Analysis of the In uence of Lifestyle on Liver Dysfunction Based on NAHANES Database


 ObjectiveTo find out the influence of various factors in life style on liver dysfunction, put forward reasonable suggestions on prevention of liver dysfunction.MethodsDatasets from 2017 to March 2020 NHANES (National Health and Nutrition Examination Surveys) required for the analysis were downloaded from the NHANES web site and R 4.1.1. Software was used for data analysis. Survey Design Logisitic regression was used to analyze the influence of various factors on liver dysfunction and screen the risk factors.Resultshypertension, depression, and sedentary activity are risk factors for liver dysfunction, while reducing salt in diet and vigorous recreational activities were protective factors for liver dysfunction. The inflection points of blood pressure, BMI and sedentary activity were 98.33mmHg, 30.6kg/m2, 420min, respectively.ConclusionsBlood pressure, BMI, mood and sedentary behavior are risk factors for liver dysfunction. We suggest that keeping MAP level at 70-98.33mmHg, controlling BMI < 30kg/m2, maintaining a positive attitude, and sedentary time less than 420 min per day are more conducive to reducing the risk of liver dysfunction.


Introduction
Liver is the largest substantial and digestive organ of the human body, with a variety of functions such as secretion, synthesis, metabolism and excretion. Various factors such as virus, pathology and chemical substances can affect the normal structure of liver parenchyma cells and tissue, causing liver dysfunction (1). With the growth of age, the structure and function of liver will change, including Alanine aminotransferase (ALT)(2), Volume(3), etc.
Different lifestyles affect liver function through a variety of factors, such as alcohol, drugs, and gut ora. Changes in gut microbiota can have a big impact on the liver, as gut bacteria and their by-products can enter the liver through the portal vein (4). Alcohol is the most common cause of liver damage. Nearly all people who consume alcohol for a long time develop fatty liver, but only a small proportion develop cirrhosis (5,6), which leads to liver dysfunction. But there are also some people in the occurrence of liver dysfunction before the discovery of liver diseases, many factors in life may be the cause of disease, but lack of relevant research. Intestinal microbiota changes occur in many liver diseases, such as nonalcoholic fatty liver disease, cirrhosis, alcoholic cirrhosis, and cirrhosis with encephalopathy (7). In addition, different lifestyles can also affect liver function to varying degrees, but there is still a lack of research on this aspect. Therefore, this study was conducted to study working time, activity, mood, blood pressure, sleep time, smoking and other lifestyle factors, in order to nd the in uence of different life factors on liver dysfunction and provide reference for the prevention of liver dysfunction.

Study Sample
Datasets from 2017 to March 2020 NHANES (National Health and Nutrition Examination Surveys) required for the analysis were downloaded from the NHANES web site and R 4.1.1. Software was used for data combined. The respondents included in this study were all over 20 years old, all accepted complete NHANES surveys, including Blood Pressure, Diabetes, Medical Conditions, Mental Health, Physical Activity, Use, Prescription, Medications, Sleep Disorders, Smoking, etc. After excluding respondents with incomplete survey data, a total of 11388 respondents were included in this study, including 5034 males and 6354 females. Characteristics of respondents are described in Table 1.

Classi cation of indicators
All indexes included in the study were converted into dichotomous variables. If the depression index occurred More than "More than half the days", the event was considered to have occurred. The time of sedentary activity is bounded by 8 hours. If the time exceeds 8 hours, the event is considered to occur. Mean arterial pressure (MAP) was used to re ect blood pressure to study the relationship between blood pressure and liver dysfunction. MAP=Diastolic blood pressure+ 1/3*Systolic blood pressure. to consider the respondents to have liver dysfunction in this study (8,9).

Statistical analysis
Analyses were conducted using R 4.1.1. Survey Design Logisitic regression was used to analyze the in uence of various factors on liver dysfunction and screen the risk factors. The risk factors of liver dysfunction were strati ed by sex, age, race and education level. Then the trend P test was carried out for several important factors. Smooth curve tting was made for mean arterial pressure, body mass index (BMI) and sedentary activity time to nd the in ection point of the curve.

Analysis of the in uence of various factors on liver dysfunction
After analyzing lifestyle-related factors in the NHANES database, we found that we found that hypertension (OR=1.

The relationship between blood pressure and liver dysfunction
We divided MAP into four groups and conducted trend P test. The results are shown in Table 3. The Model 1 has been non-adjusted. The Model 2 adjusted gender, race, education level, age, Marital status and Ratio of family income to poverty. The Model 3 adjusted for factors other than weight and liver dysfunction. Then smooth curve tting was performed for MAP and liver dysfunction, strati ed by gender, age, race and education level respectively, and the results were shown in Figure 2. We found that the risk of liver dysfunction in men increased rst and then decreased with MAP, with an in ection point of 134.33mmHg, while that in females showed an overall upward trend after a uctuation of 98.33mmHg. The total in ection point was 98.33mmHg.
In age group, we found that the curve of 20-40 years old group did not change much, and began to rise after 111mmHg, and the risk of liver dysfunction increased with it. In the 40-60 years old group, the curve reached the rst in ection point at 95mmHg. In the 60-80 years old group, the risk of liver insu ciency was higher when blood pressure was greater than 92mmHg. In the group older than 80 years, the overall trend was downward, and the risk changed little after 100.33mmHg. The total in ection point was 98.33mmHg.
In racial subgroups, we nd that the curve of non-Hispanic White is stable and begins to rise after 142.67mmHg, and the curve of Mexican American and Other Race remains at a high level around 98mmHg.Mexican American declined after 140mm. Non-Hispanic Black uctuated and began to decline after 154.67 mmHg. The total in ection point was 98.33mmHg.
In the education level group, we nd that the curve of less high school graduate group remains at the high risk level after 108.33mmHg, and the curve of high school graduate/GED or Equivalent group begins to rise after 118mmHg. While College graduate or above group increased at 92.67 mmHg. The total in ection point was 109.67 mmHg.

The relationship between BMI and liver dysfunction
We divided BMI into four groups and conducted trend P test. The results are shown in Table 4. The Model 1 has been non-adjusted. The Model 2 adjusted gender, race, education level, age, Marital status and Ratio of family income to poverty. The Model 3 adjusted for factors other than weight and liver dysfunction. Then smooth curve tting was performed for BMI and liver dysfunction, strati ed by gender, age, race and education level respectively, and the results were shown in Figure 3.
In the gender group, with the change of BMI, the risk of liver dysfunction did not change signi cantly in women, but signi cantly in men. The risk decreased rst and then increased signi cantly after 34.5kg/m 2 . The total in ection point was 30.6 kg/m 2 .
In age groups, 20-40 years old formed a steady upward trend, and the in ection points of 40-60 years old and 60-80 years old were 30.7 kg/m 2 and 25.8 kg/m 2 , respectively. However, at the same BMI, the risk was lower than that of 20-40 years old, and the risk of > 80 years old decreased signi cantly after 22.9 kg/m 2 . The total in ection point was 30.6 kg/m 2 .
In race groups, the curve of non-Hispanic White years old group and over other Race group increased steadily, but the risk of liver dysfunction in non-Hispanic White years old group was much lower than that in over other Race group. The risk of liver dysfunction in Non-Hispanic Black group began to increase after 22.4 kg/m 2 . The risk of liver dysfunction in Mexican American group increased signi cantly and reached the in ection point at 33.8 kg/m 2 . Its risk of liver dysfunction was much higher than the other three groups. The total in ection point was 30.6 kg/m 2 .
In the education level group, the less high school graduate group shows a downward trend as a whole, and the high school graduate/GED or Equivalent group drops after reaching the in ection point of 39.8 kg/m 2 . College graduate or above group starts to rise at 29.9 kg/m 2 . The total in ection point was 30.6 kg/m 2 .

The relationship between sedentary activity time and liver dysfunction
We divided sedentary activity time into four groups and conducted trend P test. The results are shown in Table 5. The Model 1 has been non-adjusted. The Model 2 adjusted gender, race, education level, age, Marital status and Ratio of family income to poverty. The Model 3 adjusted for factors other than weight and liver dysfunction. Then smooth curve tting was performed for sedentary activity time and liver dysfunction, strati ed by gender, age, race and education level respectively, and the results were shown in Figure 4.
In the gender group, the curve of male uctuated greatly and reached the lowest point at 600min, while the curve of female rose at 75min and was generally stable. The risk of liver dysfunction in female was lower than that in male. The total in ection point was 451min.
In age groups, the changes of 20-40 years old group and 40-60 years old group were not obvious and reached the in ection point at 660min.The curve of the group over 80 years old gradually decreased and leveled off after 150min.
The risk of liver dysfunction in the 60-80 years old group was signi cantly higher than that in the other three groups, and reached the in ection point at 240min. The total in ection point was 420min.
In the race groups, the curves of non-Hispanic White group changed signi cantly and increased after 150min, non-Hispanic Black group increased steadily from 90min, and Mexican American group decreased after reaching the in ection point at 300min.The Other Race group gradually rises. The total in ection point was 420min.
In the educational level group, high school graduate/GED or equivalent group and college graduate or above group gradually increase, the less High school graduate group began to decline after reaching the in ection point in 420min.
The total in ection point was 420min.

Discussion
Heart diseases such as chronic heart failure can cause abnormal liver function(10-12), which is closely related to changes in hemodynamics(13-15). Changes in hemodynamics will also be re ected in changes in blood pressure. In our study, the total in ection point of gender, age, and race was 98.33mmHg, and the total in ection point of education level was 109.67mmHg. Under the condition of ensuring the normal perfusion of body organs, blood pressure less than the in ection point may have a smaller occurrence The risk of liver dysfunction. Therefore, we consider that maintaining a MAP of 70-98.33mmHg in normal people is more conducive to reducing the risk of liver dysfunction.
Obesity is de ned as BMI> 30 kg/m 2 , while morbid obesity is de ned as BMI> 40 kg/m 2 (16). BMI is an important indicator for evaluating human health, and it is also related to most diseases, including cardiovascular disease, reduced life expectancy, increased mortality, psychosocial problems, blood pressure, etc(17, 18). We found that BMI is related to the occurrence of liver dysfunction. Regardless of gender, age, race, and education level group, the total in ection point of BMI is 30.6 kg/m 2 . This is consistent with the diagnostic criteria for obesity with BMI<30 kg/m 2 . Therefore, a lower BMI can reduce the risk of liver dysfunction.
Sedentary behavior is independently associated with adverse health outcomes(19-21), including obesity, metabolic syndrome (METS), non-alcoholic fatty liver disease (NAFLD), and type 2 diabetes(22-26). KELLY et al. also found that sitting was associated with fatty liver disease(27). However, these studies did not give the relationship between sedentary time and liver dysfunction, nor did they give a speci c time. In the sedentary activity study, I found that the total in ection point in the gender group was 451 minutes, and the total in ection point in age, race, and education level group was 420 minutes. Therefore, we believe that a sedentary time of less than 420 minutes a day can reduce the risk of liver dysfunction (not including sleep time).
In our study, several evaluation indicators of depression are risk factors for liver insu ciency, which is consistent with the results of several previous studies. Studies have shown that about 20-30% of patients with liver disease suffer from depression(28-30). The speci c cause is not clear, but this may be related to alcoholism and drug use(31, 32). Our research also found that seeing a psychiatrist in the past month is a risk factor for liver insu ciency, but taking prescription drugs is a protective factor. The prescription drugs here are not only antidepressants, probably because these drugs improve the patient's body condition.
In conclusion, blood pressure, BMI, mood and sedentary behavior are risk factors for liver dysfunction. We suggest that keeping MAP level at 70-98.33mmHg, controlling BMI < 30kg/m 2 , maintaining a positive attitude, and sedentary time less than 420 min per day are more conducive to reducing the risk of liver dysfunction.

Declarations Ethics declarations
Ethics approval and consent to participate   The analysis of the in uence of various factors on liver dysfunction.   Smoothing curve of education level; 1: Less High school graduate; 2: High school graduate/GED or equivalent; 3: College graduate or above.