A nonlinear model for mammary gland growth and regression in lactating sows

The objective was to propose an empirical mathematical model to describe ma mmary gland growth and regression in lactating sows. A nonlinear dynamic mo del based on the logistic function was constructed, and data from 61 sows w ere used to illustrate the model. Sows were fed four diets with two levels of energy and of protein during lactation, and individuals were slaughtered over a 30-d period to produce a cross sectional data set on weight and com position variables from suckled mammary glands. Data (y(x)) were obtained f or each day of lactation (x) and fitted by nonlinear regression. The logist ic distribution function was modified for different durations of growth (f; days/gram of weight or composition) and regression (g; days/gram of weight or composition): Y-x=y(max)e(x/f+xmax/g)(f+g/fe(xmax/2f+xmax/2g)+ge(x/2f+x/2g))(2) where y(max) is maximum weight or composition and x(max) is day of lactatio n at maximum. Based on results for wet weight, for example, individually su ckled mammary glands grow until between Day 21 and 28 of lactation and reac h a maximum of about 500 to 600 g, depending on diet. Growth pattern of mam mary glands can be described well with an asymmetric nonlinear model, using different durations for growth and regression. From this model, it was pos sible to estimate directly biologically important parameters: maximum weigh t or composition, day of lactation at maximum weight or composition, and du rations of growth and regression. This model can be applied to describe mam mary gland growth patterns for other species and to describe similar growth or production patterns.