemodialysis is to to methanol and during treatment for methanol poisoning.Kidney Int. 2014; Full Text Full Text PDF PubMed Scopus Google Scholar, J. D.M. et and operational parameters to with extracorporeal treatments.Semin Dial. 2014; 27: PubMed Scopus Google Scholar is potentially by a of in the delivered blood flow in all subjects a and a dialysate flow of and in the of dialysis in HD parameters been our it would been to the between HD parameters and J. D.M. et and operational parameters to with extracorporeal treatments.Semin Dial. 2014; 27: PubMed Scopus Google Scholar In the 71 episodes of in our study, only 4 episodes were of and blood flow during HD. high-efficiency HD was for of the episodes, that our can an on the management of the of MP, when HD is M. Hoffman R.S. Mowry J.B. et al.Trends in toxic alcohol exposures in the United States from 2000 to 2013: a focus on the use of antidotes and extracorporeal treatments.Semin Dial. 2014; 27: 395-401Crossref Scopus (36) Google Scholar, 9.Roberts D.M. Yates C. Megarbane B. et al.recommendations for the role of extracorporeal treatments in the management of acute methanol poisoning: a systematic review and consensus statement.Crit Care Med. 2014; 43: 461-472Crossref Scopus (93) Google Scholar In most cases of MP result from an acute intentional The and of patients treated for MP in large are from In the has been to be as high as et of cases of methyl alcohol poisoning in Assoc 60: Scholar, M. et of clinical and of methanol poisoning in Med Scholar, et poisoning in from 2000 to Toxicol. Scopus Google Scholar, et poisoning in of Toxicol PubMed Scopus Google Scholar However, there is available on the clinical characteristics and the of patients with MP in the of intentional MP. our cases and were acidotic at presentation as compared with is in of et of cases of methyl alcohol poisoning in Assoc 60: Scholar, M. et of clinical and of methanol poisoning in Med Scholar, et poisoning in from 2000 to Toxicol. Scopus Google Scholar, et poisoning in of Toxicol PubMed Scopus Google Scholar, et and ethylene glycol acute of Toxicol Scopus Google Scholar, H. to and ethylene glycol acute of Toxicol Scopus (2) Google Scholar, et for in patients with methanol Clin 2014; Scholar, H. A. A. et poisoning in role of in 2014; Scopus Google Scholar In our only one patient and four patients had persistent visual The of our cohort also be by an to and to HD which may be in et of cases of methyl alcohol poisoning in Assoc 60: Scholar, M. et of clinical and of methanol poisoning in Med Scholar, et poisoning in from 2000 to Toxicol. Scopus Google Scholar, et poisoning in of Toxicol PubMed Scopus Google Scholar, et and ethylene glycol acute of Toxicol Scopus Google Scholar, H. to and ethylene glycol acute of Toxicol Scopus (2) Google Scholar, et for in patients with methanol Clin 2014; Scholar, H. A. A. et poisoning in role of in 2014; Scopus Google Scholar In the of a presentation, most of the methanol may been to formic acid and the of the gap and methanol concentration may the severity of the In our study, there was of formic acid is a with of and to which has a elimination T1/2 under formic acid has a elimination et P. P. in methanol Toxicol. 2005; PubMed Scopus Google Scholar the elimination T1/2 of formic acid and HD and elimination T1/2 of and also that clearance by HD was However, of the HD parameters of subjects were P. P. in methanol Toxicol. 2005; PubMed Scopus Google Scholar In the study presented by et D. et hemodialysis is to to methanol and during treatment for methanol poisoning.Kidney Int. 2014; Full Text Full Text PDF PubMed Scopus Google Scholar where a of HD parameters was available, the T1/2 of was half of the methanol elimination T1/2 D. et hemodialysis is to to methanol and during treatment for methanol poisoning.Kidney Int. 2014; Full Text Full Text PDF PubMed Scopus Google Scholar To our this is the cohort of intentional MP, with clinical and proposed to predict HD duration has and it on the of methanol elimination which is intuitive to the prediction model has been using a large training and validation of has been using bootstrap and the and of the of in methanol concentrations during HD, our suggest that the delivered HD time longer is required to reach a safe level of 6 mmol/l. in our center where methanol concentrations are to the proposed still allowed to the HD duration as compared with the method of methanol in our validation The study for safe management of patients and of our prediction model may be when methanol be in a The study also has the of a retrospective were available for the of the patients, the of patients not be The proposed model is only for HD of when a blood flow and high-efficiency large are we the 90th percentile of methanol elimination there was a of but this was by targeting a lower methanol concentration of 4 mmol/l in our be in this consultation of methanol resulted in a of acidotic formic acid are not in most it is also to the acid-base during HD. as all patients were our be to the In the of high-efficiency HD, our proposed prediction which is based on the 90th percentile of the methanol elimination T1/2 and a target methanol concentration of 4 mmol/l, resulted in a safe methanol concentration of 6 mmol/l in all

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.