Bibian van der Voorn

98 CHAPTER 7 For cortisone the equation was cortisone post-pasteurization = 0.14 + 1.01 * cortisone pre-pasteurization, with the intercept ranging from -0.66 to 0.99 and the slope ranging from 0.97 to 1.05. DISCUSSION Our study demonstrated that Holder pasteurization of human donor milk does not lead to a reduction inmilk cortisol and cortisone levels. The levels found in the donated samples were within the range found in other studies 6,11 . Combined with results from our previous study in which we showed that human milk cortisol and cortisone levels are stable during 36h at room temperature and during multiple freeze-thaw cycles 11 , we can now state that processing of human milk by donor milk banks does not affect milk cortisol and cortisone levels. Steroid hormones are not thermolabile; cholesterol is the substrate for steroid biosynthesis and consists of carbon-based ring structures, making the molecule rather stabile below the melting point (148.5 ⁰C) 12 . Therefore our results which showed that heating at 62.5°C for 30 minutes did not affect breast-milk cortisol and cortisone levels, were in line with our expectations. In contrast, protein hormones, such as insulin-like-growth-factors, growth hormone, leptin, adiponectin and insulin can denature upon heating and this explains decreased concentrations upon pasteurization 1 . Although, one study previously reported a possible heat-induced transformation of steroidal anti-inflammatory drugs in bovine milk 13 , due to the thermostabile characteristics of steroid hormones, it is reasonable to assume that the unchanged milk glucocorticoidal levels retained their functionality. The presence of milk glucocorticoids in donor human milk might exert direct and indirect beneficial effects on the gastro-intestinal tract of the vulnerable preterm infant. Evidence for direct effectswere found in animal studies inwhich administration of glucocorticoids was found to result in maturation of neonatal intestinal enzymatic activity 14 . In addition in human studies on preterm infants, administration of antenatal glucocorticoids was found to result in a reduced incidence of necrotizing enterocolitis and a decrease in intestinal permeability 15,16 . Simultaneously, milk glucocorticoids might exert an indirect beneficial effect via the microbiome, and via human milk oligosaccharides (HMOs). Although Holder pasteurization is known to inactivate the bacteria of human milk 17 , HMOs remain unaltered in donor milk 2 . These HMOs have the potential to support growth of Bifidobacterium in the gastrointestinal tract, which