We read with interest the article by Meek et al. (1). The authors have measured maternal serum C-peptide concentrations at 12, 24, and 34 weeks of gestation in 127 pregnant women with type 1 diabetes and cord blood C-peptide concentrations. In 74 (58%) pregnant women, C-peptide was not detected; in 22 (17%), it was confirmed at the beginning of pregnancy; and in 31 (24%), it was detected in the 34th week of pregnancy (1). Neonates born to mothers in whom C-peptide was detected at 34 weeks of gestation had elevated cord blood C-peptide and more frequent neonatal complications than others (1). Based on the results, the authors suggest a transfer of C-peptide from fetal to maternal serum without the regeneration of pregnancy-related β-cells.

In the prospective PRE-HYPO (prevention of hypoglycemia in pregnant women with diabetes type 1) study IP-2018–01–1284 HRZZ (https://mef.unizg. hr/znanost/istrazivanje/web-stranice- projekata/projekt-hrzz-pre-hypo/), we investigated the effect of endogenous insulin in pregnant women with type 1 diabetes on the incidence of hypoglycemia and the duration of glucose at <3.9 mmol/L in continuous glucose monitoring. Pregnant women with higher levels of endogenous insulin have better-regulated glycemia, fewer hypoglycemia events, reduced total insulin dose, lower incidence of macrosomia, and lower levels of umbilical vein C-peptide, consistent with findings from Meek et al. (1)

The concentration of C-peptide was determined by the highly sensitive electrochemiluminescence immunoassay (ECLIA) from Roche: Elecsys (Roche Diagnostics, Switzerland). We found a significant increase from the first to the third trimester for fasting C-peptide concentrations across all three trimesters (2,3). This finding is consistent with the increase in C-peptide throughout pregnancy reported by Nielsen et al. (4). Maternal serum C-peptide concentrations were negatively correlated with the type 1 diabetes duration (rs = −0.586; P < 0.001). A highly significant correlation coefficient was found between umbilical vein glucose and C-peptide concentrations (rs = 0.468; P < 0.001). Eutrophic infants of mothers with type 1 diabetes have a higher concentration of C-peptide in the umbilical vein serum than eutrophic infants of healthy pregnant women (2).

During cesarean delivery, we took maternal blood samples and umbilical vein samples simultaneously. No significant correlation was obtained when the maternal serum C-peptide concentration was compared with umbilical vein serum C-peptide from their newborns at delivery (rs = −0.064).

It is worth noting that, in the mothers who had higher serum C-peptide values, lower concentrations of C-peptide and glucose and lower HOMA (insulin resistance) values were found in the umbilical vein than in mothers with lower C-peptide values (3).

Therefore, the higher concentration of C-peptide in the cord blood in group 3 found by Meek et al. (1) is evidence of poorly regulated maternal glycemia, resulting in delivery of disproportionately large-for-gestational-age newborns, and does not represent evidence of C-peptide transport from fetus to mother and cause-effect of fetal macrosomia. Our study did not find an association between maternal venous serum C-peptide concentration and umbilical vein, which does not mean that there is no possibility of passing a minuscule amount of C-peptide from fetus to mother.

According to our and other authors' research, it is plausible that pregnancy yields immunological tolerance and stimulates endogenous insulin production in women with type 1 diabetes.