The most important finding of this observational study was that amlodipine (CCB) accounts for 1.4% of ICU admissions, coupled with a high mortality rate. A recent review indicated a lack of data on the frequency of CCB in South Africa. [1] Outside of case reports, we could not find any specific local ICU data in the literature, and our data provides the first published estimate of the burden of amlodipine overdose in an ICU setting. Data from the Danish Poison Information Center (DPIC) indicated that amlodipine accounted for 71% of all CCB enquiries, supporting the significance of our finding. [14] Furthermore, the American Association of Poison Control Centers’ National Poison Data System reported 6132 cases of calcium channel blocker overdose in the United States of America. [5] This 2020 report detailed a 2.4% major adverse event rate including death. According to the DPIC data from a 5-year period in the past decade, the 30-day hospital mortality rate from intentional CCB overdose was 2% and amlodipine was involved in 29% of fatalities. [14] The mortality rate in our study was about 6-fold higher (12.5%) likely reflecting the mortality in a critically ill population requiring ICU care.
Another striking finding from our data was the female preponderance (95% female gender). The DPIC data also reflected a majority of female cases (61%). [14] The gender paradox in suicidology has been well described with a greater frequency of nonfatal suicide attempts in females compared to males. In addition, drug or toxin use in females has been reported to be more common than violent methods [15).
The frequency of interventions in the ICU used to treat CCB overdose reflects current guidelines. [12, 16, 17, 18] All patients required fluid therapy (IV), over 90% required vasopressor support with noradrenaline, adrenaline, or both, about 70% required high-dose insulin euglycaemic therapy (HIET), and 42% required ongoing calcium in ICU. These measures are indicative of first-line treatment published in consensus guidelines. [17] Rescue therapy availability is limited in our setting. Methylene blue is intermittently available, intralipid therapy and extracorporeal membrane oxygenation (veno-arterial) are unavailable. Lipid therapy has been noted to be ineffective in CCB overdose involving dihydropyridines [19]
In addition to standard of care, hemoadsorption therapy (HA) is available and used for protein and lipid-bound toxins in severe poisoning. Based on the physical characteristics of CCBs modern resin hemoadsorption devices could be useful to enhance the elimination of these drugs. [20] Data from an ex-vivo model demonstrated the effective removal of amlodipine using a modern resin adsorption filter similar to the one available in our setting. [21] Removal of amlodipine in a clinical setting, using a styrene resin adsorption filter with associated shock resolution and good clinical outcome has also been suggested. [22] Current EXTRIP guidelines recommend against the use of extra-corporeal therapy for CCB overdose. [23] However, this guideline is based on hemodiafiltration and older charcoal adsorption filter technology. [24] Consideration of the affinity for protein-bound substances that modern resin adsorption filters possess may have been overlooked. [20]
Patients receiving HA therapy had a significantly higher SAPS II score indicating greater severity of illness and higher predicted mortality compared to standard care (SoC). The total maximum vasopressor dose (adrenaline and noradrenaline) was also significantly higher. This reflects the use of HA therapy in patients who were refractory to standard care. Despite a significantly lower mean arterial pressure (MAP), there was a greater recovery in the HA group's 48-hour mean arterial pressure profile. It is plausible that HA therapy may have contributed to the haemodynamic recovery.
Due to the greater haemodynamic instability, initiation of HIET therapy was more frequent in the HA group (85% vs 55%) and significantly higher doses were administered. Despite this, we noted a trend toward a shorter duration of HIET in the HA group compared to the SC group (about 1 day vs. 3 days). There is scarce data on modern hemoadsorption therapy and amlodipine overdose. We postulate that hemoadsorption therapy enhanced the clearance of amlodipine, potentially contributing to a shorter duration of HIET therapy and improved hemodynamic stability, despite initially higher vasopressor doses. [21, 22] We further speculate that additional potential benefits of less HIET therapy could be due to the limitation of adverse effects of this therapy. These are hypoglycemia, hypokalemia, hypomagnesemia, hypophosphatemia and volume overload related to glucose and electrolyte replacement. [25, 26, 27]
Important data from our study reinforces the greater severity of disease in the HA group compared to the SC group with longer durations of vasopressors, mechanical ventilation, and length ICU of stay. Despite this greater severity of disease and the higher predicted mortality in the HA group, there was only one death in the HA group (1/13) 7.7% and two in the SoC group (2/11) 18.2%. This possible mortality benefit of HA therapy needs further exploration.
Our study provides important detail regarding the changes in lactate during the first day of shock from CCB overdose. Median lactate levels were similar at ICU admission. Median lactate levels in the SoC group peaked early, at 3 hours, and cleared slowly over the next 11 hours to reach a low at 14 hours, coinciding with the peak of the HA group. The HA peak lactate decreased to its 24-hour nadir within the next 6 hours, corresponding to approximately half the time of the SoC group. If the reduction of lactate over time is an estimate of lactate clearance this may indicate a faster resolution of a global oxygen delivery deficit (shock) in the HA group. [28] Improved lactate clearance has also been shown to affect mortality independent of its effect on organ support. [29] The potential higher lactate clearance in the HA group needs to be further explored as it may indicate faster shock resolution, increased extracorporeal lactate removal or both.
The lactate peak was significantly greater in the HA group vs. the SoC group and occurred later. This is in keeping with ongoing slower amlodipine absorption from the gastrointestinal tract, possibly with higher doses and associated delayed peak hemodynamic effects as supported by the longer time to peak vasopressor dose in the HA group. Additionally, saturation of the hepatic enzymes may reduce the first pass effect and contribute to ongoing shock from an apparent prolonged T1/2 and toxic effects with a delayed peak lactate. [8]
Lastly, an important complication of hemoadsorption/hemoperfusion therapy is thrombocytopenia. We found no significant difference in the maximum and minimum platelet counts between the HA and SC groups during the ICU stay. Previous utilisation of this hemoadsorption filter also demonstrated no significant changes in platelet counts or transfusion requirements.[30] This finding provides much-needed safety information regarding HA therapy.
Limitations
This was an observational study with a small sample size. However, it is notable that current guidelines suggest first-line treatment strategies with low levels of evidence. Being retrospective, the study is open to bias and confounding. However, we have described both groups of patients in terms of their predicted risk using underlying physiology and organ function and support. We have also used detailed chronological changes in important clinical parameters to improve characterization of the groups. Finally, the study data was from a single centre limiting the validity of findings even though it represents the only ICU description of CCB overdose that we are aware of.