Abstract
Alcohol is a common factor in many deaths. Chronic alcoholics may die suddenly from trauma, intoxication, and from preexisting disease. In many cases, there is no obvious cause of death in chronic alcoholics following initial postmortem examination. Determining the cause of death in chronic alcoholics can be challenging. Such deaths in chronic alcoholics may be due to a number of mechanisms, including alcoholic ketoacidosis and disorders of cardiac rhythm. Alcoholic ketoacidosis has been recognized as a cause of death since the 1990s. It can be diagnosed by postmortem analysis of ketone bodies, including acetone and β-hydroxybutyrate. It has been recognized clinically that patients with alcoholic liver disease have prolongation of the QT interval and a risk of sudden death. This paper reviews sudden natural deaths in chronic alcoholics, discussing pathological processes and mechanisms of death, especially where the cause of death is not obvious on initial autopsy.
Keywords
Introduction
Many deaths are associated with the use of alcohol and may be from natural causes, accidentally or intentionally inflicted trauma, or acute intoxications (1). Chronic alcoholics have an increased incidence of suicide, although rates may have been over-reported (2,3). Sudden death is common in chronic alcoholism. Forensic pathologists are regularly faced with investigating these deaths.
Chronic alcoholism will typically be identified from the historical information and scene investigation. Clark, in an analysis of 500 deaths from Scotland where the subjects were defined as chronic alcoholics, used evidence from police reports, relatives and medical records (4). To these historical findings can be added the scene information, which is often chaotic, with evidence of large numbers of alcoholic beverages typically present at the residence.
The Autopsy and Chronic Alcoholism
Autopsy findings will also assist in determining whether a person is an alcoholic, with paradigm disorders such as alcoholic liver disease with or without cirrhosis (5), esophageal varices (6), acute or chronic pancreatitis (7–9), dilated cardiomyopathy (10,11), proximal myopathy (12), testicular atrophy (13), and brain pathology (14) being variably identified.
Studies have been conducted to provide objective evidence of chronic alcoholism at autopsy. These have included an analysis of the general state of hygiene, bruises, liver disease, blood ethanol, and other biochemical markers (15). Alcoholics were found to have more bruises than controls. Considerable research has also been conducted on biochemical markers in body fluids and hair. Use of carbohydrate-deficient transferrin has been most studied. Other markers have also been studied, but these are not in routine use (15–19).
Clark divided his deaths into five main categories: 1) acute alcohol intoxication, 2) trauma, 3) effects of chronic drinking, 4) incidental natural disease, and 5) obscure deaths (4). It is the latter category with which this review is primarily concerned, and the one that most often presents forensic pathologists with a diagnostic challenge.
Sudden Death and Chronic Alcoholism
Of the 500 deaths examined by Clark, 93 were from acute intoxication, 118 from trauma, 78 were alcohol related, 121 with incidental natural disease, and 51 cases were called obscure deaths, making up 10% of all deaths in the study. Thirty-nine were suicides, which was 8% of all deaths. Of the 118 trauma deaths, there were 29 fire deaths, 25 falls, 18 homicides, 13 hypothermia, 13 vehicular collisions, 11 drownings, and 9 miscellaneous deaths. Of the deaths ascribed to the effects of chronic drinking, the two most common pathologies were pneumonia and cirrhosis.
The 51 obscure deaths involved 32 cases with fatty liver. These people often had a history of feeling unwell the day or days before death, including vomiting, abdominal pain, and seizures. The blood ethanol in these cases was zero or very low. A second group included apparent aspiration of vomitus with a few having witnessed seizures. This series of Clark's was published in 1988 and does not make reference to alcoholic ketoacidosis; neither do studies on deaths in chronic alcoholics by Copeland and Randall, published in the 1980s nor a paper analysing deaths in chronic alcoholics in Denmark between 1986–1988 (20–22).
Intoxication
Although death from acute alcohol intoxication may occur in the chronic alcoholic, in many deaths the blood ethanol is below levels attributed to causing death from acute toxicity. In an analysis of 500 deaths in chronic alcoholics, Clark reported 93 deaths from acute alcohol intoxication (18.6%) (4). The alcohol range was 261 to 612 mg/dL, with a mean of 425 mg/dL. All those with a blood ethanol below 318 mg/dL had significant coexisting disease.
In a study involving 693 deaths from acute alcohol intoxication deaths and 825 deaths from chronic alcoholism, Jones and Holmgren found the mean concentration in male chronic alcoholics was 174 mg/dL and in women 157 mg/dL, compared with a mean concentration of 356 mg/dL in men dying from acute intoxication and 373 mg/dL in women dying of acute intoxication, with a 90% range of 250–510 mg/dL (23).
Alcoholic Ketoacidosis
The recognition that nondiabetic alcoholics may develop ketoacidosis was first reported in 1940 and has been increasingly recognized in the clinical literature from 1971 onwards (24–28). However, it was not until the 1990s that alcoholic ketoacidosis was reported in the forensic pathology literature (29–32).
The typical scenario presenting to the forensic pathologist is a chronic alcoholic who is found dead unexpectedly, sometimes with a history of being unwell with nausea and vomiting. Often the liver is fatty, but no obvious pathology is present to account for sudden death at the initial autopsy. The blood ethanol concentration is typically low or negative.
In periods of starvation, where there has also been a period of abstinence from alcohol and which may be exacerbated by vomiting, an accumulation of β-hydroxybutyrate occurs because fatty acids cannot be oxidized and a surplus of acetyl-Co-A develops. This is ultimately converted to acetoacetic acid, some of which is decarboxylated to acetone and some reduced to β-hydroxybutyrate with resulting ketoacidosis.
To identify alcoholic ketoacidosis at autopsy, ketone bodies need to be measured. These include β-hydroxybutyrate (1 mg/dL = 0.096 mmol/L), acetone (1 mg/dL = 0.172 mmol/L) and acetoacetate (1 mg/dL = 0.098 mmol/L). Normal β-hydroxybutyrate concentrations in living subjects have been recorded up to 6.77 mg/dL = 0.650 mmol/L and acetoacetate up to 0.226 mmol/L (0.0013 mg/dL) (33). One postmortem study showed mean levels of acetone plus acetoacetate was 0.037 mmol/L in controls, 0.052 mmol/L in alcohol abusers and 0.168 mmol/L in ketoacidotic deaths (29). Similarly β-hydroxybutyrate was lowest in non-alcoholics at 0.56 mg/dL (0.054 mmol/L), with alcohol abusers having a higher concentration of 0.75 mg/dL (0.072 mmol/L), but in ketoacidotic deaths was 3.65 mg/dL (0.351 mmol/L). Combining the values gave a mean ketone concentration of 0.109 mmol/L in controls, 0.152 mmol/L in alcohol abusers and 0.531 mmol/L in deaths from ketoacidosis. Vitreous and pericardial fluid can be used to measure ketone bodies as well as blood (34). One study showed that there was no evidence of postmortem formation or decomposition of β-hydroxybutyrate in blood (35). β-hydroxybutyrate in blood below 5.2 mg/dL (0.50 mmol/L) can be considered normal, between 5.2 – 26 mg/dL (0.50–2.50 mmol/L) elevated; and above 26 mg/dL (2.50 mmol/L) pathological. Clinical patients with alcoholic ketoacidosis have glucose levels that are usually in the normal range, but may be mildly elevated or subnormal (36).
At autopsy, the typical findings are the stigmata of alcohol-related disease, but an otherwise negative autopsy. One clue to the presence of ketoacidosis (whether alcoholic, diabetic or other causes) is the presence of subnuclear vacuolation of the renal tubules (sometimes called the Armanni-Ebstein lesion) (37–39).
Alcohol Withdrawal
After a period of abstinence, alcoholics begin to experience withdrawal symptoms. This may occur within a few hours but typically takes places two to three days after last drinking (40). Symptoms include sweating, anxiety, tremor, auditory and visual disturbances, agitation, nausea, headache and tactile disturbances, disorientation, and clouding of the sensorium (41). The most severe form of withdrawal is alcohol withdrawal delirium (AWD), often called delirium tremens or the “DTs.” Twenty percent of alcohol dependent individuals presenting to an inner city hospital developed AWD (42). While mortality rates of 15% have been reported in the past, modern treatment has reduced mortality with Ferguson and colleagues reporting an overall mortality of 2% in withdrawal and 8% in those who developed AWD (43,44). Predictors of death in patients with AWD include persistent hyperthermia and use of restraints (45).
Chronic Alcoholism and Seizures
Another postulated mechanism of death in chronic alcoholics is the development of seizures. Seizures may develop in alcoholics as alcohol-induced epilepsy or during withdrawal. Three to ten percent of alcohol dependent patients develop seizures, with 70% associated with withdrawal. Alcohol withdrawal seizures are typically self-limiting (46,47). In a review of seizures and deaths in alcoholics, Christoffersen could not find any well-documented deaths from alcohol related seizure (48).
Alcoholics are prone to falls and the development of craniocerebral trauma and the development of post-traumatic seizures (49). Chronic alcoholism is an independent risk factor for the development of post-traumatic epilepsy (50).
Cardiac Arrhythmias, Sudden Death, and the Chronic Alcoholic
In this issue, a separate paper details the interaction of alcohol and the heart (11). Cardiac arrhythmias may cause sudden death in chronic alcoholics. Cardiac dysfunction in patients with cirrhosis is well recognized (51). Cardiovascular disease is an important cause of sudden death in alcoholics. An analysis of sudden death in patients with ischemic heart disease found that sudden death was more common in heavy drinkers (more than six drinks daily) (52). With respect to sudden death and liver disease, alcoholics with histologically proven liver disease (greater than simple fatty liver) may have prolongation of the QT interval and an increased risk of sudden death (53). Prolongation of QT interval has been reported to resolve and cardiac function may improve following liver transplantation (54–56).
Sudden death may also be related to alcoholic cardiomyopathy and coexisting cardiovascular disorders including coronary artery atherosclerosis (11).
Conclusions
Sudden death is well recognized in the chronic alcoholic. At autopsy cases often present with nonspecific findings that do not account for death. In the past, fatty liver was often found at autopsy, but as one author stated, these deaths appeared obscure. A number of mechanisms causing death are now recognized, including alcoholic ketoacidosis, cardiac arrhythmias associated with alcoholic cardiomyopathy, and alcoholic liver disease. Alcoholics are at increased risk of developing seizures, but in acute withdrawal, these seizures are usually self-limiting. They are also at risk of post-traumatic epilepsy.
When alcoholic ketoacidosis is identified, it can be listed as the cause of death. When an investigation of the death of a chronic alcoholic is negative for a sudden cause of death and toxicological and biochemical analysis is negative, it is reasonable to list the cause of death as chronic alcoholism. The mechanism of death in these cases can reasonably be postulated to be due to cardiac arrhythmias.
Footnotes
The author, reviewers, editors, and publication staff do not report any relevant conflicts of interest.
