Changes in Hepatobiliary Enzyme Abnormality After the Great East Japan Earthquake: The Fukushima Health Management Survey

Although the incidence of hepatobiliary enzyme abnormality increased immediately after the Great East Japan Earthquake and subsequent Fukushima Daiichi Nuclear Power Plant accident, longer-term trends remain unclear. The aims of this study were to determine longer-term trends in hepatobiliary enzyme abnormality and to elucidate lifestyle factors associated with such changes among residents of a nuclear-disaster-affected area. This longitudinal survey enrolled 20,395 adults living in the vicinity of Fukushima Daiichi Nuclear Power Plant. Data were obtained from the records of annual health checkups of adults aged ≥40 years between 2011 and 2012. Follow-up examinations were conducted from June 2013 to March 2014. Associations were assessed between changes in hepatobiliary enzyme abnormality immediately and 3–4 years after the disaster and lifestyle factors. The overall prevalence of hepatobiliary enzyme abnormality significantly decreased over the study period, from 29.9% to 27.1%. Multivariate logistic regression analysis revealed significant associations between improved hepatobiliary enzyme abnormality and improvements in daily physical activity and frequency of breakfast consumption. The results suggest that improvements in daily physical activity and frequency of breakfast consumption significantly reduced the incidence of hepatobiliary enzyme abnormality 3–4 years after the Great East Japan Earthquake and Fukushima Daiichi Nuclear Power Plant accident.

The early results of the FHMS have already shown increases in body weight and higher incidences of diabetes, dyslipidemia, atrial fibrillation, hypertension, renal dysfunction and metabolic syndrome among residents of the disaster-affected area [2][3][4][5][6][7][8] . According to the FHMS, a higher incidence of hepatobiliary enzyme abnormality was also observed immediately after the Great East Japan Earthquake, from 2011-2012 9 . We previously reported finding an association between increases in the incidence of hepatobiliary enzyme abnormality and evacuation due to the Fukushima Daiichi Nuclear Power Plant disaster 10 .
Although more than 5 years have passed since the disaster, more than 90,000 residents of Fukushima Prefecture have yet to return to their homes. Previous assessments of hepatobiliary enzyme abnormality using FHMS data were based on comparisons of incidences before and after the disaster 2-10 , and thus the results reflected relatively short-term effects; however, longer-term effects of the disaster on and lifestyle factors associated with hepatobiliary enzyme abnormality unclear.
Therefore, the objectives of the present study were to determine longer-term trends in hepatobiliary enzyme abnormality and to elucidate lifestyle factors associated with such changes among residents of Fukushima Prefecture 3-4 years after the Great East Japan Earthquake.
The distribution of the serum aspartate aminotransferase (AST) levels is shown in Fig. 1. AST levels were mostly mild, with about 98% of patients having AST levels less than 62 U/L. Only small numbers of participants were having AST levels higher than 61 U/L and the distribution increased in moderate/heavy drinkers (Fig. 1d). The overall proportion of AST levels more than 61 U/L significantly decreased, from 1.6% in 2011-2012 to 1.2% in 2013-2014 (p = 0.005), and among non-and light drinker groups (non-drinkers, 1.4% to 1.0%; and light drinkers, 1.2% to 0.8%). The distributions of serum alanine aminotransferase (ALT) and gamma-glutamyl transpeptidase (γ-GTP) levels are shown in Figs 2 and 3, respectively. These distributions were similar to that of AST levels. The overall proportion of ALT levels higher than 61 U/L significantly decreased, from 3.0% in 2011-2012 to 2.3% in 2013-2014 (p < 0.001), and among non-and light drinker groups (non-drinkers, 2.6% to 1.9%; and light drinkers, 2.6% to 2.0%). The overall proportion of γ-GTP levels more than 101 U/L significantly decreased, from 4.8% in 2011-2012 to 4.1% in 2013-2014 (p = 0.002), and among light and moderate/severe drinker groups (light drinkers, 2.7% to 2.2%; and moderate/heavy drinkers, 13.1% to 11.6%). The AST to ALT ratio in overall significantly increased, from 1.22 in 2011-2012 to 1.23 in 2013-2014 (p < 0.001), and among all three groups (non-drinkers, 1.25 to 1.27; light drinkers, 1.22 to 1.23; and moderate/heavy drinkers, 1.18% to 1.19). Moreover, the proportion of the AST to ALT ratio over 0.87 significantly increased in overall and all three groups (Supplement Table 1).  After the disaster, 1,648 (11.5%) of 14,288 participants without hepatobiliary enzyme abnormality in 2011-2012 were defined as having hepatobiliary enzyme abnormality in 2013-2014. The crude incidence rates (per 1000 person-years) of hepatobiliary enzyme abnormality in non-, light, and moderate/heavy drinkers were 35.7,  42.5, and 74.4, respectively. Risk factors of hepatobiliary enzyme abnormality in all three groups were male sex, increased BMI, and evacuation (Table 2). Other, risk factors of moderate hepatobiliary enzyme abnormality in all three groups were increased age, increased BMI, and evacuation (Supplement Table 2).

Non-drinkers
Multivariate logistic regression was used to investigate lifestyle factors associated with improvements in hepatobiliary enzyme abnormality among 18,070 participants (Table 3). After adjusting for age, sex, BMI, smoking status, evacuation and alcohol intake at baseline, significant associations were found between increases in daily physical activity and improved hepatobiliary enzyme abnormality. In addition, overall, more frequent breakfast  consumption was significantly associated with improved hepatobiliary enzyme abnormality. Increased daily physical activity had also significant association with moderate hepatobiliary enzyme abnormality in non-drinkers (Supplement Table 3).

Discussion
This study investigated trends associated with hepatobiliary enzyme abnormality after the Great East Japan Earthquake and subsequent Fukushima Daiichi Nuclear Power Plant accident. The results showed a decreased incidence of hepatobiliary enzyme abnormality 3-4 years after compared with immediately following the disaster for the first time. Male sex, increased BMI, and evacuation were identified as risk factors for the development of hepatobiliary enzyme abnormality. Conversely, increased levels of daily physical activity and frequency of breakfast consumption were lifestyle factors shown to be significantly associated with improved hepatobiliary enzyme abnormality. Notably, increased levels of daily physical activity was also significantly associated with improved moderate hepatobiliary enzyme abnormality in non-drinkers. Previous cohort studies have reported finding elevated liver enzymes and an increased incidence of hepatobiliary enzyme abnormality after disasters [9][10][11] . Based on a longitudinal analysis, we recently reported that hepatobiliary enzyme abnormality was associated with evacuation after a disaster 10 . That study focused on the period before and immediately after a disaster, and therefore reflected the direct effects of a disaster on hepatobiliary enzyme abnormality. However, many evacuees have yet to return to their homes, and this likely affects their lifestyles. In the present study, we found that evacuation and increased BMI were risk factors for hepatobiliary enzyme abnormality. In addition, the decreased incidence of hepatobiliary enzyme abnormality among residents was a newly identified trend.
Metabolic syndrome and BMI are closely linked to hepatobiliary enzyme abnormality [12][13][14][15] . The incidence of both overweight and hepatobiliary enzyme abnormality decreased among non-drinkers. This finding suggests an association between increased BMI and hepatobiliary enzyme abnormality among non-drinkers. Obesity has been shown not only to induce fatty liver, but also to increase the prevalence of alcoholic liver disease 16 . Therefore, decreases in BMI may reduce the incidence of hepatobiliary enzyme abnormality, regardless of drinking status. In addition to hypertension and dyslipidemia, obesity is one of leading causes of metabolic syndrome. Interestingly, in this study, we found that the incidences of hypertension, dyslipidemia, and diabetes increased despite decreases in overweight. Although the precise reasons for this finding remain unclear, the incidence of these diseases may have been overestimated as a result of drug continuation. Another possible reason could have been differences in disease mechanisms or the effect of overweight on hepatobiliary enzyme abnormality and those associated diseases.
Lifestyles were dramatically altered after the Great East Japan Earthquake and subsequent Fukushima Daiichi Nuclear Power Plant accident. To our knowledge, this is the first study to investigate the effects of various lifestyle factors on hepatobiliary enzyme abnormality. Physical activity or exercise has been reported to be associated with hepatobiliary enzyme abnormality [17][18][19] . The results of the present study support an association between increases in physical activity and improvements in hepatobiliary enzyme abnormality even in unusual life after the Great East Japan Earthquake and subsequent Fukushima Daiichi Nuclear Power Plant accident. Moreover, present study showed an association between increases in physical activity and improvements in moderate hepatobiliary enzyme abnormality in non-drinkers. In fact, the proportion of residents who reported walking or exercising more than 1 hour per day increased form 33.4% in 2011-2012 to 35.0% in 2013-2014. This finding  may reflect effective efforts to promote exercise by municipal governments. Although no association was found between improved sleep and hepatobiliary enzyme abnormality in the present study, short sleep duration has been reported to be associated with metabolic syndrome and health outcomes 20,21 . Therefore, a more detailed analysis of sleep disorder may reveal associations with hepatobiliary enzyme abnormality after a disaster. Changes in the living environment can increase the prevalence of mental health issues. The FHMS reported finding severe traumatic events among evacuees 22 . Mental disorders may increase the prevalence of hepatobiliary enzyme abnormality because obesity is associated with psychiatric disorders 23 . The incidence of psychological distress and post-traumatic stress decreased over time among residents in the evacuation zone after the Fukushima Daiichi Nuclear Power Plant accident 24 . Therefore, improved mental health after a disaster may promote a decrease in the incidence of hepatobiliary enzyme abnormality.
The present study did have several limitations. First, it was based only on an analysis of routine health checkups. Therefore, liver function such as Child-Pugh score or liver fibrosis scoring such as AST to platelet ratio index 25 and FIB-4 index 26 could not be evaluated. Other, we confirmed significant increase in the AST to ALT ratio, however, this ratio can be affected by age 27 . Moreover, no definitive diagnoses of hepatobiliary enzyme abnormality were established. In future studies, it will be critical to identify the precise cause of hepatobiliary enzyme abnormality, for example, viral hepatitis or fatty liver and medication. Second, the evaluation of lifestyle factors in the present study was based on responses to questionnaire surveys during health checkups, and therefore not very comprehensive. The associated dynamics in use of healthcare resources, such as antidiabetics, other related drugs and alcohol rehabilitation could not be evaluated in present study. Moreover, we could not evaluate food intake, mental status, or sleep duration and quality, all of which can affect obesity and hepatobiliary enzyme abnormality. It will be important in the future to elucidate which lifestyle factors are most affected by evacuation, as this could help in the development of novel strategies for protecting against the onset of hepatobiliary enzyme abnormality after a disaster. Third, there were no comparative cohorts in the present study. Future research is needed to compare the changes in hepatobiliary enzyme during the same observation period between residents living in the evacuation area and those living in an area far from the Fukushima Daiichi Nuclear Power plant.
In conclusion, in parallel with decreases in BMI, the incidence of hepatobiliary enzyme abnormality decreased 3-4 years after compared with immediately after the Great East Japan Earthquake. Improvements in lifestyle factors such as daily physical activity and breakfast consumption were associated with improved hepatobiliary enzyme abnormality. These findings are expected to be important for periodic health checkups and lifestyle recommendations among individuals at risk of developing hepatobiliary enzyme abnormality, regardless of evacuation status.

Study population.
For the purposes of this study, we analyzed a subset of participants from the FHMS. Informed consent was obtained from all subjects and community representatives to conduct an epidemiologic study based on the guidelines of the Council for International Organizations of Medical Science 28 . This study protocol was reviewed and approved by the Ethics Committee of Fukushima Medical University (#1916). The study was conducted in accordance with the approved guidelines.
Biochemical analysis. The following laboratory data were obtained from all participants: AST, U/L; ALT, U/L; γ-GTP, U/L; high-density lipoprotein cholesterol (HDL-C, mg/dL); low-density lipoprotein cholesterol (LDL-C, mg/dL); triglycerides (TG, mg/dL); fasting plasma glucose (FPG, mg/dL); and HbA1c (%). HbA1c values were estimated based on National Glycohemoglobin Standardization Program units, and the equivalent value was calculated using the following equation: HbA1c (%) = 1.019 × HbA1c (Japanese Diabetes Society) (%) +0.4% 29 . Hepatobiliary enzyme abnormality was defined as AST ≥ 31 U/L, ALT ≥ 31 U/L, or γ-GTP ≥ 51 U/L. In addition, moderate hepatobiliary enzyme abnormality was defined as AST ≥ 62 U/L, ALT ≥ 62 U/L, or γ-GTP ≥ 102 U/L. Liver fibrosis was evaluated using AST to ALT ratio, moreover, 0.87 was used as a cut off based on the Japanese criteria 30 . Diabetes was defined as fasting plasma glucose ≥ 126 mg/dL, HbA1c ≥ 6.5%, or the self-reported use of antihyperglycemic agents. Dyslipidemia was defined as fasting LDL-C ≥ 140 mg/dL, fasting TG ≥ 150 mg/dL, fasting HDL-C < 40 mg/dL, or the self-reported use of antidyslipidemic agents. Assessment of other variables. Height in stocking feet and weight in light clothing were measured, and body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared. Overweight was defined as a BMI ≥ 25 kg/m 2 . Systolic and diastolic blood pressure (BP) values were measured by trained technicians using a standard mercury sphygmomanometer on the right arm of seated participants. Hypertension was defined as systolic BP > 140 mmHg, diastolic BP > 90 mmHg, or the self-reported use of antihypertensive agents. Interviews were conducted to obtain the history of cigarette smoking, weekly alcohol intake, and medication history. Participants were classified into the following three groups according to alcohol consumption: (1) non-drinkers, defined as drinking no alcohol; (2) light drinkers, defined as drinking ≤22 g of alcohol per day; or (3) moderate/heavy drinkers, defined as drinking >22 g of alcohol per day.
Statistical analysis. Changes in data immediately and 3-4 years after the disaster were compared using the paired Student's t test and Wilcoxon signed-rank test. For participants with no liver dysfunction at baseline (immediately after the disaster), we tested associations between evacuation and other potential confounders with the incidence of liver dysfunction after the disaster using the Cox proportional hazards model. The following variables were considered potential confounding factors: age (continuous), sex, BMI (continuous), current smoking (yes or no), evacuation (yes or no), alcoholic intake (yes or no).
In addition, odds ratios and 95% confidence intervals were calculated to analyze the association between improvements in liver dysfunction and changes in lifestyle factors using a logistic regression model. Because alcohol consumption is one of the major causes of liver dysfunction, analysis was stratified according to drinking status (non-drinkers, light drinkers, or moderate/heavy drinkers). SAS version 9.3 (SAS Institute, Cary, NC, USA) was used for all statistical analyses. All probability values for statistical tests were two-tailed, and p values < 0.05 were considered statistically significant.