It remains unclear whether severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) can be shed into breastmilk and transmitted to a child through breastfeeding. Recent investigations have found no evidence of SARS-CoV-2 in human breastmilk, but sample sizes were small.
We examined milk from two nursing mothers infected with SARS-CoV-2. Both mothers were informed about the study and gave informed consent. Ethical approval for this case study was waived by the Ethics Committee of Ulm University and all samples were anonymised.
Clinical data and the timecourse of infection in the two mothers is shown in figure 1. After feeding and nipple disinfection, milk was collected with pumps and stored in sterile containers at 4°C or −20°C until further analysis. We determined viral loads using RT-qPCR for SARS-CoV-2 N and ORF1b-nsp14 genes in both whole and skimmed milk (obtained after removal of the lipid fraction). Further details of sample storage and processing are provided in the appendix. Following admission and delivery (day 0), four samples from Mother 1 tested negative (figure 2). By contrast, SARS-CoV-2 RNA was detected in milk from Mother 2 at days 10 (left and right breast), 12, and 13. Samples taken subsequently were negative (figure 2). Ct values for SARS-CoV-2 N peaked at 29·8 and 30·4 in whole milk and skimmed milk, respectively, corresponding to 1·32 × 105 copies per mL and 9·48 × 104 copies per mL (mean of both isolations). Since milk components might affect RNA isolation and quantification, viral RNA recovery rates in milk spiked with serial dilutions of a SARS-CoV-2 stock were determined. We observed up to 89·2% reduced recovery rate in whole milk and 51·5% in skimmed milk (appendix), suggesting that the actual viral loads in whole milk of Mother 2 could be even higher than detected.