Waist-to-hip ratio but not body mass index predicts liver cirrhosis in women

Abstract

Background: Being overweight can lead to fatty liver and end-stage liver disease. In men, higher body mass index is associated with higher risk of developing liver cirrhosis. The extent of association between overweight and liver cirrhosis in women is not fully elucidated.

Aims: This study aimed to investigate the association between overweight and liver cirrhosis in women, taking into account different measures of adipose tissue distribution.

Methods: A cohort of 1462 middle-aged women was followed over 40 years. Cases of liver cirrhosis were identified by linkage to Hospital Discharge and Death Certificate registries. The hazard ratios for different anthropometric measures and liver cirrhosis were obtained by Cox proportional hazard regression, using propensity score methods to adjust for important confounders.

Results: During 48,062 person-years of follow-up, 11 cases of liver cirrhosis were identified. The incidence rate in women with waist-to-hip ratio ≥ 0.8 was 131.8 (48.1–287.0), compared to 12.0 (3.9–28.1) in women with a lower ratio. A waist-to-hip ratio ≥ 0.8 was associated with an increased risk of liver cirrhosis, the hazard ratio being 5.8 (95% confidence interval 1.6–21.4). No association between body mass index and liver cirrhosis was found and the hazard ratio for body mass index >25 was 1.8 (0.5–5.8).

Conclusion: In women, an unfavorable adipose tissue distribution is more important for development of liver cirrhosis than total body fat per se. When assessing the risk for development of liver cirrhosis in women, waist-to-hip ratio is a better predictor than body mass index.

Keywords:

• Overweight
• obesity
• liver cirrhosis
• women
• waist-to-hip ratio
• body mass index

Acknowledgements

The authors thank the women of ‘The Population Study of Women in Gothenburg’ for their participation in the study which allowed answering research questions in many fields of medicine and still creates new insights in epidemiologic questions.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by Faculty funding only.