Bacterial Growth in Expressed Mother's Milk Stored and Transported Under Different Simulated Conditions in a Tropical Country

Abstract

Background:

Research on how storage and transport of expressed human milk in a tropical country affect the milk bacterial count is limited.

Materials and Methods:

A cross-sectional descriptive study of 50 mothers of infants in a Sri Lankan tertiary neonatal unit was performed. Expressed mother's milk was divided into three bottles and kept under varied environmental conditions to simulate different storage and transport methods. Initial bacterial culture of milk was performed <30 minutes after expression, with further cultures at predetermined times. Bottles A and B were stored at room temperature and in a cool bag for the first 6 hours, respectively, and then refrigerated; and bottle C was refrigerated for 24 hours, transported in a cool bag for 6 hours, and rerefrigerated until 72 hours. Total colony counts >10⁵ colony-forming units (CFU)/mL of viable microorganisms or >10⁴ CFU/mL of either Enterobacteriaceae or Staphylococcus aureus were considered positive.

Results:

Initial culture was positive in 30% (15/50) of samples; majority, 87% (13/15), of these were S. aureus. For bottle A, 26% (13/50), 36% (18/50), 34% (17/50), and 26% (13/50) of samples were positive at 4, 6, 24, and 72 hours, respectively. For bottle B, positive cultures were found in 26% (13/50) and 17% (8/47) of samples at 24 and 72 hours, respectively. For bottle C, results were similar to bottle B.

Conclusions:

Transportation of expressed mother's milk for 6 hours in a tropical climate using a low-cost cool bag, with refrigeration at other times, maintained acceptable bacterial counts for up to 72 hours after expression. Hygienic practices at collection are extremely important as most samples with significant bacterial growth were positive on initial culture.

Introduction

Mothers' milk feeding, either as direct breastfeeding or expressed milk, is the best method of providing nutrition to a newborn infant as it has numerous advantages for both the infant and mother.¹ Some infants born prematurely or those who are sick need to spend several weeks in a neonatal unit. In most western countries such as the United States, United Kingdom, and Australia, once a mother is discharged from the hospital after the birth of her infant, she has to bring her expressed milk from home or other accommodation outside the hospital as facilities for families to stay in the hospital 24/7 are not common.^2,3 Our study in Melbourne, Australia, showed the importance of using cooler containers for transporting refrigerated mother's milk even for a short duration from the home to the hospital and the need for diligent monitoring of milk storage and transport conditions to ensure that milk remains at a low temperature.⁴

In contrast, the current practice in most Sri Lankan neonatal units (based on corresponding author's personal experience as a pediatric/neonatal doctor in Sri Lankan hospitals for 15 years) is for mothers to spend the entire time that their newborn infants are in the neonatal unit as inpatients themselves at the same hospital, but in a different ward or room from the infant. One reason mothers need (or are usually expected) to remain in the hospital is to provide the infant with fresh expressed milk for every feed. Storage and transport of expressed mother's milk are not standard care in Sri Lanka. In Sri Lankan public hospitals, most mothers express their milk by hand into sterilized metal cups.

Study context

This article describes bacterial growth in human milk samples stored and transported under different conditions in Sri Lanka, an island situated within the tropics where the mean annual temperature varies between 27°C in coastal lowlands and 16°C in central highlands.⁵ Ragama, where this study was conducted, is situated in the coastal lowlands. The average temperature is 30–32°C, with a maximum of 38.4°C recorded in the past 40 years.⁶

Bacteriological safety of expressed mother's milk

A 2019 review analyzing 242 studies from 38 countries identified the presence of a total of 820 species in human milk microbiota, with Staphylococcus aureus and Streptococcus agalactiae being the most common bacteria.⁷ Minimizing the possibility of expressed human milk containing pathogenic bacteria is extremely important for neonatal units. A hazard analysis approach to handling expressed human milk in a neonatal unit in Belgium identified critical control points that can be optimized to ensure infant safety: hand hygiene, cleaning/disinfection of breast pump equipment, need to refrigerate or freeze the expressed milk within 1 hour, transportation in an isothermal bag to the hospital, maintenance of the home refrigerator or freezer, and steps for ensuring correct identification of a milk sample to be given to a particular infant.⁸

Previous literature on this topic was searched with the aim of identifying studies that looked at bacterial growth in mother's milk, expressed under a range of conditions and stored under different temperatures for varied periods, to determine the safety (using cutoff values for bacterial counts) of feeding the milk to hospitalized infants. Studies or study components that looked at the bacteriology of donor milk in milk banks or those where storage of the expressed milk was only in a freezer were not included. All original articles similar to ours, which were identified by searching PubMed using the terms “expressed breast milk,” “expressed human milk,” “storage time,” and “bacteria,” are included in Table 1. Relevant original articles cited by those identified on the PubMed search and mentioned as being similar when accessing these identified articles were also included (last update September 10, 2020).

Table 1.

Safe Storage Duration/Microbiology of Expressed Mother's Milk Based on Storage Temperature

Author (year)^Ref.	Country	No. of samples	No. of mothers	Methodology and/or cutoff values for SC	Study findings and/or recommendations for a safe storage duration based on bacterial growth
Larson et al. (1984)⁹	United States	30	30	Bacterial counts checked at time of expression and after 24 and 48 hours of refrigeration.	No differences in bacterial counts
Sosa and Barness (1987)¹⁰	United States	41	41	Milk cultured within 1 hour of collection and placed in the refrigerator. Further cultures at 24, 48, 72, 96, and 120 hours. Skin and nipple of breast were swabbed.	No bacteria in 8 samples; bacteria in 33 samples were similar to skin/nipple; initial concentration of bacteria in milk was low; bacterial counts decreased over 5 days of refrigeration
Olowe et al. (1987)¹¹	Nigeria	87	63	Cultures at six-hour intervals for 24 hours while stored in a domestic refrigerator.	Bacteria multiplied in 50 samples and their growth was inhibited in 32 of the samples; mean bacterial count during the 24 hours was not significantly different from the initial count
Nwankwo et al. (1988)¹²	Nigeria	NM	20	SC = mean bacterial counts either >10⁴ CFU/mL or becoming significantly higher than the initial counts at the time of expression.	Storage recommendations at room temperature (27–32°C): mature milk from mothers with term infants—6 hours; colostrum at term—12 hours; and mature milk from mothers with preterm infants—4 hours
Ajusi et al. (1989)¹³	Kenya	NM	25	SC = bacterial counts >10³ CFU/mL.	Storage recommendations: 8 hours at room temperature (10–24.7°C); bacterial counts decreased during storage
Hamosh et al. (1996)¹⁴	United States	NM	16	SC = 2.5 × 10³ CFU/mL for nonpathogenic maternal skin flora and 10²–10³ for Staphylococcus aureus.	Storage recommendations: 24 hours at 15°C; 4 hours at 25°C; and should not be stored at 38°C
Igumbor et al. (2000)¹⁵	Zimbabwe	NM	24	Definite cutoff values are not given; bacterial counts were studied when milk was stored at different temperatures.	Nonpathogenic skin flora found in all fresh samples; storage recommendations: 24 hours in the refrigerator (4–10°C); 8 hours at room temperature (15–27°C); and 4 hours at high temperature (30–38°C)
Slutzah et al. (2010)¹⁶	United States	36	36	Fresh milk samples were divided and stored at 4°C for 0, 24, 48, 72, and 96 hours.	Total and Gram-negative bacterial counts were similar at each time point; Gram-positive bacterial counts reduced from 0 to 96 hours
Bertino et al. (2013)¹⁷	Italy	15	17	Bacterial count analyzed at 0, 24, 48, 72, and 96 hours following refrigeration.	96-Hour-old refrigerated milk (6.8°C ± 1.1°C) was equivalent to fresh milk in bacterial count
Ukegbu et al. (2013)¹⁸	Nigeria	20	20	Milk was kept at 30°C or 26°C (in a cold water container) and tested at 0, 3, 6, and 9 hours. SC = bacterial counts of >10⁴ CFU/mL.	Bacterial count was lower in samples immersed in a cold water container (26°C); storage recommendation: 9 hours at 30°C
Fogleman et al. (2018)¹⁹	United States	12	12	Total aerobic bacterial count assessed at 0, 1, 2, 7, 14, and 21 days.	Significant increases in bacterial count were noted when stored for 14 days at 4°C and 2 days at 24°C
Widjaja et al. (2019)²⁰	Indonesia	30	30	Milk samples were kept at room temperature (30–32°C) for 2, 4, and 6 hours and bacterial counts assessed.	At 0 hours, 2 (7%) samples were contaminated; all <10⁵ CFU/mL; at 2 hours, 16 (53%) samples were contaminated; all <10⁵ CFU/mL; at 4 hours, 27 (90%) samples were contaminated; all >10⁵ CFU/mL; and at 6 hours, 30 (100%) samples were contaminated; all >10⁵ CFU/mL

CFU, colony-forming units; NM, not mentioned; SC, significant contamination.

In addition to the previous search terms on PubMed, the terms of “hygienic practices,” “home vs. hospital,” and “manual vs. pump” were used to identify articles that have assessed the effects of different places and techniques of expression and hygienic practices on bacterial counts of milk. These articles are summarized in Table 2. Studies on storage or transport of human milk have not been conducted in Sri Lanka previously.

Table 2.

Effects of Hygienic Practices, Place, and Method of Expressing Mother's Milk on Suitability for Administration to Hospitalized Newborn Infants

Author (year)^Ref.	Country	No. of samples	No. of mothers	Conditions/methods of the study	Findings
Jones et al. (1979)²¹	United Kingdom	14	7	Bacterial counts compared with sterile collection bottles used as they are or rinsed with 1% hypochlorite solution	Rinsing with hypochlorite solution significantly reduced bacterial counts at 4 hours when stored at 4°C
Pittard et al. (1991)²²	United States	NM	16	Each woman used sterile and clean containers for collection and both manual and mechanical techniques for expression; milk samples from each situation were cultured	No difference in significant growth for milk collected in clean versus sterile containers or collected manually versus by breast pump and milk collected initially versus later in the expression
el-Mohandes et al. (1993)²³	United States	108	40	Milk expressed at home brought in frozen, thawed for 24 hours in the refrigerator, and then tested bacteriologically; instructions on hygienic practices were given before collection; those with high milk bacterial counts were reinstructed and milk retested	A total of 12% (13/108) of samples had no growth; in 62% (25/40) of mothers, the bacterial count was >3 × 10³ CFU/mL in at least one milk sample; reinstruction improved contamination in 32% (7/22) of participants
Boo et al. (2001)²⁴	Malaysia	92	28	Manual expression of milk versus use of a breast pump at home and the hospital	Higher bacterial counts when expressed by breast pump compared with manual expression; no difference in bacterial counts when in the hospital; and higher bacterial counts with pumps when expressed at home
Rozolen et al. (2006)²⁵	Brazil	90	32	Appropriateness for raw consumption of milk expressed at home; appropriate if the mesophilic bacterial count was <2,500 CFU/mL and potential pathogens were not detected	A total of 36 (40%) milk samples expressed at home were not appropriate for raw consumption
Karimi et al. (2012)²⁶	Iran	244	15	Effect of educational intervention on bacterial contamination of milk; samples taken by hands, breast, pumps, mother's milk, and milk storage containers; contamination before and after education on hygienic practices was compared	Contamination of at least one sample: 80% before and 36% after; contamination of milk samples: 25% before and 8% after; and main sources of contamination: milk containers and pumps
Serra et al. (2013)²⁷	Argentina	280	53	Compare the risk of microbiological contamination of milk expressed at home versus a health care facility	A total of 139 (50%) samples were contaminated; contamination was more frequent at home than at the health care facility 60% versus 40%; p = 0.0008)
Haiden et al. (2016)²⁸	Austria	1,466	NM	Standard versus strict hygienic practices at home and hospital; all mothers expressing milk for infants in the neonatal intensive care unit over a 2-year period were included	All mothers used breast pumps; higher bacterial counts were observed when expression was done at home, irrespective of the infection control regimen used
Carre et al. (2018)²⁹	France	NM	29	Instructions on milk collection and breast pump decontamination were given to mothers; bacterial counts of milk samples were tested; questionnaire and interview were conducted for maternal adherence to instructions	Mothers with increased bacterial counts in their milk frequently declared several divergences from the guidance

CFU, colony-forming units; NM, not mentioned.

Recommendations from human milk banks and their relevance

Human milk banks across the globe have set standards on acceptable limits of bacterial content in expressed human milk that is donated for infants of other mothers, for example, the Human Milk Banking Association of North America, PREM Bank (Perron Rotary Express Milk Bank) in Perth, Australia, and Mercy Hospital Breastmilk Bank, in Melbourne, Australia.^30–33 There are guidelines published by the National Institute for Health and Care Excellence (NICE), United Kingdom too.³⁴ Guidelines prepared for temperate climates have limited applicability in tropical countries such as Sri Lanka.

Furthermore, there are no standards for what is considered an acceptable level of bacterial content when a particular infant is fed freshly expressed human milk from its own mother. Even in developed countries, when a mother is providing expressed milk for her own infant in a neonatal unit, the milk is not tested microbiologically before administration. We therefore decided to use values that were most lenient, as given by NICE, United Kingdom, as the cutoffs for this study: that is, a count of >10⁵ colony-forming units (CFU)/mL for total viable microorganisms or >10⁴ CFU/mL for Enterobacteriaceae or S. aureus as a significant growth, which constituted a positive sample.³⁴

Study rationale

Fresh human milk for each feed is the optimal nutrition that can be provided to infants in a neonatal unit.³⁵ However, if social and other circumstances make it necessary for a mother to be at home and bring in milk expressed at home for her newborn in the hospital, it is important to understand the best way that this can be done.

Therefore, the objective of this study was to explore the bacterial content in expressed human milk transported and stored using different methods, under Sri Lankan environmental conditions, and to identify which methods and duration measures would maintain bacterial count levels within acceptable limits over 72 hours from the time of expression.

Materials and Methods

Design

A descriptive exploratory study was conducted at the University Neonatal Unit of Colombo North Teaching Hospital, Ragama, Sri Lanka, and the Faculty of Medicine, University of Kelaniya, Ragama. As it would be unethical to change current practice and send mothers home before the infant's discharge from hospital for the sake of a study, simulated conditions were used.

Eligibility

Women whose infants were currently receiving care as inpatients at the above neonatal unit, who had already been trained in standard hygienic practices for human milk expression, and who were able to express 30 mL of milk within 30 minutes were included in the study. Those women whose milk supply was inadequate to provide 30 mL of expressed human milk without compromising the amount available for her infant were excluded. Women taking antibiotics within the previous 48 hours were also excluded along with women who had nipple damage or nipple/breast infection or a skin condition.

Sample size and recruitment

The number of mothers who provided expressed human milk samples for similar previous studies ranged from 12 to 63 individuals.^{11,12,14,18,19} A bacteriology study of human milk expressed at home and hospital showed that a significant number of pathogenic organisms were identified by 4 hours when milk was kept at 25°C in 11% of those samples expressed at home. Therefore, using the calculation N = 4 × Zα² × p (1 − p)/w² (Zα = confidence level, w = width of confidence interval, and p = prestudy estimate of the proportion to be measured), the sample size required to show the percentage of milk samples that were positive as a reflection of what it would be in a similar larger population, with a confidence level of 95% and confidence interval of ±10%, was 37.³⁶ We planned to recruit 50 mothers to allow for issues such as laboratory staffing and equipment malfunction. Each mother's expressed milk was bacteriologically tested 14 times under 3 storage/transport protocols. A research assistant not involved in clinical care recruited the participants. Consecutive convenience sampling was used.³⁷

Flow of samples and microbiological testing

Mothers expressed their milk in the neonatal unit, simulating home conditions, and the microbiological testing was done at the Department of Microbiology, Faculty of Medicine, University of Kelaniya (located within 10 minutes walking distance from the hospital). The 30 mL of milk provided by each participant was divided equally into 3 bottles, A, B and C, which were each placed inside a Ziploc bag along with a data logger thermometer. The bottles remained in this bag for 72 hours of monitoring, commencing from completion of milk expression, except when samples were being obtained for culture. The protocols for the three scenarios were prepared to simulate expressed milk brought in to the hospital within 6 hours (A and B) or the following day after being stored at home for 24 hours (C) and used in the unit until 72 hours from time of expression (Fig. 1). Immediate transport of required bottles ensured that initial culture was done within 30 minutes.

FIG. 1.

Sample flowchart.

Simulation of home conditions

Mothers followed standard advice already provided regarding hand hygiene and breast preparation for milk expression. The containers used for collection and storage of expressed milk were glass feeding bottles sterilized in the neonatal unit using boiling water on a stove. The staff kitchen refrigerator was used for storage before transportation (for bottle C) rather than the well-controlled medicine/vaccine refrigerator of the neonatal unit. In the laboratory, milk was stored in their sample refrigerator, which simulated a potential future milk refrigerator in the neonatal unit. When placing samples in the refrigerator, the door compartment was avoided.

Microbiological testing

Samples were inoculated onto sheep blood agar and MacConkey agar plates with a standard 1 μL wire loop at specified intervals. The plates were read for growth of microorganisms after overnight incubation at 37°C in the incubator. The total aerobic bacterial colony count (total viable organisms) and the colony counts for each of the organisms in a particular sample were calculated manually by counting the CFU and interpreted as CFU/mL. Identification of organisms was done by the colony morphology, gram stain, and routine biochemical tests such as the catalase test, staphylase test, coagulase test, oxidase test, streptococcal grouping by agglutination test, indole test, and inoculation into KIA (Kligler iron agar).

Details of data collection tools

Demographic, pregnancy, and birthing details of the mother were obtained by face-to-face interview at recruitment.

Temperature assessment instruments

(1)

Refrigerator thermometers (CaterChef brand digital fridge/freezer thermometer) in the staff refrigerator of the neonatal unit and laboratory refrigerator were used to maintain daily temperature logs throughout the study period.

(2)

Temperature was recorded every 5 minutes using the Tinytag Transit 2 temperature data logger TG-4080 in Ziploc bags next to the milk bottles. The recordings were downloaded to the computer after 72 hours.

Data analysis

Analysis was conducted using SPSS, version 16.0, and descriptive statistics (means and proportions, etc.) presented.

Ancillary study

To test mother's milk that was actually expressed at home and transported to the hospital, we conducted a second smaller study 2 months after the main study. We recruited 11 consecutive eligible and consenting women, who had not participated in the main study, at the time of their infants' discharge from hospital. The eligibility criteria were the family could travel from the home to hospital within 1 hour, the infant would be receiving expressed mother's milk at home, and they would be attending the follow-up clinic within 2 weeks of discharge. Information provision and consent were similar to the main study. Mothers boiled the collection bottles at home before use and were asked to follow written instructions for collection and storage of the three bottles of milk—A, B, and C. Ziploc bags and data loggers were used as in the main study. The only difference in the flow of samples from the main study was that milk in bottle C was collected the day before coming to the hospital and it remained in the participant's home refrigerator for 24 hours. The mothers brought all three bottles to the hospital clinic in the specified containers. The bottles were immediately taken to the laboratory. Telephone reminders were provided to participants.

Ethical considerations and approval

Ethical approval was obtained from the Ethics Review Committee of the Faculty of Medicine, University of Kelaniya, Ragama, Sri Lanka (P/237/09/2017), and the Human Research Ethics Committee of the College of Science, Health and Engineering, La Trobe University (Melbourne, Australia).

Results

Demographic characteristics

All eligible mothers were approached during the study period (February to August 2018) except for approximately 12 weeks when laboratory staff were intermittently engaged in union action and recruitment was paused. All those approached consented to participate (n = 51), but one baby's condition deteriorated overnight after a woman consented, so this woman was withdrawn, leaving 50 women. Demographic and birthing details are available in Table 3.

Table 3.

Mother's Characteristics (N = 50)

	Mean (SD)	Range
Demographic variables
Age (years)	29.7 (4.5)	19–38
Gestation (weeks)	33.7 (5.1)	24–41
	N	%
Mode of birth
Normal vaginal	23	46
Assisted vaginal	3	6
Cesarean section	24	48
Education level
Completed a degree or higher	6	12
Completed a certificate/diploma/apprenticeship	3	6
Completed secondary school to year 12	21	42
Not completed secondary school	20	40
Parity
Primiparous	32	64
Multiparous	18	36
Method of milk expression
Manual (hand expression)	50	100
Breast pump	0	0

SD, standard deviation.

Microbiology

Main study: in bottle A, three samples (6%) did not have any growth even after 72 hours, while 30% (15/50) of samples were positive initially. Of the 28% (13/47) of samples that were positive at 48 and 72 hours, 85% (11/13) were positive from time of collection. The organism found in 87% (13/15) of samples that were positive on initial culture was S. aureus. In bottles B and C, the number of positive samples gradually went down over time with refrigeration. Further details are available in (a) in Table 4.

Table 4.

Microbiological Results

	Number of samples (percentage)
	0 Hours	4 Hours	6 Hours	24 Hours	30 Hours	48 Hours	72 Hours
(a) Microbiological results of the main study (N = 50)
Bottle A					X
No growth	3 (6)	3 (6)	3 (6)	3 (6)		3 (6)	3 (6)
Insignificant growth	32 (64)	34 (68)	29 (58)	30 (30)		31 (66)	31 (66)
Significant growth	15 (30)	13 (26)	18 (36)	17 (17)		13 (28)	13 (28)
Culture missed^a	0 (0)	0	0	0		3 (6)	3 (6)
Bottle B	X	X			X
No growth			3 (6)	3 (6)		3 (6)	4 (8)
Insignificant growth			33 (66)	34 (68)		34 (72)	35 (74)
Significant growth			14 (28)	13 (26)		10 (21)	8 (17)
Culture missed^a			0	0		3 (6)	3 (6)
Bottle C	X	X	X
No growth				4 (8)	2 (4)	2 (4)	5 (11)
Insignificant growth				32 (64)	35 (70)	34 (72)	34 (72)
Significant growth				14 (28)	13 (26)	11 (21)	7 (15)
Culture missed^a				0	0	3 (6)	4 (8)
(b) Microbiological results of the ancillary study (N = 11)
Bottle A					X
No growth	1 (9)	0 (0)	0 (0)	0 (0)		0 (0)	0 (0)
Insignificant growth	4 (36)	5 (46)	3 (27)	3 (27)		3 (27)	2 (18)
Significant growth	6 (54)	6 (54)	8 (73)	8 (73)		8 (73)	9 (81)
Bottle B	X	X			X
No growth			0 (0)	0 (0)		0 (0)	0 (0)
Insignificant growth			3 (27)	5 (46)		5 (46)	6 (54)
Significant growth			8 (72)	6 (54)		6 (54)	4 (36)
Bottle C	X	X	X
No growth				0 (0)	0 (0)	0 (0)	0 (0)
Insignificant growth				8 (72)	7 (63)	7 (63)	7 (63)
Significant growth				3 (27)	4 (36)	4 (36)	4 (36)

Culture missed: performance of culture inadvertently missed by laboratory staff.

X, no culture as per the study protocol.

Ancillary study (milk brought from home): (b) in Table 4 provides the percentages of samples with positive bacterial counts at various time points under different conditions. Initial culture was positive in 54% (6/11) of samples, with four of the positive samples having significant growth of S. aureus and two of Escherichia coli.

Temperature

Temperature results were obtained by perusing data logger recordings and refrigerator temperature logs. Expressed milk was collected in the hospital at room temperature that ranged from 23°C to 28°C with a mean of 25.9°C. During the 6 hours that bottle A was at room temperature, the maximum temperatures to which each sample was exposed over 6 hours ranged from 26°C to 30.6°C, with a mean temperature of 27.5°C for recordings made at 6 hours. The temperature recorded by the data loggers, when in the laboratory refrigerator, ranged from 0.5°C to 6.2°C. The mean refrigerator temperature was 4.1°C at 24 hours.

Discussion

The most prominent finding from this study was that the majority of milk samples with positive bacterial counts showed growth at initial culture soon after collection. Therefore, it was not storage and transport that were the main contributors to bacterial growth in expressed human milk, but rather bacteria already present in the milk or introduced during expression/collection. Human milk contains a wide range of microbiota.^7,38 Strict versus standard hygienic practices have not shown a difference in milk bacterial counts.²⁸ Education and reminders about hygienic practices have consistently resulted in lower bacterial counts in milk in other studies.^23,26,29 The evidence regarding use of pumps versus hand expression and bacterial counts is equivocal.^22,24 In Sri Lanka, most mothers expressing milk for hospitalized infants do so by hand. Higher contamination in milk expressed at home rather than the hospital has been noted in several studies.^24,27,28 Our ancillary study showed a similar trend. This raises the possibility of lower adherence to hygienic practices at home where mothers have many other commitments and distractions. Therefore, if Sri Lankan mothers are to be allowed to express milk at home for hospitalized infants, the importance of adhering to recommended hygienic practices needs to be stressed.

There was a rise in bacterial counts (statistically nonsignificant, probably due to the small sample size) from 4 to 6 hours when milk was kept at room temperature. This rise in bacterial counts suggests that 6 hours may be too long—under Sri Lankan environmental conditions of our study (26–30.6°C)—for mother's milk to be kept at room temperature if it is to be stored further in the refrigerator and administered later to infants in the neonatal unit. Previous studies recommend 4 hours or less as a safe storage duration when room temperature is 27–32°C,¹² 30–38°C,¹⁵ 25°C,¹⁴ or 24°C¹⁹ and 8–9 hours with temperatures of 15–27°C¹⁵ or 30°C.¹⁸ The Human Milk Banking Association of North America and the World Health Organization recommend 4 hours at room temperature.^35,39

Milk samples that were in the cool bag for the 6 hours of simulated transport did not show an increase in the number of samples with significant bacterial counts, which implies that if expressed mother's milk is to be transported in Sri Lanka, it should be done using a cool bag that would maintain bacterial counts at static levels. Expressed human milk that was stored and transported to simulate conditions where mothers would store milk expressed at home for over 1 day in the home refrigerator and travel with it to the hospital the next day, with the milk in a cooler container, also had shown acceptable bacterial counts for the process of storage and transport. In the two sets of milk samples that were always stored in the refrigerator and transported only using a cool bag, the number of samples with significant bacterial counts dropped by 72 hours to about half of the number that had significant counts at the time of collection. Human milk that has not been pasteurized is biologically active and has numerous antipathogenic properties that could have contributed, along with refrigeration, to the drop in number of positive samples.¹⁰ This was an encouraging result for those mothers wishing to travel from home to the hospital. Several authors have reported that bacteriology of expressed mother's milk has remained stable during refrigeration for up to 96 hours,^16,17 120 hours,^10,35 and 6 days,¹⁹ while other studies have monitored the milk for shorter refrigeration periods, 24 hours^11,15 and 48 hours,⁹ and deemed these periods as safe.

Implications

Our findings suggest that 4 hours appears to be a safe duration at room temperature for expressed human milk for hospitalized infants in Sri Lanka. Bacteriology of milk samples stored and transported using economical methods for 72 hours suggests that mothers who wish to travel from home can safely bring milk expressed and stored at home to the hospital. The importance of adhering to recommended hygienic practices at home needs to be stressed to such mothers as most samples that had significant bacterial counts had these from the time of collection.

Long-term surveillance of infection-related conditions, for example, necrotizing enterocolitis, should be in place if the current practice of providing only fresh expressed mother's milk to infants in the neonatal unit is changed to allow mothers to bring stored mother's milk to the hospital to monitor safety of this new practice on infants' health.

Limitations

This study has a small sample size and was conducted in a single center: a larger sample size under real, rather than simulated, conditions could enable firmer conclusions and should be replicated in other Sri Lankan centers to make generalized recommendations for the entire country.

Conclusions

In this simulation study, it was found that mother's own milk, which was expressed for later use by infants in neonatal units, could be transported in an economical cool bag for up to 6 hours and kept refrigerated for the remainder of 72 hours from the time of expression in a tropical climate without compromising microbiological safety.

Footnotes

Authors' Contributions

Authors of this article contributed substantially to the concept of the study and data collection, analysis, or interpretation. All contributed to drafting and revising the article and provided final approval.

Acknowledgments

The authors wish to acknowledge the contribution made by mothers of infants in the neonatal unit who willingly participated in the study.

Disclosure Statement

All authors have no conflicts of interest to declare.

Funding Information

This study was funded by a research grant from the University of Kelaniya, Sri Lanka.

References

Victora

, Bahl

, Barros

, et al. Breastfeeding in the 21st century: Epidemiology, mechanisms, and lifelong effect. Lancet, 2016; 387:475–490.

Anderson

. Families kept apart: Barriers to parents' involvement in their baby's hospital care. London: Bliss, 2017.

Australasian Health Infrastructure Alliance. Australasian health facility guideline. North Sydney: Australasian Health Infrastructure Alliance, 2016.

Rodrigo

, Amir

, Forster

, et al. Human milk expression, storage and transport by women whose infants are inpatients at a tertiary neonatal unit in Melbourne, Australia: An exploratory study. Adv Neonatal Care 2020. [Epub ahead of print]; DOI: 10.1097/ANC.0000000000000825.

Department of Metereology Sri Lanka. Climate of Sri Lanka. 2017. Available at www.meteo.gov.lk/index.php?option=com_content&view=article&id=94&Itemid=310&lang=en (accessed August 31, 2020).

United Nations Human Settlements Programme (UN-Habitat) Negombo, Sri Lanka: Climate change vulnerability assessment, 2011. Kenya: UN-Habitat, 2013.

Togo

, Dufour

, Lagier

, et al. Repertoire of human breast and milk microbiota: A systematic review. Future Microbiol, 2019; 14:623–641.

Cossey

, Jeurissen

, Thelissen

, et al. Expressed breast milk on a neonatal unit: A hazard analysis and critical control points approach. Am J Infect Control, 2011; 39:832–838.

Larson

, Zuill

, Zier

, et al. Storage of human breast milk. Infect Control, 1984; 5:127–130.

10.

Sosa

, Barness

. Bacterial growth in refrigerated human milk. Am J Dis Child, 1987; 141:111–112.

11.

Olowe

, Ahmed

, Lawal

, et al. Bacteriological quality of raw human milk: Effect of storage in a refrigerator. Ann Trop Paediatr, 1987; 7:233–237.

12.

Nwankwo

, Offor

, Okolo

, et al. Bacterial growth in expressed breast-milk. Ann Trop Paediatr, 1988; 8:92–95.

13.

Ajusi

, Onyango

, Mutanda

, et al. Bacteriology of unheated expressed breast milk stored at room temperature. East Afr Med J, 1989; 66:381–387.

14.

Hamosh

, Ellis

, Pollock

, et al. Breastfeeding and the working mother: Effect of time and temperature of short-term storage on proteolysis, lipolysis, and bacterial growth in milk. Pediatrics, 1996; 97:492–498.

15.

Igumbor

, Mukura

, Makandiramba

, et al. Storage of breast milk: Effect of temperature and storage duration on microbial growth. Cent Afr J Med, 2000; 46:247–251.

16.

Slutzah

, Codipilly

, Potak

, et al. Refrigerator storage of expressed human milk in the neonatal intensive care unit. J Pediatr, 2010; 156:26–28.

17.

Bertino

, Giribaldi

, Baro

, et al. Effect of prolonged refrigeration on the lipid profile, lipase activity, and oxidative status of human milk. J Pediatr Gastroenterol Nutr, 2013; 56:390–396.

18.

Ukegbu

, Uwaegbute

, Ijeh

, et al. Bacterial load in expressed and stored breast milk of lactating mothers in Abia state, Nigeria. Afr J Food Agric Nutr Dev, 2013; 13:8139–8154.

19.

Fogleman

, Meng

, Osborne

, et al. Storage of unfed and leftover mothers' own milk. Breastfeed Med, 2018; 13:42–49.

20.

Widjaja

, Hardiyani

, Hanindita

, et al. Microbiological assessment of fresh expressed breast milk on room temperature at Dr. Soetomo Hospital, Neonatal Unit. Folia Med Indonesia, 2019; 55:30–36.

21.

Jones

, Jennison

, D'Souza

. Bacterial contamination of expressed breast milk. Br Med J, 1979; 2:1320–1322.

22.

Pittard

, 3rd, Geddes

, Brown

, et al. Bacterial contamination of human milk: Container type and method of expression. Am J Perinatol, 1991; 8:25–27.

23.

el-Mohandes

, Schatz

, Keiser

, et al. Bacterial contaminants of collected and frozen human milk used in an intensive care nursery. Am J Infect Control, 1993; 21:226–230.

24.

Boo

, Nordiah

, Alfizah

, et al. Contamination of breast milk obtained by manual expression and breast pumps in mothers of very low birthweight infants. J Hosp Infect, 2001; 49:274–281.

25.

Rozolen

, Goulart

, Kopelman

. Is breast milk collected at home suitable for raw consumption by neonates in Brazilian public neonatal intensive care units?. J Hum Lact, 2006; 22:418–425.

26.

Karimi

, Eslami

, Shamsi

, et al. The effect of educational intervention on decreasing mothers' expressed breast milk bacterial contamination whose infants are admitted to neonatal intensive care unit. J Res Health Sci, 2012; 13:43–47.

27.

Serra

, Teves

, Lopez de Volder

, et al. Comparison of the risk of microbiological contamination between samples of breast milk obtained at home and at a healthcare facility. Arch Argent Pediatr, 2013; 111:115–119.

28.

Haiden

, Pimpel

, Assadian

, et al. Comparison of bacterial counts in expressed breast milk following standard or strict infection control regimens in neonatal intensive care units: Compliance of mothers does matter. J Hosp Infect, 2016; 92:226–228.

29.

Carre

, Dumoulin

, Jounwaz

, et al. Maternal adherence to guidance on breast milk collection process. Arch Pediatr, 2018; 25:274–279.

30.

Rodrigo

, Amir

, Forster

. Review of guidelines on expression, storage and transport of breast milk for infants in hospital, to guide formulation of such recommendations in Sri Lanka. BMC Pediatr, 2018; 18:271.

31.

Human Milk Banking Association of North America (HMBANA). Guidelines for the establishment and operation of donor human milk bank. Fort Worth, Texas: HMBANA, 2015.

32.

Hartmann

, Pang

, Keil

, et al. Best practice guidelines for the operation of a donor human milk bank in an Australian NICU. Early Hum Dev, 2007; 83:667–673.

33.

Mercy Health. Pre and post pasteurisation bacteriological testing of donor breastmilk. Melbourne: Mercy Hospital Breast Milk Bank, 2010.

34.

National Institute for Health and Care Excellence (NICE). Donor milk banks: Service operation. In Vol. Clinical guideline (CG93). London: NICE, 2010.

35.

Jones

Best Practice for Expressing, Storing and Handling Human Milk in Hospitals, Homes, and Child Care Settings, 4th ed. Fort Worth, Texas: Human Milk Banking Association of North America, 2019.

36.

Pratap

. Sample size in clinical research, the number we need. Int J Med Sci Public Health, 2012; 1:5–9.

37.

Taherdoost

. Sampling methods in research methodology; how to choose a sampling technique for research. Int J Acad Res Manage, 2016; 5:18–27.

38.

Fitzstevens

, Smith

, Hagadorn

, et al. Systematic review of the human milk microbiota. Nutr Clin Pract, 2017; 32:354–364.

39.

World Health Organization

ICEF, Wellstart

International

. Baby-friendly hospital initiative: Revised, updated and expanded for integrated care. Section 3-Breastfeeding promotion and support in a baby-friendly hospital: A 20-hour course for maternity staff. Geneva: World Health Organization and UNICEF, 2009.