Liver disease due to alcohol consumption is becoming more frequent worldwide. In North America and Europe, it is the fifth leading cause of death in the young population and is a frequent cause of disability.1,2 In Mexico, liver cirrhosis due to alcohol consumption represents approximately 50% of cases, constituting a public health problem.3 A significant increase in alcohol consumption during the COVID-19 pandemic could have impacted the development of liver disease,4 with a substantial increase in Mexican women.

According to the Mexican National Survey of Addictions, ENCODAT 2016, the prevalence of excessive alcohol consumption increased by 5% (19.7–24.4%) and regular consumption increased up to 3 times, from 1.4 to 3.5%.5 The same ENCODAT survey showed that younger women drank more alcohol and started consuming it at earlier ages. Mellinger et al.6 found that liver disease due to alcohol consumption in women in the United States increased by 50% from 2009 to 2015, compared with 30% in men, over the same period. Grant et al.7 reported an 80% increase in heavy alcohol consumption in women, compared with 30% in men between 2001–2003 and 2012−2013. A similar pattern has been observed globally, with a two to four-fold increase in alcohol consumption among Japanese women.8

The increase in alcohol consumption in both sexes and at younger ages increases the toxic exposure time, which favors liver damage from alcohol-associated steatohepatitis and its progression to fibrosis, cirrhosis, and hepatocellular carcinoma (HCC).9 The complications of cirrhosis include portal hypertension with increased variceal hemorrhage risk; splanchnic vasodilatation, which induces activation of the renin-angiotensin-aldosterone system and development of ascites; hepatic encephalopathy; and liver cancer. Some patients develop alcoholic hepatitis (AH), with a 3-month mortality risk of 20–50%.10 The leading causes of AH hospital admission are variceal hemorrhage, aggregate infectious processes, grade III ascites, renal failure, and multiple organ dysfunction. Decompensation causes and mortality rates differ between sexes and are higher in women due to their greater susceptibility to alcohol-related liver damage.11–13

Our aim was to evaluate the quantity and duration of alcohol consumption and its relation to mortality in patients with alcohol-related cirrhosis during their first hospital admission.

We conducted a retrospective cohort study on patients hospitalized for the first time at the Gastroenterology and Hepatology Service of the Hospital General de Mexico "Dr. Eduardo Liceaga" (HGMEL) and diagnosed with alcohol-related cirrhosis between 2018 and 2021. The HGMEL Research Ethics Committee approved this study. Patients with a different cause of cirrhosis (chronic viral hepatitis, autoimmunity, and metabolic syndrome, among others) and patients over 70 years of age (with possible associated geriatric syndromes) were excluded. From the records of patients who met the selection criteria, we collected data regarding their survival status (yes/no), liver complications at hospital admission (digestive tract bleeding, jaundice, ascites, encephalopathy, alcoholic hepatitis, renal failure, and acute-on-chronic liver failure), together with liver disease severity scores (Child-Turcotte-Pugh and MELD scores). The onset of alcohol consumption and its duration until diagnosis were also collected.

From the routine questionnaire applied to patients at hospital admission, we obtained data on the usual type of alcoholic beverage consumed (fermented, distilled, or mixed) and the frequency of alcohol intake. We identified two patterns of alcohol consumption: excessive (>60 g/day for men and >50 g/day for women) and binge drinking (>60 g/occasion for men and >50 g/occasion for women).14,15 The average drink for both types of alcohol consumption was estimated by the formula: grams/day of alcohol = (amount consumed in milliliters) (alcohol content of the ingested drink) (0.8)/100.

We reviewed a total of 220 medical records of patients diagnosed with alcohol-related cirrhosis; we excluded 28 patients who did not meet the inclusion criteria. Ninety-six women (50%) and 96 men (50%) were finally included (n = 192) in the study.

Reaching statistical significance, age at chronic alcohol intake onset was 1.5 years lower in men than in women (16.5 vs 18 years of age, respectively) (p = 0.0001), excessive alcohol consumption was lower in women (140 g/day) than in men (275 g/day) (p = 0.0001), and binge drinking was lower in women (196 g/occasion) than in men (320 g/occasion) (p = 0.0001). The number of years of chronic alcohol consumption was also statistically significant and was 5.5 years lower in women than in men. Women had a higher frequency of complications than men (p = 0.0001) (Table 1).

There were no statistically significant differences between women and men, with respect to the MELD score at admission, age at first hospital admission (50 and 51 years), and age at cirrhosis diagnosis (Table 1).

Excessive consumption patterns were more frequent in men than in women (60 vs 51%, with no statistical association regarding drinking pattern (p = 0.191). In the excessive consumption and binge drinking patterns, there was an independent association with the type of drink, for women and for men. There was a higher percentage of distilled and fermented drink intake in women with excessive alcohol consumption (p = 0.001), whereas the percentage of distilled drink intake and binge drinking was higher in men. The percentage for mixed drink intake was high in both sexes (p = 0.005) (Table 2).

The total mortality rate during first hospital admission was 32.8% (n = 63), of which 61.9% (39) were women and 38.1% (24) were men.

Regarding liver disease severity, the Child-Turcotte-Pugh score in women was A: 9 (10%), B: 32 (33%), and C: 55 (57%). In men, the Child-Turcotte-Pugh score was A: 4 (4%), B: 25 (26%), and C: 67 (70%). The chi-square test did not reflect statistical significance: (2) = 3.96; p = 0.138.

Mean survival in women and men was 33.8 ± 1.6 years (95% CI 30.5–37.1) and 37.0 ± 1.2 years (95% CI 35.4–38.6), respectively (log-rank test [chi-square = 9.688; p = 0.002]).

The multivariate analysis conducted with a Cox proportional hazards regression produced an association between mortality and lower age at hospital admission; a 98% increase in mortality with excessive alcohol consumption, compared with binge drinking; a 6% increase in mortality with a higher MELD score; and a 4.2% increase in mortality for women, compared with men (Table 3 and Fig. 1).

Figure 1.

Kaplan–Meier survival curve at 50 months from 192 patient records (50% women and 50% men) with alcohol-related cirrhosis at first hospital admission.

Fig. 2 and Table 4 showed a lower survival curve for women than for men. The independent Cox proportional hazards regression for men and women produced an association between age and a higher mortality rate, in women and men (p = 0.0001), whereas only the MELD score was associated with mortality in women.

Figure 2.

Unadjusted Kaplan–Meier survival curves for the effect of the alcohol consumption pattern on mortality, separated for women and men.

Age at hospital admission was 15.6% lower in women, as a mortality factor for chronic alcohol consumption duration. The average age at death was 49.9 and 50.9 years for women and men, respectively. The MELD score was 4% higher in women, but the difference from men was not statistically significant. Grams of alcohol consumption per occasion, the number of symptom manifestations during hospitalization, consumption pattern, and drink type were not significantly associated with mortality and the length of time of continued chronic alcohol consumption (Table 3).

Alcohol consumption has become a significant concern, both in Mexico and worldwide. Increased alcohol consumption in women, particularly young women, is a public health problem.15 In our results, we found that women started drinking alcohol two years later than men. This behavior in women could have a cultural and social explanation or be due to a higher prevalence of anxiety and depression or to social pressure from peers to drink alcoholic beverages. In our study group, the mean amount of alcohol consumed by women was significantly less than that of men (160 g, when compared between groups, and 290 g per occasion, respectively). However, this was not reflected in the consumption pattern, in which no statistical difference was found throughout the time of active consumption. Although patients required hospitalization, it did not contribute to mortality. It is noteworthy that the different patterns of alcohol consumption and the type of beverage (fermented or distilled) showed no significant difference, when compared between groups. Conceivably, the accuracy of answering the questionnaire could be biased. Nevertheless, in other studies, alcoholic beverage type and consumption pattern have shown distinct toxic liver effects and are related to mortality.16 In our cohort, mortality in women was higher, at 40%, compared with 25% in men.

The survival curve showed a disadvantage for women with chronic excessive alcohol consumption. Age was a significant factor for mortality in both sexes and was less protective for women. The lower alcohol consumption and shorter duration in women, compared with men, suggest a greater predisposition to liver damage in women, with a 3-year difference in mean survival, in favor of men. This result was also reflected in increased complications at hospital admission, and the MELD score was only significant in women as a factor associated with mortality. Alcohol-related mortality is known to be potentially higher in women and can be explained by the effect of sex hormones on the oxidative and metabolic pathways and the differential transcription of genes, resulting in a greater susceptibility to alcohol-induced toxicity,17,18 as well as by the absence of a gastric first pathway of alcohol metabolism.

Other factors that negatively impact women who consume alcohol are slower gastric emptying time, lower lean body mass that induces greater plasma volume of circulating alcohol, and low activity of the cytochrome P450 enzyme, which generates a difference in the rate of alcohol metabolism.19,20 Paradoxically, however, progesterone is associated with faster alcohol elimination rates in women.21,22

Chronic alcohol consumption (in two patterns of alcohol consumption: daily or per occasion) increases dysbiosis and bacterial overgrowth by increasing the intestinal bacterial load. In addition, it modifies the composition of the existing microbiota, with increased intestinal permeability and translocation of pathogen-associated molecular patterns (PAMPs) derived from the intestinal microbiome and bacterial DNA, activating the innate and adaptive immune response, thus causing an excessive synthesis of pro-inflammatory cytokines and chemokines. In women, this response is much greater at lower blood alcohol concentrations.23,24

The limitations of our study include the fact that inquiring about alcohol consumption is complicated by sociocultural taboos, and women may have a more significant response bias. Other factors, such as the coexistence of women with eating disorders (e.g., bulimia or anorexia) was not considered, and said conditions could have a more significant harmful effect due to poor food intake in those patients.

Mortality due to alcohol-related liver damage is higher in women than in men; it is associated with a lower amount of g/day alcohol consumption, a shorter period of continued chronic consumption, and a higher age at the onset of drinking. Mortality in women in the coming years could increase because there is evidence of a prevalent increase in female alcohol consumption.

No financial support was received in relation to this study/article.