Commentary: physical activity and NAFLD - cause or effect? Authors' reply


Drs Ahmed and Forrest provide an important contribution to the discussion about physical activity and NAFLD.[1, 2] They express concern that it is important to reiterate the fact that the data reported in this publication are derived from a regularly conducted population survey, National Health and Nutrition Examination Survey (NHANES).[1, 2] Advantages of this type of study are that one is able to access a large sample with a mix of diagnoses and from a broad geographical and demographic distribution. This limits the ability to obtain additional information from what is routinely collected. In particular, the survey does not include liver biopsy or scan, but relied upon the fatty liver index (FLI). While this may not be optimal, it has been widely used, is reliable and has been shown to be a reliable predictor of hepatic steatosis in the general population.[3] FLI was used consistently and hence does serve to define the group which we suspect strongly as having NAFLD.

The study data are derived from survey data and hence do not enable us to comment on whether low activity is a cause of NAFLD, or the reverse, that NAFLD causes low activity. The editorial comment suggests that this would be useful information to obtain. We agree with this view.

We are grateful to Drs Ahmed and Forrest for pointing out the importance of physical activity. Data are overwhelmingly convincing that physical activity plays an important role in disease modulation and possibly prevention (examples: breast cancer, colon cancer, prostate, endometrial and lung cancer[4]). Data support its usefulness in secondary prevention in breast cancer survivors,[5] cardiovascular diseases and diabetes mellitus.[6] On the other hand, its effect on weight loss is modest, at best.[7] Calorie restriction is the most effective means of reducing weight. Physical activity is effective in maintaining weight loss.[8, 9]

The use of an objective activity monitor to assess both the quantity and the intensity of physical activity is an important step forward. In fact, it is suggested that cardiovascular disease modulation requires 150 min/week of exercise.[10] Health benefits are directly proportional to the intensity of activity in modification of hepatic metabolism,[11] but not necessarily in improving HbA1c.[12] Therefore, it is important to tailor the intervention to the correct population. It is equally important to ascertain the details about type, duration and intensity of exercise in which the study group participated.

We do not agree with the editorial comment that 10 h of daily activity recording is ‘only a small portion of their activity’. Most people record sleeping 7 h/day and more on the weekend. If one adds time for arising, bathing, preparing for bed etc., 10 h of recorded activity is likely to represent a substantial portion of the day's activity.

Comparing the diagnostic sub-groups of individuals with chronic diseases [NAFLD, diabetes mellitus (DM), both or neither] sheds light on the likely behaviours of the group. The findings, not entirely surprising, support the view that people with DM and NAFLD are less active than those without. In particular, the combination of DM and NAFLD is associated with the lowest level of activity.

Nonetheless, the editorial raises an important challenge: we need to explore the causes of inactivity. The significant contributors are likely to be multi-factorial, including habitual behaviours, abnormalities of glucose uptake/metabolism, energy production and level of inflammation. They are also likely to be difficult to treat.


The authors' declarations of personal and financial interests are unchanged from those in the original article.2