Although heavy alcohol intake is known to be one of the most common causative factors of liver disease, pancreatitis, upper gastrointestinal and neurological disorders, the influence of the drinking pattern is largely unknown. The study investigated the relationship of alcohol-related medical disorders in alcoholics and their drinking pattern. Two hundred and forty-one chronic alcoholics were referred consecutively for detoxification and their drinking pattern was sufficient for them to be included in this study. History of alcohol abuse as well as drinking behaviour in the last 6 months were assessed by a semi-structured interview. Findings included intensive clinical examination with abdominal ultrasound in most subjects. Heavy drinking with frequent inebriation was most often found in our sample (44.4%), whereas continuous heavy alcohol consumption without intoxication (33.6%), and an episodic drinking style (22.0%) were less frequent. The heavy drinkers suffered more often from pancreatitis, oesophageal varices, polyneuropathy or erectile dysfunction than episodic drinkers. They also showed more upper gastrointestinal disorders, although the estimated life-time alcohol intake was comparable to continuous drinkers. No difference relating to withdrawal delirium or seizures could be found between the groups of alcoholics. Frequent heavy drinkers showed a trend to more alcohol-related medical disorders than alcoholics with a different drinking pattern, although they were younger and their estimated life-time alcohol intake was comparable to that of continuous drinkers. Thus, the drinking pattern, particularly frequent inebriation, has an influence on the occurrence of alcohol-related disorders. Definition of an alcoholism.
Alcoholism, one of the most common chronic disorders in the western world, causes or promotes a plethora of diseases and injuries. The social health care costs of harmful alcohol consumption are enormous ( Rice et al., 1986 ). Since ethanol is a hydrophilic and lipophilic substance, it may harm nearly every organ, but only some disorders are thought to be related to heavy alcohol intake (so-called alcohol-related disorders, such as liver disease, pancreatitis, upper gastrointestinal, and neurological disorders, e.g. polyneuropathy), ( Charness et al., 1989 , Lieber 1998 , Piette et al., 1998 ). However, until recently, few studies were available which focused on the influence of drinking behaviour on the frequency of alcohol-related disorders. For example, there is little information on whether drinkers with frequent intoxications suffer more often from withdrawal complications, such as delirium or seizures, than continuous drinkers or vice versa. Furthermore, the high amounts of alcohol drunk by frequently intoxicated alcoholics per day may injure the gastrointestinal tract directly and may also harm the liver as a detoxifying organ. Therefore, this study was aimed at evaluating the relationship between drinking pattern and occurrence of medical complications due to alcohol misuse.
METHODS AND SAMPLE
The sample was collected from 322 chronic alcoholics referred consecutively for detoxification to our department. History of alcohol misuse including drinking pattern and physical illnesses, was assessed by a semi-structured interview according to the documentation standards of the German Society for Addiction Research and Therapy ( Deutsche Gesellschaft für Suchtforschung und Suchttherapie, 1991 ) containing 262 items providing data on the socio-economic status, drinking history and behaviour (frequency, amount, etc.), drug abuse, history of somatic and psychiatric disorders, etc. The drinking pattern was classified into three categories according to frequency of drinking (during the previous 6 months) and amount of alcohol intake: (1) continuous drinkers = (almost) daily alcohol consumption without binges, (2) frequent heavy drinkers = frequent alcohol consumption (more than 3 days/week) with frequent intoxication (more than one/week), (3) episodic drinkers = less frequent, irregular alcohol consumption with longer (>, 5 days) sober periods, and some binges (less than one/week).
It was possible to categorize the drinking pattern in 241 patients (74.8%), as the remaining patients showed irregular drinking behaviour. Thus, 64 females (mean age ± SD: 43.8 ± 8.8 years, mean duration of harmful drinking: 10.4 ± 6.8 years) and 177 males (mean age: 41.0 ± 9.9 years, mean duration of harmful drinking: 11.6 ± 8.7 years) were included in this study. All subjects underwent comprehensive clinical examination including laboratory tests (at admission and 3 weeks after admission) and tests for viral hepatitis A, B, and C. Subjects with viral hepatitis were excluded. Abdominal ultrasound was performed in 194 cases.
The life-time alcohol intake was estimated as the product of the drinking frequency, the mean alcohol intake/drinking day, duration of harmful alcohol intake, and a ‘tolerance factor'. This ‘tolerance factor' was estimated as the reciprocal of the ratio of reported increase of alcohol intake at the onset of harmful drinking to the index drinking period. Longer abstinence periods (>, 3 months) were taken into consideration when estimating the duration of harmful drinking. Alcohol intake was calculated in g/kg. All statistical calculations were performed using the SPSS-PC program package (version 7.5).
Definition of alcoholism by consumption
Heavy drinking with frequent intoxication was found most often in our sample (44.4%), whereas continuous alcohol consumption (33.6%) and an episodic drinking style (22%) were less frequent. The proportion of females was significantly lower in the group of frequent heavy drinkers (15.0 vs 33.3% in continuous and 39.6% in episodic drinkers, χ
= 13.9, d.f. = 2, P = 0.0009). The frequent heavy drinkers were significantly younger (37.6 ± 9.4 years) than the continuous drinkers (46.5 ± 8.5) and the episodic drinkers (42.8 ± 8.3 years) (Scheffé-test: P <, 0.05).
Severity of alcohol dependence
The frequent heavy drinkers showed more severe alcoholism according to ICD-10 criteria ( World Health Organization, 1992 ) for alcohol dependence than the other groups. Nearly all of them (91.6%) qualified for at least three of the six ICD criteria, so that they were diagnosed as alcohol-dependent, whereas only 75% of the continuous drinkers and about 60% of the episodic drinkers fulfilled ICD-10 criteria for alcohol dependence. Each ICD-10 criterion was fulfilled by frequent heavy drinkers more often than by other groups, particularly an impaired capacity to control drinking (66.4 vs 16% in continuous and 26.4% in episodic drinkers, χ
36.6, d.f. = 2, P <, 0.0001), and preoccupation with drinking (74.8 vs 49.4% and 41.5%, χ
The alcohol history (Table 1 ). The continuous and the frequent heavy drinkers showed a history of pancreatitis and oesophageal varices more often than episodic drinkers. Furthermore, the frequent heavy drinkers' group suffered from chronic gastritis and gastrointestinal bleeding more frequently. Polyneuropathy as well as erectile dysfunction occurred more often in continuous and frequent heavy drinkers. The rates of withdrawal delirium or seizures were no different between the groups. In summary, the frequent heavy drinkers tended to show a higher number of alcohol-related disorders than episodic drinkers, but no more than continuous drinkers. In particular, more upper gastrointestinal and neurological disorders were detected in frequent heavy drinkers. Furthermore, they required emergency treatment and had a history of severe brain trauma with unconsciousness more often. They also attempted suicide more often than continuous drinkers.
In order to evaluate the impact of cumulative alcohol consumption on the occurrence of alcohol-related disorders, the life-time alcohol intake was estimated (see the Methods and sample section). The estimated average life-time alcohol intake was similar in continuous drinkers (8.8 ± 13.1 kg alcohol/kg body weight) and in frequent heavy drinkers (8.5 ± 9.1 kg/kg), whereas that of episodic drinkers was significantly lower (3.3 ± 7.1 kg/kg). Female alcoholics drank significantly less alcohol (4.4 ± 5.6 kg/kg body weight) than male alcoholics (8.5 ± 11.6 kg/kg) (U-test, P = 0.0062).
The number of alcohol-related disorders was strongly related to life-time alcohol intake (Table 3