Recommendations for the treatment of calcium channel blocker poisoning
St-Onge M, Anseeuw K, Cantrell L, Gilchrist IC, Hantson P, Bailey B, Gosselin S, Kerns W 2nd R, Laliberté M, Mégarbane B, Jang D, Lavonas EJ, Juurlink DN, Muscedere J, Yang CC
Sinuff T, Rieder MJ, Lavergne V
ABSTRACT
INTRODUCTION: The main objective of these recommendations is to decrease practice variation and improve the management of patients with acute calcium channel blocker (CCB) poisoning by providing evidence-based treatment guidance. These recommendations are endorsed by 12 international critical care, emergency medicine and toxicology associations and address interventions for in-hospital management. The target users are physicians, consultants, other healthcare providers and poison control centres.
METHODS: A working group was established and followed the process outlined by the AGREE II instrument. The working group initially used the evidence documented in a systematic review that pre-dated the establishment of the group. The search was then completed by complementary sources. Members were separated into subgroups to evaluate each candidate intervention, outlining the level of evidence (based on the GRADE system), risks, benefits, and costs. Initial voting statements were constructed based on these evidence summaries, which were then refined using four rounds of modified Delphi. The strength of recommendation was determined by the vote results, using the median votes, lower/upper interquartiles and disagreement indexes (RAND/UCLA Appropriateness Method).
RESULTS: The overall level of evidence was very low. The main recommendations are as follows: 1) In asymptomatic patients, we recommend observation and decontamination (following the position statement published by the EACCT/AACCT) after an ingestion of a potentially toxic amount of CCB (1D); 2) As a first-line treatment, we recommend intravenous calcium (1D), high-dose insulin therapy (in combination with IV fluids, calcium and vasopressors) (1-2D) and norepinephrine and/or epinephrine (1D) in the presence of shock. We also suggest dobutamine in presence of cardiogenic shock (2D) and/or atropine in presence of symptomatic bradycardia or conduction disturbance (2D); 3) In patients refractory to first-line treatments, we suggest incremental doses of high-dose insulin therapy if evidence of myocardial dysfunction is present (2D), intravenous lipid emulsion therapy (2D), and using a pacemaker in presence of unstable bradycardia or high-grade AV block, without significant alteration in cardiac inotropism (2D); 4) In patients in refractory shock or peri-arrest, we recommend, as rescue treatments, incremental doses of high-dose insulin therapy (1D) and/or intravenous lipid emulsion therapy (1D) if not already tried. We also suggest VA-ECMO (or ECLS) in presence of cardiogenic shock in centres where the treatment is available (2D), and/or using pacemaker in presence of unstable bradycardia or high-grade AV block without significant alteration in cardiac inotropism (2D) if not already tried; 5) In patients in cardiac arrest, we recommend intravenous calcium in addition to standard advanced cardiac life-support (1D), lipid emulsion therapy (1D), and we suggest VA-ECMO (or ECLS) in presence of a low flow, for less than 5 min, and in centres where the treatment is available (2D).
CONCLUSION:
These recommendations for the treatment of CCB-poisoned adults include the stepwise administration of various therapies as a function of poisoning severity. However, the level of evidence for all interventions was very low in adults and absent in children. Further research is needed to better characterize the utility and comparative effectiveness of available treatment options.
INTRODUCTION
Calcium channel blocker (CCB) poisonings occur relatively frequently and can result in significant morbidity and mortality. According to the United States National Poison Data System, cardiovascular drugs are the second most common category associated with the largest number of fatalities.1 A recent retrospective study found that CCB poisoning is associated with significant morbidity in 50% of patients including acute renal failure (35%), aspiration pneumonia (15%) or cerebral anoxia (4%), with a mortality rate of 6%. Additional reports suggest that only 42% of treatments provided accorded with advice provided by a regional poison control centre.2 Similar findings were noted by Darracq et al.3 and Espinoza et al.,4 with respect to the use of high-dose insulin (HDI). An evidence-based consensus guidelines has been published by Olson et al for out-of-hospital management of calcium channel blocker ingestion,5 but current recommendations for evidence-based in-hospital care have not been systematically developed and may vary from one toxicologist to another due to differences between individual, local, or regional appraisal of the literature or access to resources. Hence, our goal was to develop evidence-based recommendations to guide the management of patients with CCB poisoning.
Calcium channel blockers, can be separated into two categories: dihydropyridines such as nifedipine; and non-dihydropyridines such as diltiazem (a benzothiazepine) and verapamil (a phenylalkylamine). Dihydropyridines preferentially block calcium channels in vascular smooth muscles, causing vasoplegic hypotension and, in many patients, compensatory tachycardia. Dihydropyridines have lower affinity for myocardial calcium channels, but this selectivity attenuates at higher doses.6 In addition to their peripheral effects, non-dihydropyridines generally slow the heart rate by inhibiting L-type calcium channels in the sinoatrial and atrioventricular nodes.7
The oral bioavailability, onset of action and plasma half-life vary among the available CCBs.8 Moreover, it is difficult to estimate the half-life of an individual CCB in an overdose situation, since the pharmacokinetics may be altered by coingestants, the presence of an ileus or decreased gut perfusion due to shock, or vasopressor administration. Bezoars have been reported to occur in patients taking CCBs, notably with certain formulations of extended release nifedipine.9
In light of the possible loss of selectivity at very high CCB doses and the altered pharmacokinetics following overdose, the following recommendations adopt a clinically-oriented approach and do not focus on specific CCBs or formulations.
METHODS
Objective, scope, target users and analytical framework
The objective for the development of recommendations for the treatment of CCB poisoning was to provide an evidence-based document to help reduce practice variation, with the ultimate goal of improving the management of CCB-poisoning, The recommendations address what types of in-hospital interventions should be considered for patients who have ingested a potentially toxic amount of CCB, according to their clinical status. Targeted users include the following: bedside or telephone-consulting physicians, other healthcare providers (pharmacists and nurses), and poison control centres.
An analytical framework illustrating the links between key questions to be answered during the recommendations development process was created (Figure 1). This was done as suggested by the US Preventive Task Force13 to integrate heterogeneous evidence in the development of a treatment approach including multiple interventions. The AGREE II statements10 provided the basis for the process of developing these recommendations.
Recommendations development working group
A working group representing all participating professional international healthcare organizations in emergency medicine, critical care, paediatrics, and toxicology (Table 1) was created. Working group members were selected based on their content expertise. The Canadian Association of Poison Control Centres (CAPCC) acted as the leading association, named a chair and a co-chair, and a representative was appointment by each participating organization.
The evidence
As a starting point, the working group used a systematic review that pre-dated the establishment of the consensus group and had been published as a distinct article (registry number: CRD42012002823).14 The systematic review included all study types involving humans or animals poisoned with a CCB and examined the effects of all interventions on the outcomes targeted by our recommendations up to December 31, 2013 (see online Appendices 1 and 2). The eligibility criteria (study types, participants, interventions, comparisons and outcomes) and search strategy were similar to the ones used for the systematic review.14 Nevertheless, in order to be thorough and complete, additional literature brought to the attention of the working group by any member was considered for inclusion. Data extraction, synthesis and presentation was performed as previously detailed in the systematic review.14
Type of studies: Systematic reviews, controlled trials, observational studies, case series and animal studies were considered appropriate evidence for the first three questions in Figure 1. While guidelines often discount animal studies in their synthesis of evidence, Lamontagne et al. (2010)15 suggest that animal research may enhance knowledge useful for clinical practice especially when the evidence is limited in human subject trials. In order to address the fourth key question, we examined any study that could identify an association between the intermediate outcomes and patient-centered health outcomes. Thus, only controlled trials and observational studies were included. To answer the fifth key question (adverse effects of therapy), case reports were considered in addition to other types of evidence.
Type of study subjects: Studies involving adult and children or on animals poisoned with any CCB were eligible, but not enough paediatric articles were find to include this population in the recommendations.
Type of intervention: Any treatment for the CCB poisoning was eligible as long as outcome measures were reported.
Type of outcome measures: Outcomes included mortality (survival at discharge for human studies or LD50 or time of survival for animal studies), functional outcomes (defined as a return to functional baseline), and duration of stay in intensive care unit or in hospital. Intermediate outcomes included prevention of (or attenuation of) toxicity, a decrease in CCB serum level, improved hemodynamics and a decrease in duration of vasopressor use. Differences in outcome were evaluated on experience, preferences and values. Studies measuring the association between intermediate outcomes and health outcomes were selected in order to answer the fourth key question. Adverse effects of treatment and costs were also documented to answer the fifth key question. Adverse effects were documented by the systematic review previously mentioned.14 However, since costs were not reported, two working group members conducted a cost-effectiveness analysis with other co-authors concerning the use of venoarterial extracorporeal membrane oxygenation (VA-ECMO), the most expensive intervention.16
Definitions and terminology
The stepwise approach used by the working group included the following clinical categories: asymptomatic patients, symptomatic patients requiring first-line treatment, patients refractory to first-line treatment, patients in refractory shock or peri-arrest, patients in cardiac arrest. First-line treatment was defined as treatments initially provided to a symptomatic CCB-poisoned patient. Patients refractory to first-line treatment were classified as such when desired effects (see Table 3) were not significantly reached with first-line treatments, whereas rescue treatments were those provided to patients in refractory shock or peri-arrest. Refractory shock was defined as persistent cardiovascular failure associated with organ failure despite the administration of supportive care and adequate antidotes.
The members defined signs of CCB toxicity as hemodynamic abnormalities, such as low heart rate (60 per minute in adults), low blood pressure (systolic 100 mmHg or mean arterial pressure 65 mmHg in adults), myocardial dysfunction or abnormal peripheral vascular resistances. Reference values and definitions supported by the American Heart Association17 were adopted by the working group, but the members also recognized the importance of clinical judgement. Therefore, as suggested by the American Heart Association (2008),17 myocardial dysfunction was defined as a decrease in myocardial contractility seen on the echocardiography or a documented cardiac index of less than 2.2 L/min/m2; while shock was defined as a state characterized by inadequate blood flow and oxygen delivery to organs and tissues.17
The recommendation statements
The working group was divided in subgroups. Each subgroup responsible for a specific intervention presented a summary outlining the evidence, the benefits and the risks and costs. The subgroups developed statements focusing on specific interventions to be used in certain clinical presentations (asymptomatic patients, symptomatic patients needing first-line treatments, patients refractory to first-line treatments, patients in refractory shock or peri-arrest requiring rescue treatments or patients in cardiac arrest) in order to achieve a specific outcome. Those statements were used for the subsequent modified Delphi process.
Each statement was associated with a specific level of evidence, which was determined using the GRADE11 system (Table 2). In order to determine the strength of recommendation (Table 2), the working group proceeded to a modified Delphi method (four rounds of anonymous online votes using a 9-point Likert scale followed by telephone meetings and a face-to-face meeting held in Brussels in May 2014). When a statement needed to be clarified, it was modified and voted on again until the working group members agreed that the results reflected a clear understanding of what the statements meant and implied. For each statement, the strength of recommendation (Table 2) was determined by the results of the votes at the last round using the medians, the lower/upper interquartiles, and the disagreement indexes (RAND/UCLA Appropriateness Method)12 as described in Figure 2.
Values and preferences
In accordance with AGREE II methodology, the perceived influence of the working group values and preferences on the vote results was documented at each round. In addition to considering values and preferences of decision makers, clinicians, patients and relatives, the draft recommendations were posted on a blog for two weeks. The public was asked to provide comments and suggestions to which the working group responded. In order to encourage participation, messages were posted on social media websites (Facebook, Twitter, Google +) of relevant organizations of patients and relatives (search key words: patient care, suicide prevention, mental health, substance of abuse, heart disease, hypertension, health ethics, research ethics), poison control centers, and professional organizations (search key words: emergency medicine, critical care medicine, toxicology, psychiatry, pharmacy).19
Internal and external review
The recommendations for the treatment of CCB poisoning were submitted to all participating association for internal review and to anonymous reviewers chosen by the associations for external review. The external reviewers evaluated the guideline development process with the AGREE tool.10 A second face-to-face meeting was held in New Orleans in October 2014 to discuss the documented values, preferences, and the results of the internal and external reviews.
RESULTS
Table 3 details of the rationale for each recommendation and Figure 3 illustrates the progression of care for key recommendations.
RECOMMENDATIONS
Therapy in asymptomatic patients
For the treatment of patients who ingested a potentially toxic amount of CCB, the working group recommends observation and decontamination following the position statement published by the EACCT/AACCT22 (1D).