Date Published: January 17, 2019
Publisher: Public Library of Science
Author(s): Carsten Simonsen, Kristinn Thorsteinsson, Rikke Nørmark Mortensen, Christian Torp-Pedersen, Benedict Kjærgaard, Jan Jesper Andreasen, William Durante.
Carbon monoxide (CO) poisoning is frequent worldwide but knowledge regarding the epidemiology is insufficient. The aim of this study was to clarify the extent of this intoxication, its mortality and factors associated with mortality.
National databases from Statistics Denmark were used to identify individuals who suffered from CO-poisoning during 1995–2015, as well as information regarding co-morbidities, mortality and manner of death.
During the period from 1995 to 2015, 22,930 patients suffered from CO-poisoning in Denmark, and 21,138 of these patients (92%) were hospitalized. A total of 2,102 patients died within the first 30 days after poisoning (9.2%). Among these, 1,792 (85% of 2,102) were declared dead at the scene and 310 (15% of 2,102) died during hospitalization. Deaths due to CO-poisoning from smoke were intentional in 6.3% of cases, whereas deaths due to CO containing gases were intentional in 98.0% of cases. Among patients who survived >30 days, there was no significant difference in survival when comparing hyperbaric oxygen therapy (HBO) treatment with no HBO treatment after adjustment for age and co-morbidities such as drug abuse, psychiatric disease, stroke, alcohol abuse, arterial embolism, chronic obstructive pulmonary disease, cerebrovascular disease and atrial fibrillation. Several co-morbidities predicted poorer outcomes for patients who survived the initial 30 days.
Poisoning from smoke and/or CO is a frequent incident in Denmark accounting for numerous contacts with hospitals and deaths. Both intoxication and mortality are highly associated with co-morbidities interfering with cognitive and physical function. Treatment with HBO was not seen to have an effect on survival.
Since carbon monoxide (CO) has no smell, no taste and no color, it is very hard to detect without the use of a CO-detector. This stealthiness, along with its highly toxic properties, makes it extremely dangerous, and victims might not suspect any danger until they are incapacitated and need help to survive.
The study was accepted by the Data Protection Agency (GEH-2014-013 I-Suite nr: 02731) and it is a retrospective nationwide cohort study. All citizens in Denmark are equipped with a unique personal identification number (CPR number), which is registered in all contacts with the health care system. Settlement of the correct rate of reimbursement is based on the correct registration of both CPR number and diagnoses. According to current legislation in Denmark, it is not necessary to obtain approval by the individual patient or general ethics approval for register-based studies. All contacts are registered in various databases administered by Statistics Denmark. These databases can be accessed for health management purposes and epidemiological research. The CPR number is encrypted in the same way across databases, which enables cross-referencing without accessing personally identifiable information. Data were accessed using encrypted remote access to Statistic Denmark´s servers. Three different databases managed by Statistics Denmark were used. The Danish National Patient Register (DNPR) has registered all hospital admissions and discharge coding diagnoses in Denmark since 1978. Diagnoses and procedures are coded according to the International Classification of Diseases (8th revision (ICD-8) through the end of 1993, and 10th revision (ICD-10) thereafter) . Since 1875, the the Danish Register of Causes of Death has registered all deaths and cause(s) of deaths among citizens that died in Denmark. Registration has been in accordance with WHO’s rules since 1994 by ICD-10  and The National Prescription Registry which contains information regarding all claimed prescriptions since 1994, including date of dispensation, strength and quantity. This database has been proven to contain valid information regarding the use of medication in Denmark .
A total of 22,930 patients suffered from CO-poisoning from 1995–2015 in Denmark (Fig 1). A total of 2,102(9.2%) patients died within the first 30 days after CO-poisoning directly or with CO-poisoning as a contributing cause of death. Out of this group, 1,792 (85.3%) patients were declared dead at the scene and thus never reached a hospital. Of the 21,138 patients who were hospitalized or in contact with an emergency department 310(1.5%) died within 30 days. Autopsies were performed in 40.0% of deaths after 2002. No data prior to 2002 were available. The median follow-up time in this study was 8.1 years.
This study shows that HBO treatment was not associated with survival benefit compared with that of no HBO treatment after adjustment for several confounding factors. It also demonstrated that a large percentage of CO-poisoned victims had a history of psychiatric diseases and alcohol abuse. These co-morbidities, along with increasing age, were strongly associated with mortality. Approximately, 50% of deaths were intentional; however, a higher proportion of deaths caused by CO-poisoning from gas were intentional compared to intentional deaths due to poisoning from smoke from fire.
Poisoning from smoke and/or CO is a frequent occurrence in Denmark, accounting for numerous contacts with hospitals and numerous deaths every year. Suicide and suicide attempts contribute heavily to these numbers. Both intoxication and mortality are highly associated with co-morbidities, especially conditions that might interfere with cognitive and physical function. Treatment with HBO is not associated with an effect on short-term or long-term survival after adjusting for CO-morbidities and age.