Objective. Data are scarce regarding the effects of nutrient fortification and storage on the immunoprotective properties of human milk. These effects are important considerations when feeding premature infants. We hypothesized that total bacterial colony counts (TBCC) and immunoglobulin A (IgA) concentration were not affected by the addition of fortifier even when tested under extreme storage conditions and that osmolality of fortified human milk does not increase with storage.
Methods. Ten frozen and five fresh milk samples from mothers of premature infants were divided into fortified and unfortified milk, and stored for 72 hours at either refrigerator or room temperature. Aliquots were obtained at 0 to 72 hours for TBCC, osmolality, and total IgA, and analyzed by repeated measures analysis of variance (ANOVA).
Results. Log10 TBCC in milk stored at refrigerator temperature for 0, 24, 48, and 72 hours were significantly greater in fortified vs unfortified milk; both increased similarly with storage. Osmolality was greater in fortified than unfortified milk; both increased by approximately 4% with storage. IgA concentration was not affected by fortification or storage. To simulate the usual nursery use of fortified human milk, a separate evaluation was performed. Fortified milk was stored at refrigerator temperature for 20 hours, warmed in a 40°C laboratory incubator for 20 minutes, and placed in a 34°C infant incubator for 4 hours. Samples for TBCC were obtained at 0, 20, and 24 hours and analyzed by repeated measures ANOVA. Log10 TBCC in fortified, refrigerated milk did not change over the 20-hour storage but increased during the simulated 4-hour usage.
Conclusions. These findings may warrant consideration when using human milk in the neonatal nursery but support recommendations to use commercially fortified human milk within 24 hours.
Human milk is being used increasingly to feed the premature infant.1 The high bioavailability of its nutrients coupled with its immunoprotective factors make human milk unique.1,2 Nurseries have established protocols for the use of mother's own milk,3,4 because storage becomes necessary when feeding is delayed.5,6 To ensure the appropriate growth of the premature infant, fortification of human milk with protein and minerals is essential.1 However, data are scarce regarding the effects of fortification on the immunoprotective properties of human milk.
Measures to control the bacterial quality of unfortified human milk have been reported.7 Bacterial colony counts in fresh human milk progressively decreased throughout a 5-day refrigeration period.8 Refrigerator storage of human milk up to 8 days had a significant inhibitory effect on bacterial growth that was not observed after freezing.9,10 However, specific quality control measures have not been established for fortified human milk, despite the potential for alteration of protective mechanisms due to nutrient fortification.
The concentration of total immunoglobulin A (IgA), the major immunoglobulin in human milk, did not decrease over time in sterile milk, but decreased in potentially contaminated milk, whether frozen or refrigerated.9 The decrease in total IgA in contaminated milk may be a result of the presence of IgA proteases of bacterial origin. Total IgA concentration, therefore, may be used as a determinant of milk integrity. The effect of commercial fortification of human milk on total IgA concentration is less clear. One study reported that the total IgA concentration in human milk was not significantly lower after the addition of fortifier, but the effects of time and storage were not investigated.11
Bacterial colony counts and flora identification are specific indicators of milk contamination and spoilage. Osmolality also is used as an indicator of milk lipolysis and/or protein degradation, signs of potential rancidity.
This study was designed to determine the effects of commercial nutrient fortification on the intrinsic host defense properties of human milk. We hypothesized that when tested under a variety of extreme storage conditions, the addition of nutrient fortifier to human milk would not increase total bacterial colony counts (TBCC) or promote nutrient degradation, as measured by osmolality and total IgA concentration compared with unfortified milk.
MATERIALS AND METHODS
Single samples of milk from mothers of premature infants (n = 10) that had been collected using electric breast pumps and stored for 1 to 6 weeks at −20°C in the Texas Children's Hospital Milk Bank were studied. Five additional samples of milk from mothers of premature infants were obtained fresh using similar methods for collection. These samples were refrigerated at 4°C for less than 6 hours until studied. Frozen milk samples were thawed using a tepid water fast-thaw method. The milk samples were divided into fortified and unfortified milk, and further divided for storage at refrigerator temperature (4°C) or room temperature (26°C) for 0 to 72 hours.
In Part A of the study, TBCC, osmolality, and total IgA concentration were measured from aliquots obtained at 0, 24, 48, and 72 hours. In Part B of the study, frozen milk was thawed and fortified using the standard protocol and samples were refrigerated for 20 hours, warmed per usual nursery practice (using a 40°C dry heat laboratory incubator for 20 minutes), and placed in a 34°C infant incubator for 4 hours to simulate routine use of milk in the neonatal intensive care unit. Samples for TBCC were obtained at 0, 20, and 24 hours.
TBCCs were determined within 1 hour after sterile collection. Each milk sample was quantitatively cultured in trypticase soy agar using a minimum of five dilutions for each sample. After incubation for 48 hours at 35°C in an environment containing 10% carbon dioxide, standard plate counts were determined. Colonies were then inoculated into blood, mannitol, and MacConkey's agar and incubated at 35°C for 24 to 48 hours. Bacteria were identified with standard biochemical tests, colony morphology, and the microscopic characteristics found in Gram's-stained preparations. Osmolality was measured using a vapor pressure osmometer (Wescor Inc, Logan, UT). Total IgA concentration was determined by enzyme linked immunosorbent assay (Jackson Immunoresearch Laboratories, Inc, West Grove, PA) using double rabbit antihuman IgA directed toward the heavy chain of human IgA. Fragments are not detected by this method.
Mothers were asked to donate a sample of milk if they were more than 1 month postdelivery. For stored samples, the study milk was taken from the remaining milk thawed for daily use. Enfamil Human Milk Fortifier (Mead Johnson Nutritional Division, Evansville, IN) was used for fortification in accordance with the manufacturer's instructions, 4 packets of fortifier per 100 mL of human milk. The protocol was approved by the Institutional Review Board for Human Research.
Data were analyzed as fortified vs unfortified milk using repeated measures analysis of variance (ANOVA) to account for changes in storage time. Milk state, defined as either fresh or frozen milk, was used as a covariate in these analyses. Part A of the study was conducted both at refrigerator temperature and at room temperature. Part B of the study was conducted at refrigerator temperature.
TBCCs in refrigerated milk were significantly greater from 0 to 72 hours in fortified vs unfortified human milk (Fig 1). Differences in TBCC between fortified and unfortified milk were not affected by milk state (whether the study milk initially used was fresh or frozen). TBCC in milks stored at room temperature for 0, 24, 48, and 72 hours did not differ between fortified (n = 15, log10 mean TBCC were 4.9, 6.3, 6.8, and 7.8) vs unfortified milk (n = 15, log10 mean TBCC were 4.8. 5.6, 6.2, 7.5), respectively, P = .26. TBCC increased significantly from 0 to 72 hours in both fortified and unfortified milk, whether stored at refrigerator or room temperature,P < .001. In clinical use, fortified human milk may be stored at refrigerator temperature for up to 24 hours. The change in TBCC from 0 to 24 hours and the TBCC at 24 hours did not differ between fortified and unfortified human milk, P = .24.
There were fewer no bacterial growth cultures in the fortified compared with the unfortified milk at 0 hours (P = .007) but there were no differences between milks from 24 to 72 hours. The distribution of Gram-positive and Gram-negative bacterial flora was similar between fortified and unfortified human milk. The fortifier was cultured (×3) and no growth of bacteria was noted.
Milk osmolality was greater in fortified (408 ± 4 mOsm/kg H2O, mean ± SEM) than unfortified (302 ± 7 mOsm/kg H2O) milk, as expected; both increased similarly, by approximately 4%, from 0 to 72 hours, P = .012.
Total IgA concentrations were similar between fortified (357 ± 23 μg/mL) and unfortified (376 ± 29 μg/mL) milk, and did not increase from 0 to 72 hours, P = .3.
Overall, in the simulated design, TBCC increased in refrigerated fortified human milk from 0 to 24 hours, P = .008 (Fig2). However, TBCC were stable during the 20-hour storage in the refrigerator but increased during the subsequent 4 hours that simulated the nursery conditions of warming to room temperature and feeding. The increase in TBCC was dependent on whether fresh or frozen (stored) milk was studied. Although TBCC were significantly lower in fortified fresh human milk at 0 hours, the rise in TBCC in that group was significantly greater than in the fortified, previously frozen milk.
We showed that TBCCs in fortified human milk were significantly greater than in unfortified milk, but that TBCC increased similarly in both fortified and unfortified milk from 0 to 72 h storage. Data are scarce on the effects of fortification on the anti-infective properties of human milk. It has been shown that the inhibition of Escherichia coli growth was not lessened significantly by the addition of fortifier to frozen, and subsequently thawed human milk.11 In our study, although TBCC increased over time even in milks stored at refrigerator temperature, the magnitude of the difference in TBCC between fortified and unfortified human milk was small (∼1 log10 difference). Moreover, at 24 hours, the difference in TBCC between fortified and unfortified milk was not significant, supporting the current recommendation of using refrigerated milk within 24 h.3,4,12,13
We chose the storage conditions to provide the most extreme test of the intrinsic properties of the milk, room temperature, and storage to 72 hours. In developing countries where refrigeration is not available, milk used for premature infants may be stored at room temperature. The appropriateness of this practice has been questioned, especially if nutrient fortification is used. Furthermore, in developing regions, room temperature may be significantly greater than in the conditions imposed in the current study.
Current feeding practices in the neonatal intensive care unit often involve continuous orogastric feeding with syringes remaining at infant incubator temperature for 4 hours. Syringes usually are changed every 4 hours to avoid bacterial proliferation, especially if frozen milk is used. In this study, we have shown that TBCC significantly increased in fortified refrigerated milk that was warmed per usual nursery protocol and maintained in an infant incubator for 4 hours, simulating current nursery practice. TBCC at 24 hours, however, did not differ significantly from values obtained at baseline (0 hours). These findings may warrant evaluation of TBCC in milk that is allowed to remain within the infant incubator during feeding. Evaluation of shorter time intervals for changing feeding syringes and/or maintaining the milk syringe outside the incubator during feeding should be considered.
We showed that the total IgA concentration in milk was not affected by nutrient fortification, storage temperature, duration of storage, or milk state. Hamosh14 reported no increase in proteolytic activity of human milk stored at room temperature for 24 hours but suggested a small increase after 24 hours. Our data suggest that the addition of fortifier did not inhibit the natural antiprotease activity of human milk. The lack of a major increase in milk osmolality over time is indirect evidence that the fortifier did not enhance milk lipolysis and/or protein degradation.
Nutrient fortification and storage duration alter some of the host defense properties of human milk. The data should be interpreted with caution because the magnitude of the differences in TBCC and the changes in osmolality were small and may not be of biologic importance. Our findings support current recommendations of using fortified human milk within 24 hours of preparation.
We acknowledge support of the Wyeth-Ayerst Neonatology Research Grants Program.
We thank Nancy Hurst and the Texas Children's Hospital Lactation Program for assisting in the milk collection; the neonatal nursery staff of Texas Children's Hospital for their expertise; Linda Lamberth, Charles Imo, and Ali Bahar for laboratory assistance; J. Kennard Fraley for statistical support; Idelle Tapper for secretarial assistance; and Leslie Loddeke for editorial review.
- Received August 8, 1996.
- Accepted December 20, 1996.
Reprint requests to (R.J.S.) Children's Nutrition Research Center, Baylor College of Medicine, 1100 Bates St, Houston, TX 77030.
This work is a publication of the USDA/ARS Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, and Texas Children's Hospital, Houston, TX.
The contents of this publication do not necessarily reflect the views or policies of the USDA, nor does mention of trade names, commercial products, or organizations imply endorsement by the US Government.
- IgA =
- immunoglobulin A •
- TBCC =
- total bacterial colony counts •
- ANOVA =
- analysis of variance
- ↵Garza C, Hopkinson J, Schanler RJ. Human milk banking. In: Howell RR, Morriss EH, Pickering LK, eds. Human Milk in Infant Nutrition and Health. Springfield, IL: Charles C. Thomas; 1986:225–255
- ↵Arnold LDW, ed. Recommendations for Collection, Storage, and Handling of a Mother's Milk for Her Own Infant in the Hospital Setting. West Hartford, CT: The Human Milk Banking Association of North America, Inc; 1993
- Lucas A,
- Roberts CD
- Hernandez J,
- Lemons P,
- Lemons J,
- Todd J
- Quan R,
- Yang C,
- Rubenstein S,
- Lewiston N,
- Stevenson D,
- Kerner J
- Fleischaker JW,
- Nowak NM,
- Quinby GE
- Hack M,
- Boxerbraum B,
- Fanaroff A
- Hamosh M,
- Ellis LA,
- Pollock DR,
- Henderson TR,
- Hamosh P
- Copyright © 1997 American Academy of Pediatrics