This study aims to determine the appropriate order of Legendre polynomial (LP) for a three-parity random regression (RR) model using the Japanese Holstein production data. The goodness of fit of the RR models with second-, third-, fourth-, and fifth-order LP for the fixed and random effects was evaluated based on five statistical criteria: the three residual variances of the first three parities, the log likelihood function (2logL), and Akaike's information criterion (AIC). A model that produces the smallest AIC indicates the best fit to the data. The percentage of reduction in residual variances with increasing order of LP was greater for the 2nd and 3rd parity than for the 1st parity, suggesting that the 1st lactation milk is more robust to the order of LP than the 2nd or the 3rd lactation milk because the first lactation curve is more persistent than the 2nd and 3rd lactation curves. All five statistical criteria decrease with increasing order of LP for the additive genetic and permanent environmental effects, suggesting that the higher the order of LP fitted for RR coefficients, the more accurate the RR model is. However, the rate of decrease for the five statistical criteria (as measured by the percentage of reduction) decreases as the order of LP goes beyond 4. When the order of LP for the fixed regression coefficients increases from 2 to 4, there are little changes in the five statistical criteria, indicating that the fitting of a RR model with a second-order LP for the fixed regression coefficients is justified. Based on the results of this study, it is advisable to fit a RR model with a fourth-order LP for the random effects and a second-order LP for the fixed effect in genetic evaluation of the milk EBV of the first three
1820 The presence of leptin resistance was previously reported in both obese adults and children. However, there have been few reports evaluating the influence of weight-reducing program combined with exercise on serum leptin levels. In the present study, we administered a weight-reducing program consisting of exercise and diet treatment to severely obese children and evaluated their body fat distributions, serum leptin levels and risk factors for chronic diseases before and after treatment. Subjects consisted of 7 pediatric inpatients with simple obesity (mean age 10.6 years) who had been hospitalized for a long time. As the weight-reducing program, an aerobic exercise using lightweight dumbbells and a stepping platform as well as diet therapy corresponding to the individual degree of obesity were performed for 11 weeks. Areas of subcutaneous and visceral fats were quantitatively determined using CT images. Blood samples were collected during early morning, and risk factors for chronic diseases, such as serum levels of leptin, TG, HDL-C, LDL-C and uric acid were determined. After completing the weight-reducing program, both areas of subcutaneous and visceral fats obtained from CT images were significantly decreased. Many items relating to chronic diseases, such as serum levels of lipids, hepatic enzymes and uric acid improved. After completing the weight-reducing program, serum leptin levels were significantly decreased (14.4 ng/ml -> 5.1 ng/ml, p <0.01). There was a strong positive correlation between the reduction rate of serum leptin levels and that of total fat areas obtained from abdominal CT images (r = 0.904, p <0.05). Moreover, serum leptin levels per unit of fat mass obtained after the weight-reducing program tended to be lower than those obtained before treatment. These results suggested that the combination of diet and exercise treatment for obese children might contribute to decreasing body fat mass and risk factors for chronic diseases as well as improving leptin resistance.
Nitric oxide (NO) produced in the airways can be either detrimental or protective to the host. To investigate the role of NO in the pathogenesis of exercise-induced bronchoconstriction (EIB), we measured exhaled NO (ENO) after exercise challenge in 39 asthmatic and six normal children. FEV1 and ENO were measured before and at 0, 5, 10, and 15 min after exercise performed on a treadmill for 6 min. EIB was defined as a decrease in FEV1 of more than 15% after the exercise. Normal children (control group) did not have EIB. Twenty-one patients with asthma had EIB (EIB group) whereas the remaining 18 patients did not (non-EIB group). The baseline ENO value was significantly higher in the asthmatic children than in the normal children, and there was a positive correlation between the maximal percent decrease in FEV1 and the baseline ENO value (r = 0.501, p = 0.012). At the end of the exercise, ENO had decreased in all the subjects. In the non-EIB and control groups, ENO rebounded to above the baseline at 5 min after the exercise and thereafter. In contrast, ENO remained at a decreased level in the EIB group. The change in ENO did not correlate with the change in minute ventilation, and β -agonist inhalation at the peak of EIB that accelerated the recovery of FEV1 did not affect the depressed level of ENO, demonstrating that the reduction of ENO is not a simple consequence of increased ventilation nor airway obstruction. Among the EIB group, steroid-treated patients showed sooner recovery in ENO after the exercise than steroid-naive patients. Our study suggests that NO production in response to exercise may be impaired in patients with EIB, and that ENO represents not only airway inflammation but also a protective function of NO in EIB.
