Assessment of Extracorporeal Treatments in Poisoning criteria for the decision of extracorporeal toxin removal in lithium poisoning

Abstract

Aims: To assess recommendations provided by the EXtracorporeal TReatments In Poisoning (EXTRIP) workgroup on extracorporeal toxin removal (ECTR) in lithium poisoning.

Methods: Retrospective assessment in a 128 lithium-poisoned patient cohort previously used to identify ECTR initiation criteria that could improve outcome (Paris criteria). ECTR requirement using EXTRIP criteria was compared to the actual practice or if Paris criteria were used. The potential impact on outcome if these different criteria were used was investigated.

Results: Using the recommended (Rec-EXTRIP) or recommended + suggested (All-EXTRIP) EXTRIP criteria, ECTR would have been indicated in more patients than was actually done (P < .001), or if Paris criteria were used (P < .01). The non-actually ECTR-treated patients fulfilling Rec-EXTRIP or All-EXTRIP criteria had shorter intensive care unit stay (P < .05) and no significant increase in fatalities and neurological impairment on discharge in comparison to the actually ECTR-treated patients. ECTR requirements using EXTRIP vs Paris criteria were not concordant (P < .001). In the non-actually ECTR-treated patients, 31/106 and 55/106 patients fulfilled Rec-EXTRIP or All-EXTRIP but not Paris criteria, respectively. Those patients had longer stay (P < .01) but no worse neurological impairment on discharge than the patients not fulfilling any of these criteria (50/106 and 26/106, respectively). In the non-actually ECTR-treated patients, 7/106 fulfilled Paris but not Rec-EXTRIP criteria. Those patients had longer stay (P < .05) and worse neurological impairment on discharge (P < .01) than the 50/106 patients not fulfilling any of these criteria.

Conclusion: In this cohort of lithium poisonings, EXTRIP criteria may lead to more ECTR than actually performed or if the Paris criteria were used, with no demonstrated improvement in outcome.

Keywords: EXTRIP; extracorporeal toxin removal; haemodialysis; lithium; outcome; poisoning.

Figure 1

Comparison of extracorporeal toxin removal (ECTR) requirement and outcome of patients fulfilling Paris criteria or EXTRIP criteria for ECTR initiation according to whether they were actually treated with ECTR or not. Due to the lack of consensual definition in EXTRIP guidelines, decreased consciousness, significant confusion and serum lithium concentration expected to be >1 mmol/L at 36 hours with optimal management, were interpreted. Lithium poisoning was defined as severe in the presence of at least 1 of: (i) seizures; (ii) catecholamine infusion; (iii) mechanical ventilation lasting >48 h. data are expressed as percentages or median [25^th; 75^th percentiles]. *P < .05 and **P < .01. ICU LOS, intensive care unit length of stay

Figure 2

Comparison of Rec‐EXTRIP (a) and All‐EXTRIP (B) criteria vs Paris criteria to decide ECTR initiation patients. Concordance of ECTR indication between Paris criteria and Rec‐EXTRIP or All‐EXTRIP criteria was tested using McNemar's test. Due to the lack of consensual definition in EXTRIP guidelines, decreased consciousness, significant confusion and serum lithium concentration expected to be >1 mmol/L at 36 hours with optimal management, were interpreted

Figure 3

Comparison of the outcome of the non‐actually‐ECTR treated patients according to the ECTR requirement indicated by Rec‐EXTRIP, All‐EXTRIP or Paris criteria. Severity and outcome of the patients fulfilling none of the Paris criteria + Rec‐EXTRIP criteria were compared to the patients fulfilling Paris criteria + Rec‐EXTRIP criteria, only Paris criteria, and only Rec‐EXTRIP criteria (a). Severity and outcome of the patients fulfilling none of the Paris criteria + All‐EXTRIP criteria were compared to the patients fulfilling Paris criteria + All‐EXTRIP criteria, only Paris criteria, and only All‐EXTRIP criteria (B). Due to the lack of consensual definition in EXTRIP guidelines, decreased consciousness, significant confusion and serum lithium concentration expected to be >1 mmol/L at 36 hours with optimal management, were interpreted. Lithium poisoning was defined as severe in the presence of at least 1 of the following conditions: (i) seizures; (ii) catecholamine infusion; (iii) mechanical ventilation lasting >48 hours. Data are expressed as percentages or median [25^th; 75^th percentiles]. *P < .05,**P < .01 and ***P < .001. ICU LOS, intensive care unit length of stay