Bibian van der Voorn

82 CHAPTER 6 At ∼0700 (0600 to 0800), milk cortisol was 10.3 nmol/L (4.2 to 46.4 nmol/L), milk cortisone was 38.6 nmol/L (23.1 to 52.2 nmol/L), and the milk cortisone-to-(cortisol + cortisone) ratio was 0.75 (0.46 to 0.89). In saliva obtained at the same time, cortisol was 6.1 nmol/L (2.2 to 19.9 nmol/L), cortisone was 24.7 nmol/L (13.2 to 50.1 nmol/L), and the cortisone-to-(cortisol + cortisone) ratio was 0.80 (0.69 to 0.87). TABLE 3. Baseline characteristics of participants: Study 2 Mother’s age, yr 33 (28 to 37) Gestational age, wks 39 (37.1 to 41.3) Prepregnancy BMI, kg/m 2 22 (20 to 30) Pregnancy weight gain, kg 12 (6 to 16) EPDS score 4 (1 to 10) Values are median (ranges), n= 10. EPDS = Edinburgh Postnatal Depression Scale Table 4 presents the results of the GEEs. When analyzed linearly, time of collection was significantly associatedwithmilk cortisol and cortisone concentrations, as well as the milk cortisone-to-(cortisol + cortisone) ratio. When splitting the day into 2 parts, a significant increase in milk cortisol and cortisone concentrations was observed before the peak at ∼0700 and a significant decrease was observed after the peak. A reverse pattern was observed for the milk cortisone-to-(cortisol + cortisone) ratio ( Table 4, Figure 2 ) Cubic modeling provided the best fit (P < 0.01) to describe the diurnal rhythm of milk cortisol, milk cortisone, and the milk cortisone-to-(cortisol + cortisone) ratio ( Supplemental Table 1, Supplemental Figure 1 ). TABLE 4. GEE-analysis of the association between time of collection and cortisol and cortisone levels, and the cortisone/(cortisol + cortisone) ratio in breast milk: Study 2 Dependent variable Independent variable ß (95% CI) P -value Cortisol Time of collection 0.92 (0.89 ; 0.95) <0.01 0.00 - 7.00h 1.77 (1.55 ; 2.02) <0.01 7.00 - 23.59h 0.85 (0.83 ; 0.87) <0.01 Cortisone Time of collection 0.96 (0.93 ; 0.98) 0.01 0.00 - 7.00h 1.35 (1.27 ; 1.44) <0.01 7.00 - 23.59h 0.91 (0.90 ; 0.92) <0.01 Ratio Time of collection 1.01 (1.00 ; 1.01) <0.01 0.00 - 7.00h 0.97 (0.95 ; 0.98) <0.01 7.00 - 23.59h 1.01 (1.01 ; 1.01) <0.01 Values are exponentiated betas (95% CIs), n=10 mothers, who collected 92 milk samples in total. Cortisol and cortisone levels are measured in nmol/L Milk and salivary concentrations of cortisol (r = 0.92, SEE = 2.8 nmol/L; P < 0.01) and cortisone (r = 0.93, SEE = 4.4 nmol/L; P < 0.01) as well as the cortisone-to-(cortisol + cortisone) ratio (r = 0.64, SEE = 0.06; P < 0.01) were correlated with one another. For cortisol, the increase in breast-milk concentration seemed to be disproportionate to the higher range of salivary concentrations ( Figure 3 ).