If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Hôpital Pitié-Salpêtrière, Institute of Cardiometabolism and Nutrition, Assistance Publique-Hôpitaux de Paris, Université Pierre et Marie Curie, Paris, France
Liver disease is a common cause of morbidity and mortality worldwide [1]. Chronic liver diseases can in some individuals lead to cirrhosis, which is associated with an increased risk of mortality. The prognosis is particularly unfavourable if decompensated liver disease occurs [2].
]. Fatty liver occurs in 20–30% of the general adult population, in strong association with obesity and type 2 diabetes, and may progress to non-alcoholic steatohepatitis (NASH), cirrhosis and eventually hepatocellular carcinoma (HCC). While these facts are now well established, the current epidemics of obesity is shifting towards an onset earlier in life: between 1980 and 2013 overweight and obesity have increased by 47% in children. This opens a large field of investigation as we try to anticipate how this earlier advent of overweight will impact significant liver disease.
In the current issue of the Journal of Hepatology, Hagström et al. present an insightful study showing that overweight in late adolescence is a significant predictor of severe liver disease later in life. The authors studied a general population cohort of 44,248 men aged 18–20 years that attended between 1969 and 1970 an outpatient visit for conscription into military service in Sweden [
]. They tested the association between BMI and a diagnosis of decompensated liver disease, cirrhosis, or liver-related death as per specific coding in remarkably well-kept and exhaustive National Patient and Causes of Death Registries. After a long, mean follow-up of almost 38 years, 393 patients developed end-stage liver disease (ESLD) outcomes of which 213 died during follow-up. The main finding is that BMI in late adolescence is a strong and independent predictor of ESLD and mortality later in life: after adjustment for a variety of available confounders (alcohol, narcotics, smoking, high blood pressure, physical activity, and cognitive ability at time of conscription), liver-related outcomes were increased by 5% for each unit increase in BMI; being overweight at time of conscription increased the risk of outcomes by 64% compared to a low-normal range BMI (18.5–22.5) taken as reference. Since the authors measured BMI at ages 18–20, their findings basically reflect the impact of overweight or obesity acquired during childhood and adolescence. Regretfully, the incidence of ESLD per person/year of follow-up in obese vs. non obese individuals was not computed, therefore the magnitude of the risk cannot be compared with that of other well-known causes of ESLD.
The strengths of this report are manifold: the large, unbiased (except for male sex), healthy general population cohort within a narrow age window at inclusion; the longitudinal acquisition of data with a long follow-up; the exceptionally high rate of subject attrition in the study, a tribute to the Scandinavian centralized health-care system; and, finally, the choice of hard outcomes such as ESLD, further confirmed by the high liver-related mortality rate mentioned above. There are however, methodological issues inherent to this study design and some further considerations raised by these results. Because the proportion of overweight and obesity was rather low in this late 60s’ cohort, the power of the study to detect obesity-related events was limited. This might explain the absence of a clear-cut dose response with weight categories (Table 3). Also, since outcomes were ESLD-related and the ascertainment was based on hospital codes, asymptomatic, well compensated cirrhosis (a risk factor for liver cancer and future cirrhotic morbid-mortality) could not be captured. Therefore, the study clearly underestimates the association between overweight/obesity and potential for severe liver disease.
But is obesity itself to blame? Important confounding factors such as genetic factors or type of diet could modify this relationship. The PNPLA3 genetic variant is associated with progressive NASH and fibrosis in obese children and adolescents [
European Association for the Study of the Liver, European Association for the Study of Diabetes, European Association for the Study of Obesity. EASL-EASD-EASO Clinical Practice Guidelines for the management of non-alcoholic fatty liver disease.
]. Diet, particularly fructose or transfats could promote steatohepatitis and fibrosis progression, independent of total body fat. Other obesity-associated comorbidities could go against a direct role of overweight in liver damage. Unfortunately, Hagström et al. derive their analysis from administrative data, without available information on the clinical phenotype of their cohort at the end of the observation period. Liver disease in overweight/obesity may be largely mediated through obesity-related comorbidities such as hypertension and diabetes. Arterial hypertension seems to increase the rate of fibrosis progression [
], and diabetes is a risk factor for advanced or progressive NASH. The possibility that diabetes might be the link is not ruled out in the present setting. Overweight or obesity are a risk factor for type 2 diabetes at any age, including during childhood; diabetes carries a high risk of cirrhosis [
While this study does establish that the obesity-related risk for future severe liver disease exists early in life, the type and magnitude of this risk needs further refinement. In the current study, data collection was performed at baseline only, so changes in body weight or other risk factors during follow-up were not recorded. Alcohol and smoking were stronger risk factors than overweight/obesity per se and major modifications in exposure to these risk factors can be anticipated with increasing age, over a long follow-up. Although it can be argued that beginning to smoke later during adult life is uncommon (most smokers start smoking during late adolescence), alcohol consumption can frequently increase during adulthood. Also, fat distribution in this cohort of young conscripts is unknown. In overweight individuals, truncal fat is better correlated with cardiometabolic risk than total fat-mass per se [
Visceral fat accumulation is an independent risk factor for hepatocellular carcinoma recurrence after curative treatment in patients with suspected NASH.
], then “obesity” will only imperfectly capture the risk of advanced liver disease; in contrast fat distribution, if altered during adulthood, might have a substantial impact. Along the same lines, cardiovascular fitness independently predicted subsequent ESLD. This is a remarkable finding corroborating a large body of data describing negative associations with liver steatosis. But here again, cardiovascular fitness is highly likely to change during follow-up, as a reflection of adopted lifestyle.
For the same reasons of study design and data collection, information about the weight trajectory in these participants is missing. This, however, is not expected to drastically change the results presented by Hagström et al. Obviously, many adult, obese individuals may not have been obese in late adolescence. In an observational study of Italian obese subjects entering weight loss programs, 41% of men (median age of 46) reported to be normal weight at age 20, whereas 39% were overweight and only 20% were obese at the time of their military service [
], and weight gain was positively related to mortality. Notably, early adulthood weight gain carried a higher risk for mortality than weight gain in late adulthood. Nonetheless, the point of the study by Hagström et al. was to identify the role of childhood overweight/obesity in the development of ESLD later in life. If anything, later weight gain in their non-overweight 18–20 year old conscripts blunted the effect of early obesity on hepatic outcomes later in life. Equally important, the weight trajectory does not seem to be bi-directional: in a population-based risk factor surveillance program, 32% of normal weight men (median age, 31) moved into the overweight class over 10 years, and 19% moved from overweight to obesity [
]; however only 9.5% of the initial overweight or obese participants reduced their body weight into the normal range.
Unfortunately, the report by Hagström et al. could not provide any information on HCC, a dreadful complication of chronic liver disease that could be influenced by longstanding obesity [
]. There are some reports linking obesity early in life to the risk of HCC arising decades later. A study in Copenhagen schoolchildren aged 7–13, showed that childhood obesity increased the risk of HCC 30 years later [
]. This association was independent of diabetes. However, this study was based on patient recall of their weight in a distant past, an imperfect method prone to recall bias. Had the current Swedish report been able to collect meaningful data on HCC as well as on any other cancer not related to obesity [
], the study would have gained in internal validity.
What are the implications of the findings by Hagström et al.? Chronic liver diseases progress slowly and cirrhosis typically takes decades to occur. In diseases such as NASH or chronic hepatitis C, advanced fibrosis and cirrhosis are exceedingly rare in young individuals and age higher than 50 is one of the strongest profibrogenic factors [
]. In light of these well established facts, childhood or adolescence overweight/obesity may not raise sufficient concern, especially among adult hepatologists. Many practitioners may overestimate the ability of overweight children to revert body weight gain during adolescence or early adulthood. The current study shows that even at a young age obesity matters, and that it leads to an increased risk of severe hepatic complications decades later (Fig. 1). This gives additional reason to fight obesity early in life and further credence to the paradigm that excess weight (and related comorbidities) drives significant hepatic morbidity and mortality. It is not yet entirely clear how the interaction between early obesity and subsequent liver disease unfolds, given the strong impact of organ senescence on fibrogenic processes. Does obesity at an early age hasten the occurrence of ESLD or does it simply increase the risk of occurrence at an advanced age? Currently, advanced cirrhosis and HCC occur relatively late in life in the 7th or 8th decade: mean age at diagnosis of HCC ranges from 68 to 72 years in reports from Italy, England and Japan [
]. If obesity early in life precipitates the occurrence of ESLD, including HCC, at a younger age rather than simply increasing the risk, then this will translate into a significant additional societal burden. In the report by Hassan et al., obesity in the mid 20s hastened by 4 years the occurrence of HCC (vs. normal weight in the mid 20s) while obesity in the mid 30s was associated with a diagnosis of HCC 5 years earlier [
]. Interestingly, in the study by Hagström et al. ESLD occurred after a mean 24.7 years of follow-up, that is around age 45 in this population. This is an early occurrence for ESLD, supporting the dire hypothesis of an increased risk at a younger age. The consequences in terms of lost years of life and productivity could be considerable. If the epidemics of obesity in childhood and adolescence translate into advanced NASH later in life, severe NASH will no longer be a disease of the elderly.
Fig. 1Descriptive diagram showing the interaction between genes and lifestyle in the development of excessive fat accumulation and liver-related outcomes in the course of life. For the sake of clarity, all cases are expected to start with normal body fat, but in a few cases in utero changes may also occur (babies born either small or large for gestational age), at high risk of future obesity or diabetes. Subjects may remain normal weight or minimally overweight throughout their life (green line), and will never intercept the level of liver disease (light orange area) or type 2 diabetes. Other individuals (brown line) will develop overweight or obesity during adulthood; they will intercept the level of liver disease and will develop NAFLD, at various stages of disease severity along the years (from pure fatty liver, to NASH, fibrosis, cirrhosis). The occurrence of type 2 diabetes will lower the threshold of their liver disease, as it lowers the threshold of liver-related death (dark orange area), in a time-dependent manner. In some cases, overweight or obesity might occur in early adulthood (red line), as in cases presented in the study by Hagström et al. In these cases, NASH is likely to occur early, and the threshold of liver-related death is reached at a relatively young age. Finally, in a few cases (blue line) overweight and obesity is present since childhood. In these cases, liver disease and liver-related death are anticipated further, with a remarkable reduction of life expectancy.
In conclusion, the deleterious impact of obesity on liver health should be taken seriously. Clinicians involved in the battle against obesity are well aware of the difficulties in promoting weight loss in adults, with only a minority of cases being able to maintain weight loss in the absence of surgical interventions. The recent European non-alcoholic fatty liver disease (NAFLD) guidelines include obesity control as a priority to reduce the burden of NAFLD in the community [
European Association for the Study of the Liver, European Association for the Study of Diabetes, European Association for the Study of Obesity. EASL-EASD-EASO Clinical Practice Guidelines for the management of non-alcoholic fatty liver disease.
], but obesity control does not depend on hepatologists. What is needed are strategies developed at the population level to control childhood and early adulthood obesity hoping that we can reduce the cumulative incidence of severe liver disease later into adulthood.
Conflict of interest
The authors who have taken part in this study declared that they do not have anything to disclose regarding funding or conflict of interest with respect to this manuscript.
References
Kelly T.
Yang W.
Chen C.S.
Reynolds K.
He J.
Global burden of obesity in 2005 and projections to 2030.
European Association for the Study of the Liver, European Association for the Study of Diabetes, European Association for the Study of Obesity. EASL-EASD-EASO Clinical Practice Guidelines for the management of non-alcoholic fatty liver disease.
Visceral fat accumulation is an independent risk factor for hepatocellular carcinoma recurrence after curative treatment in patients with suspected NASH.
30207-0&id=gr1.jpg){kind=link}