Quality indicators of raw human milk under prolonged household freezing: titratable acidity, energy content, and presumptive coliform screening
Description
This dataset comes from a prospective longitudinal study evaluating the quality of raw human milk (RHM) under household freezing for up to 150 days, conducted in July 2025 at the Nutrition and Dietetics Laboratory of the Faculdade Israelita de Ciências da Saúde Albert Einstein (FICSAE), São Paulo, Brazil. It was motivated by the discrepancy between Brazilian regulations, which recommend a maximum of 15 days for expressed human milk, and international guidelines allowing three to twelve months for healthy full-term infants. Milk was expressed by 28 lactating individuals under domestic hygiene conditions, yielding 50 samples divided into aliquots and stored in the freezer compartment of a domestic two-door refrigerator at -18 °C. A total of 261 aliquots were analysed at seven storage time points (1, 15, 30, 60, 90, 120 and 150 days), producing 153 donor-level observations. Two routine indicators were measured at each point: titratable acidity (°D) by Dornic titration, and energy content (kcal/L) estimated by crematocrit. Total coliform occurrence was determined by the presumptive phase of the Brilliant Green Bile Lactose (BGBL) broth method and is reported as an ancillary indicator of hygienic condition; without confirmatory testing, it supports no conclusion regarding microbiological safety. Raw measurements are distributed across five sheets: Database_sample_wide, Database_aliquot_long, Database_participant_wide, Database_participant_long and Dictionary_RHM. Jamovi output files are also provided to support reproducibility. Acidity varied longitudinally during storage while energy content did not; coliform occurrence was infrequent and recorded as binary (Absent/Present). Missing observations (21.9% of planned donor–time-point combinations) reflect variability in expressed volume and are documented individually. To the authors' knowledge, this is the first Brazilian longitudinal dataset evaluating RHM quality at seven storage time points extending to 150 days under household freezing. Version 2 corrects a spreadsheet-formula artefact in Database_sample_wide, where 16 cells evaluated to zero rather than remaining empty, and a header label altered from n=20 to n=21. No other value was changed. Database_participant_long, used for all reported analyses, never contained the affected values; the number of analysed aliquots (261) and donor-level observations (153) and every published result are unaffected. The dataset and its documentation have also been fully translated into English in this version. Version 1 was deposited under the earlier title "Quality assessment of raw human milk under prolonged household freezing: titratable acidity, energy content, and microbiological safety". The title was revised because the microbiological component is restricted to the presumptive phase of a qualitative screening method and supports no conclusion regarding microbiological safety.
Files
Steps to reproduce
A prospective longitudinal storage study was conducted at the Nutrition and Dietetics Laboratory of the Faculdade Israelita de Ciências da Saúde Albert Einstein (FICSAE), São Paulo, Brazil, in July 2025. Lactating adults were recruited through social networks and community dissemination channels. Eligible participants were aged 18 years or older, had been lactating for at least 15 days, and possessed a personal breast pump. Written informed consent was obtained from all participants. An a priori calculation in G*Power 3.1 indicated a minimum of 18 participants. Before milk expression, breast pump components were cleaned with water and soap, sanitised by immersion in boiling water for 15 minutes, and air-dried. Milk was expressed using personal electric or manual breast pumps and collected directly into sanitised glass containers. Immediately after expression, samples were refrigerated at ≤4 °C. Within 24 hours, milk was aliquoted under laminar airflow into autoclaved 10 mL screw-cap tubes using automatic pipettes with sterile tips. Aliquots were assigned to seven storage time points: day 1, 15, 30, 60, 90, 120 and 150. Participants unable to provide the full 70 mL contributed aliquots in a pre-established priority order (T1, T150, T90, T30, T120, T60, T15) defined before any analysis. The day-1 aliquot remained refrigerated and was analysed within 24 hours; the rest were stored at −18 °C in the freezer compartment of a domestic two-door refrigerator until analysis. Before testing, frozen aliquots were thawed in a water bath not exceeding 37 °C, with the heat source switched off before immersion, and swirled gently throughout. Three analyses were performed at each time point from the same thawed aliquot: Titratable acidity, in triplicate by Dornic titration, 3 mL per replicate, expressed as degrees Dornic (°D); Energy content, in triplicate by crematocrit, 1 mL per replicate, with cream and total column readings recorded and cream content, fat content and energy content (kcal/L) calculated per the rBLH/FIOCRUZ protocol; Total coliform occurrence by the presumptive phase of the Brilliant Green Bile Lactose (BGBL) broth method, 4 mL inoculated into BGBL broth with Durham tubes, gas formation recorded as Absent or Present after incubation. Sociodemographic, lactation and milk-storage information were collected through a structured questionnaire at the collection session. Raw data, derived variables, participant-level summaries and a data dictionary were compiled into a spreadsheet database. Analyses were performed in jamovi 2.3 with the GAMLj module: continuous outcomes by generalised mixed models (Gamma distribution, identity link) and Spearman correlation; coliform associations by Fisher's exact test, Mantel-Haenszel linear trend test and Mann-Whitney U test. All statistical outputs are provided as supplementary PDF files.
Institutions
- Hospital Israelita Albert EinsteinSão Paulo, São Paulo