Bibian van der Voorn

97 STABILITY OF CORTISOL AND CORTISONE IN HUMAN BREAST MILK Passing and Bablok regression analysis. Assuming that pre- and post-pasteurization concentrations are equal, the intercept (95% CI) of the Passing and Bablok regression line should not differ from zero and the slope (95% CI) should not differ from one. RESULTS Thirty mothers provided milk samples at a median postpartum age of 18.1 weeks (range: 1.3 to 57.7). Their samples were collected at several moments during a time period ranging from 2 to 70 days (median 13.5). Their infants were born at a median gestational age of 40.2 weeks (range: 37.7 to 42.1), and 15 (50%) of them were male. Table 1 shows mean cortisol and cortisone levels pre- and post-pasteurization. Pre- and post-pasteurization levels correlated highly: r = 0.997, R 2 =0.99, P < 0.01 for cortisol, and r = 0.990, R 2 = 0.98, P < 0.01 for cortisone. TABLE 1. Mean breast-milk glucocorticoid levels pre- and post-pasteurization, n =30. Pre-pasteurization Post-pasteurization Mean ± SD (range) Mean ± SD (range) Milk cortisol (nmol/L) 7.93 ± 4.88 (0.8 to 19.7) 7.96 ± 5.05 (0.8 to 19.8) Milk cortisone (nmol/L) 20.98 ± 6.24 (3.7 to 29.6) 21.19 ± 6.36 (3.7 to 30.2) Figure 1 shows the Passing and Bablok regression analysis, displaying good agreement between pre- and post-pasteurization concentrations. For cortisol the equation was cortisol post-pasteurization = 1.00 * cortisol pre-pasteurization, with the intercept ranging from -0.22 to 0.07 and the slope ranging from 0.98 to 1.04. FIGURE 1. Passing and Bablok regression analysis of cortisol and cortisone in paired pre- and post-pasteurization milk samples. N = 30 paired samples.