Predicting dry weight change in Hemodialysis patients using machine learning.

BACKGROUND: Machine Learning has been increasingly used in the medical field, including managing patients undergoing hemodialysis. The random forest classifier is a Machine Learning method that can generate high accuracy and interpretability in the data analysis of various diseases. We attempted to apply Machine Learning to adjust dry weight, the appropriate volume status of patients undergoing hemodialysis, which requires a complex decision-making process considering multiple indicators and the patient's physical conditions. METHODS: All medical data and 69,375 dialysis records of 314 Asian patients undergoing hemodialysis at a single dialysis center in Japan between July 2018 and April 2020 were collected from the electronic medical record system. Using the random forest classifier, we developed models to predict the probabilities of adjusting the dry weight at each dialysis session. RESULTS: The areas under the receiver-operating-characteristic curves of the models for adjusting the dry weight upward and downward were 0.70 and 0.74, respectively. The average probability of upward adjustment of the dry weight had sharp a peak around the actual change over time, while the average probability of downward adjustment of the dry weight formed a gradual peak. Feature importance analysis revealed that median blood pressure decline was a strong predictor for adjusting the dry weight upward. In contrast, elevated serum levels of C-reactive protein and hypoalbuminemia were important indicators for adjusting the dry weight downward. CONCLUSIONS: The random forest classifier should provide a helpful guide to predict the optimal changes to the dry weight with relative accuracy and may be useful in clinical practice.

Weight status change from birth to childhood and the odds of high blood pressure among Chinese children.

Background: It is well documented that birth weight and childhood weight are associated with the blood pressure (BP) levels in childhood. However, the impact of weight status change from birth to childhood on BP among children is less well described. We aimed to assess the association between changes in weight status from birth to childhood and high BP in childhood. Methods and results: Data were obtained from a cross-sectional survey conducted in Jinan, China, and a total of 5,546 children aged 6-17 years were included in this study. Based on the birth weight status [high weight (> 4,000 g) vs. normal weight (2,500-4,000 g)] and childhood weight status during the survey period [high weight (overweight and obesity) vs. normal weight], children were assigned into four groups: persistently normal weight (normal birth weight and normal childhood weight), resolved high weight (high birth weight but normal childhood weight), incident high weight (normal birth weight but high childhood weight), and persistently high weight (high birth weight and high childhood weight). After adjustment for sex and age, BP in childhood was more responsive to current body mass index (BMI) than birth weight. After adjustment for the potential covariates, compared with children who had persistently normal weight from birth to childhood, those with incident high weight (odds ratio [OR] = 3.88, 95% confidence interval [CI] = 3.29-4.57) and persistently high weight (OR = 3.52, 95% CI = 2.71-4.57) were associated with the increased odds of childhood high BP. However, children who had resolved high weight did not have significantly increased odds of high BP in childhood (OR = 0.86, 95% CI = 0.59-1.25). Conclusion: The association of BP with recent BMI was stronger than with birth weight. Children who had incident or persistently high weight from birth to childhood had increased odds of high BP in childhood, whereas the odds was not significantly increased among those with high birth weight but changed to normal weight in childhood. Our findings highlight the importance of maintaining an appropriate weight in the early lifetime for the prevention of high BP and other related diseases, especially for those with high birth weight.

Association of body mass index and weight change with pneumonia mortality in a Japanese population: Japan Public Health Center-based Prospective Study.

BACKGROUND: Accumulating evidence suggests that pneumonia mortality is lower for individuals with high body mass index (BMI) compared to normal BMI, but it remains unclear whether weight change during adulthood influences subsequent mortality due to pneumonia in Asian populations, who have a relatively lean body mass. This study aimed to examine the association of BMI and weight change over 5 years with the subsequent risk of pneumonia mortality in a Japanese population. METHODS: The present analysis included 79,564 Japan Public Health Center (JPHC)-based Prospective Study participants who completed a questionnaire between 1995 and 1998 were followed for death through 2016. BMI was categorized into four groups: underweight (<18.5 kg/m2), normal weight (BMI: 18.5-24.9 kg/m2), overweight (25.0-29.9 kg/m2), and obese (BMI: ≥30.0 kg/m2). Weight change was defined as the difference of body weight between questionnaire surveys with a 5-year interval. Cox proportional hazards regression was used to estimate hazard ratios of baseline BMI and weight change for pneumonia mortality. RESULTS: During a median follow-up of 18.9 y, we identified 994 deaths from pneumonia. Compared with participants with normal weight, an elevated risk was observed among those who were underweight (hazard ratio = 2.29, 95% confidence interval [CI]: 1.83-2.87), whereas a decreased risk was found among those who were overweight (hazard ratio = 0.63, 95% CI: 0.53-0.75). Regarding weight change, the multivariable-adjusted hazard ratio (95% CI) of pneumonia mortality for a weight loss of 5 kg or more versus a weight change of less than 2.5 kg was 1.75 (1.46-2.10), whereas that for a weight gain of 5 kg or more was 1.59 (1.27-2.00). CONCLUSION: Underweight and greater weight change was associated with an increase in the risk of pneumonia mortality in Japanese adults.

Melanocortin Receptors: Emerging Targets for the Treatment of Pigmentation, Inflammation, Stress, Weight Disorders and Sexual Dysfunction.

Melanocortins are tiny protein molecules formed by the post-translational cleavage of proopiomelanocortin. These are bioactive peptides that are responsible for human and lower animal pigmentation patterns, energy homeostasis, and sexual function modulation. These peptides regulate numerous physiological functions by being generated in the central nervous system and peripheral tissues. Melanocortins elicit their varied biological effects by binding to a separate family of G protein, two primary proteolytic enzymes, proconvertases 1 and 2, according to recent research. These breakthroughs have opened up new avenues for research into the role of melanocortins, antagonists, and receptors in a number of physiological activities.

The intestinal fatty acid binding protein is not essential for dietary fat absorption in mice.

The intestinal fatty acid binding protein (I-FABP) belongs to a family of 15 kDa clamshell-like proteins that are found in many different tissues. So far, nine types have been identified. Their primary structures are highly conserved between species but somewhat less so among the different types. The function of these proteins, many of which are highly expressed, is not well understood. Their ability to bind lipid ligands suggests a role in lipid metabolism, but direct evidence for this idea is still lacking. We tested the hypothesis that I-FABP serves an essential role in the assimilation of dietary fatty acids by disrupting its gene (Fabpi) in the mouse. We discovered that Fabpi-/- mice are viable, but they display alterations in body weight and are hyperinsulinemic. Male Fabpi-/- mice had elevated plasma triacylglycerols and weighed more regardless of the dietary fat content. In contrast, female Fabpi-/- mice gained less weight in response to a high-fat diet. The results clearly demonstrate that I-FABP is not essential for dietary fat absorption. We propose that I-FABP functions as a lipid-sensing component of energy homeostasis that alters body weight gain in a gender-specific fashion.

Leptin reduces body weight gain in neonatal rats.

Leptin (OB protein) elicits a neuroendocrine response to starvation and states of nutritional abundance to stabilize the proportion of body fat. Leptin has dramatic effects on food intake and energy expenditure in adult and juvenile rodents. However, whether the neonatal period is associated with the development of an effective leptin feedback system is still not known. In this study, we evaluated the effects of peripherally administered leptin on body weight changes in neonatal rats during the early suckling period (from birth to 10 d). Our results show that daily i.p. injections of leptin (0.3 microg/g and 1.0 microg/g) to neonatal rats led to a significant reduction in weight gain over 10 d compared with the control group (p < 0.01 and p < 0.01, respectively). Concomitant with a reduction in weight gain, retroperitoneal fat pad weight also significantly decreased in the leptin-treated group. Our data indicate that the potential for energy balance regulation by leptin occurs in the first day after birth. In addition, we also observed that 3 d after discontinuing leptin treatment, the body weight as well as the fat pad weight of leptin-treated pups returned to the control level. Our results demonstrate that leptin reduces body weight gain in neonatal rats.

A prospective study of weight change and health-related quality of life in women.

CONTEXT: The mean body weight of US adults increased by 3.6 kg (7.6 lb) during the past 15 years, but few studies exist that examine the impact of such weight change on functional health status. OBJECTIVE: To investigate, prospectively, the association between weight change and health-related quality of life in women. DESIGN AND SETTING: Nurses' Health Study, a 4-year prospective observational study from 1992 to 1996, using the Medical Outcomes Study Short-Form 36 Health Status Survey (a self-administered 36-item questionnaire) to measure quality of life. PARTICIPANTS: A cohort of 40098 women (from 46-71 years old in 1992) grouped according to 3 patterns of weight change over the 4-year period: women whose weight remained within 2.25 kg (5 lb) of their baseline weight, women who lost 2.25 kg (5 lb) or more, and women who gained 2.25 kg (5 lb) or more. MAIN OUTCOME MEASURES: Change in scores on 7 health-related quality-of-life dimensions: physical functioning, vitality, bodily pain, limitations in role functioning due to emotional or physical problems, social functioning, and mental health, measured by the Short-Form 36 Health Status Survey. RESULTS: A total of 15602 women (39%) maintained their weight, 15160 (38%) gained between 2.25 and 9.0 kg (5-20 lb), and 6667 (17%) lost between 2.25 and 9.0 kg (5-20 lb). Weight gain was associated with decreased physical function and vitality, and increased bodily pain regardless of baseline weight. For example, the odds ratio for developing role limitations due to physical problems was 2.05 (95% confidence interval, 1.69-2.49) for the leanest women who gained 9.0 kg (20 lb) or more. Weight loss in overweight women was associated with improved physical function and vitality as well as decreased bodily pain. Weight change was more strongly associated with physical rather than mental health. The impact of weight change, especially weight gain, was just as strong in women 65 years and older as in women younger than 65 years. CONCLUSIONS: These longitudinal data support current US guidelines for women of all body mass index levels to avoid weight gain. Weight maintenance and, in cases of overweight, weight loss are desirable and likely to be beneficial for physical function, vitality, and bodily pain.

Obesity, weight change, fasting insulin, proinsulin, C-peptide, and insulin-like growth factor-1 levels in women with and without breast cancer: the Rancho Bernardo Study.

Postmenopausal overweight women have an increased risk of breast cancer. The link between obesity and breast cancer could be mediated through hyperinsulinemia. Insulin and insulin-like growth factor-1 (IGF-1) stimulate mammary cell proliferation in vitro, and cell proliferation is directly linked to the risk of breast cancer. Our objective was to investigate the relationship between breast cancer and body composition, IGF-1, proinsulin, C-peptide, and fasting insulin. A case-control study was conducted of 438 community-dwelling women aged 53-90 years in 1992-1994 who had no history of cancer at the baseline visit in 1972-1974. Women were excluded who were using estrogen replacement therapy (ERT) or tamoxifen at the 1992-1994 visit, when IGF-1, proinsulin, fasting insulin, and C-peptide levels were measured. Prior ERT, alcohol and tobacco use, exercise, and reproductive history were recorded. Weight, height, and waist/hip ratio were measured. The 45 women with breast cancer had similar baseline body mass indices to the 393 women without breast cancer but had gained significantly more weight between the baseline visit in 1972-1974 and 1992-1994, (age-adjusted relative risk [RR] 1.05/kg, 95% confidence interval [CI] 1.01-1.09, p = 0.016). Proinsulin, fasting insulin, and C-peptide were each significantly positively correlated with both current weight and weight gain. However, levels of these hormones and IGF-1 did not differ significantly between women with and without breast cancer (all 95% CI within 0.996-1.004). Past ERT was significantly more common among women with breast cancer (p = 0.015), and duration of use was significantly longer (age-adjusted RR 1.13 per year of use, 95% CI 1.08-1.18, p = 0.000). The risk of breast cancer was significantly increased in women who had gained weight or used ERT. This increased risk was not associated with circulating levels of IGF-1, fasting insulin, proinsulin, or C-peptide.