Is Carbon monoxide testing required on all deaths in the elderly?

Abstract

We were approached by our Coroners as there had been concern raised that there were deaths occurring in the elderly from unrecognised carbon monoxide toxicity that were not being investigated at post mortem. Rather than conduct additional testing in post mortem cases, we reviewed the data from admission to the hospital, where all admissions to our “major illness” have blood gases including carbon monoxide measured. All data was collected in 2017. All admissions to the “majors” unit will have had either a venous or arterial blood gas analysis, Radiometer ABL 90 Flex Blood Gas analyser. The data was anonymised to the laboratory team but all cases in which the carboxyhaemoglobin was greater 10%, a clinician reviewed the clinical notes to investigate if there was a history of excess carbon monoxide exposure. The population covered by the emergency department was approximately 700,000. After removal of duplicates from the same admission, and failed analysis there were 48,686 carboxyhaemoglobins analysis conducted. The age range of patients was from <1 year to 107 years. There was no difference between arterial and venous carboxyhaemoglobin concentrations. In all age groups the mean and median carboxyhaemoglobin were less then 2%. There was no difference in carboxyhaemoglobin levels between any of the age groups (divided by decades), gender, or time of year. There was only 50 cases (0.1%) in which the carboxyhaemoglobin was greater then 10% Hb, with only three cases having levels about 20%, with the highest cases having a level of 37%. Only one case had a history possible excess carbon monoxide inhalation (measured at 10.6%). 6/50 cases presented with a drug overdose, 2/50 presented with seizures and 5/50 with infections. With regard to the elderly population, the cases of COHb>10% Hb were 60–70 years = 3/5448 cases, 70–90 = 0/15350, >90 years 4/2544 cases. This data shows that the probability of presenting with carbon monoxide toxicity when it is not suspected was vanishingly rare in the year in question. Carboxyhaemoglobin has a half life of 4–6 hours when breathing air, so it is possible there may have been a slight reduction in levels from admission, but as each of these cases was seen in the “majors” area of the casualty department they will have been assessed within a 1–2-hour time period. It is reassuring that there was no increase in toxicity in the winter months compared to the summer months, suggesting that at least in the catchment area of this hospital, there is good maintenance of gas appliances. There is no evidence that any age group is suffering unrecognised carbon monoxide toxicity. There is no justification, at least in the jurisdiction that was studied, and probably for most of the United Kingdom, for conducting carbon monoxide analysis on post mortem cases unless there is a reasonable history of excess exposure.