Measuring Milk Synthesis in Breastfeeding Mothers

Abstract

Aim:

Breastfeeding mothers expressed their breasts hourly for periods up to 7 hours, without exogenous oxytocin, to determine the consistency of milk removal and the relationship between the volume of milk removed and the mother's 24-hour milk production.

Methods:

Mothers (n = 20) measured their 24-hour milk production from each breast by test-weighing their babies at home. Mothers came to the research laboratory for a day visit and expressed milk from either both or one breast (left or right) for 10–15 minutes, every hour for 2–7 hours. Mothers were not instructed to restrict breastfeeding prior to the first (0-hour) expression.

Results:

It was found that the milk volume at 1 hour (59.7 ± 36.5 and 46.5 ± 25.6 mL/hour for the right and left breast, respectively) was significantly (p < 0.05) higher than the milk volume from subsequent hourly intervals. The average hourly volumes obtained from the 2^nd to 7^th hour (18.0 ± 3.1 and 14.0 ± 4.8 mL/hour for the right and left breast, respectively) were not significantly different from the average hourly rates calculated for the 24-hour milk production (18.4 ± 6.3 and 15.5 ± 5.1 mL/hour for the right and left breast, respectively). Furthermore, local inhibition was not observed during hourly expressions of up to 7 hours.

Conclusion:

The average hourly volumes of milk from the 2^nd to 7^th hour expressions could provide an estimate of the rate of milk secretion. Further study is required to fully validate these findings.

Introduction

Milk production in suckling mammals, such as goats, rabbits, and sheep, has been measured by the administration of exogenous oxytocin and then, at a given time interval, milking again following oxytocin administration to determine the hourly rate of milk production. Residual milk always remains in the mammary gland following the removal of milk by the suckling neonate, a breast pump, or manual milking. Thus variations in residual milk can complicate the measurement of milk production in mammals. Exogenous oxytocin has been used in animals to obtain a more consistent level of residual milk and thus a more reliable estimate of milk production.¹ However, the doses of oxytocin used have been nonphysiological, and it has been demonstrated that exogenous oxytocin can increase milk synthesis.¹

Linzell² demonstrated that a proportion of the residual milk could be removed by minimum effective “physiological” dose of exogenous oxytocin. In addition, when mammary glands of goats were subjected to milk expression at 1-hour intervals, the quantity and quality of milk were equivalent to those of milk obtained from glands that were subjected to twice-daily milking. This indicates (1) that the mammary myoepithelium is able to respond to small exogenous doses of oxytocin and (2) that hourly expression is a practical procedure for studying milk synthesis without the use of nonphysiological doses of exogenous oxytocin.

The aim of this study was to express mothers' breasts, hourly for periods up to 7 hours and without exogenous oxytocin, to determine the volume of milk removed and the relationship between the volume of milk removed each hour to the mother's 24-hour milk production.

Materials and Methods

Mothers

Mothers (n = 20) of healthy term babies were recruited through either the lactation clinic at Texas Tech University (Amarillo, TX) (n = 15) or through the Australian Breastfeeding Association WA Branch (Perth, WA, Australia) (n = 5). All mothers supplied written informed consent to participate in the studies, which were approved by the Human Research Ethics Committee of The University of Western Australia, Crawley, WA, Australia and the Institutional Review Board at Texas Tech University.

The same experimental protocol was used in Texas and Western Australia.

Hourly breast expression session

All mothers came to the research laboratory for a day visit, arriving in the morning. Mothers were not instructed to restrict breastfeeding prior to the first (0-hour) expression for the hourly breast expression session. Only mothers whose babies could be fed from a bottle were recruited for the study. This ensured that the babies were not deprived of breastmilk over the duration of hourly breast expression sessions.

During the visit, the mother expressed milk either from both breasts (n = 15) (simultaneous/sequentially) or from one breast (left or right) (n = 5) for 10–15 minutes, every hour for up to 7 hours, using an electric breast pump (Symphony, Medela AG, Baar, Switzerland). It was found that to obtain reliable results it was necessary for the mothers to remain at the study site for a 4–6-hour period (i.e., five to seven expressions). The expressed milk was weighed and was given back to the mother to feed her baby. Mothers participated in one (n = 9), two (n = 7), or three (n = 4) pumping sessions, depending on their availability. One mother (M4) participated in three sessions at approximately monthly intervals, of which the first two sessions commenced in the morning and the third session commenced in the evening.

24-hour milk production measurement

Participating mothers were invited to measure their 24-hour milk production using the following protocol. Of these, 11 completed the 24-hour production protocol (n = 9 on both breasts, n = 2 on the left breast only).

Milk yield was determined for each breast by test weighing the infant before and after each breastfeed from each breast over a 24-hour period using an electronic balance (BabyWeigh Scales, Medela AG) in each mother's home³ within 2 weeks prior to undertaking the hourly breast expression session. Mothers were instructed on the use of the balance, were provided with written instructions, and were required to demonstrate the use of the balance to the visiting researcher. Mothers also recorded the duration and frequency of breastfeeds from each breast. The hourly rate of milk production for each breast was calculated as the 24-hour milk production from that breast divided by 24.

Statistics

Results are presented as mean ± SD values unless otherwise stated. All analyses were carried out using the R analysis program⁴ and using the base and nlme⁵ packages. Linear mixed models were used to compare the volumes at each hourly expression and to compare the hourly milk yield and the 24-hour milk production measurement.

For the comparison of volumes at each hourly expression, the data for all women, for all sessions in which they participated, were analyzed. Random effect was the mother, and the models considered had expression number (as factor) as the main explanatory variable. Other explanatory variables considered were the session number (1, 2, or 3) and breast (left or right).

For comparison of the hourly volumes from the breast expression sessions with the hourly volume calculated from the 24-hour milk production measurement, only the data from the 11 women for whom 24-hour production data were available were used. Random effect was the mother, and the models considered had hourly volume calculated from the 24-hour milk production measurement as the main explanatory variable. Other explanatory variables considered were the session number (1, 2, or 3) and breast (left or right).

Results

The age of 14 of the mothers was 31 ± 5 years, and the stage of lactation (n = 20) was 6 ± 2 months. For hourly breast expression sessions, the average number of breast expressions per session was 6 ± 1 for both breasts, and the overall average volume per expression was 22.7 ± 27.7 mL and 23.8 ± 19.8 mL for the left (n = 216) and right (n = 178) breast, respectively. In the 24-hour milk production, the average frequency of breastfeeding sessions was 6.9 ± 1.3 and 6.7 ± 1.1 per 24 hours, and the average of milk volume produced was 466 ± 162 mL/24 hours and 526 ± 151 mL/24 hours for the left (n = 11) and right (n = 9) breast, respectively.

The volumes of milk expressed by the mothers (n = 20) each hour for the hourly pumping sessions are shown in Figure 1. Significantly higher volumes of milk were pumped at the first (0-hour) and second (1-hour) expression (p < 0.05), but there were no significant differences between the third (2-hour) and subsequent breast expressions (Fig. 1). The results of the mother (M4) who participated in three different hourly pumping sessions showed a similar pattern of changes in milk volume at each hourly expression (Fig. 2). There was no significant difference between the mean volume of milk expressed from the third (2-hour) and subsequent expressions for the three different hourly pumping sessions.

FIG. 1.

Examples of volume of milk expressed from the right and left breast at hourly intervals for three mothers (M13, M14, and M15). All sessions began in the morning.

FIG. 2.

Volume of milk expressed from right and left breasts of one mother (M4) at three different pumping sessions. Sessions 1, 2, and 3 were approximately 1 month apart. The first two sessions began in the morning, and the third began in the evening.

The mean (±SD) hourly rates for the 24-hour milk production were 18.4 ± 6.3 mL/hour (n = 11) and 15.5 ± 5.1 mL/hour (n = 9) for the right and left breast, respectively (Fig. 3). The mean (±SD) hourly milk productions from 2 to 7 hours for the mothers who measured their 24-hour milk production were 19.4 ± 6.8 mL/hour (n = 11) and 21.9 ± 6.3 mL/hour (n = 9) for the right and left breast, respectively. The overall mean (±SD) values were 18.0 ± 3.1 mL/hour (n = 47) and 14.0 ± 4.8 mL/hour (n = 38) for the right and left breast, respectively. The Mixed model analysis showed that the hourly volume of milk was not significantly different between session and breasts. Furthermore, there were no significant differences (p > 0.05) between the hourly rates of milk production from the 24-hour breastfeeding for each breast and the average hourly volume recorded between 2 to 7 hours for the hourly pumping study (Fig. 3).

FIG. 3.

Box plot of milk volume expressed from right and left breasts at hourly intervals for all 14 mothers together with the hourly rate of milk production calculated from the 24-hour breastfeed measurements. The time 0 is the volume obtained at the first expression session (the interval between the previous breastfeed/breast expression and time 0 was not standardized). **p < 0.001 for difference between 0 hour and all other times; *p < 0.05 for difference between 1 hour and all other times.

Discussion

The higher volumes of milk obtained at the first expression (Fig. 3) were expected because the time of the mother's previous breastfeed/breast expression was not standardized and was more than 1 hour before the start of the session (Figs. 1 and 2). A longer interval before the 0-hour expression would result in a higher degree of fullness of the breast, and more milk would be available than after the intervals of exactly 1 hour since the last expression. It has been reported that approximately 70% of the available milk was removed either during a breastfeed or breast expression.³ Furthermore, the finding that the volume of milk obtained at the second expression after 1 hour was significantly higher than third and subsequent expressions (Fig. 3) suggests that there was a carryover of this effect. The consistent volumes of milk expressed during the second and subsequent hours (Figs. 1 –3) suggest that with only 1-hour intervals between expressions the breasts were drained to a consistent degree after the second and subsequent hours. The physiological release of oxytocin during expression⁶ was sufficient to enable the consistent removal of milk in women who expressed milk at hourly intervals. This conclusion is consistent with the findings that breastfeeding women experience similar patterns of multiple milk ejections during breastfeeding and breast expression.^6,7 Thus, the endogenous release of oxytocin appears to be equally efficient in facilitating milk removal during both breastfeeding and breast expression in women. Presumably this is because, unlike other suckling mammals, both breast expression and alterations to the breast expression routine are not stressful procedures for lactating women.

The consistent volume of milk obtained from breast expression from the 2^nd to 7^th hour suggests that milk was synthesized at a constant rate over this period (Fig. 3). Because the mean volume of milk expressed per hour from 2 to 7 hours was only 33% of the volume of milk expressed at zero-time, we conclude that the accumulation of relatively small volumes of newly synthesized milk at each hour was not sufficient to invoke any local inhibition of milk synthesis.⁸

Studies of mothers who were either breastfeeding or expressing their breastmilk have demonstrated that over both long-term (weeks) and short-term (hours) periods, the degree to which the breast was drained of milk influenced the subsequent rate of milk synthesis.

For example, Dewey and Lonnerdal⁹ observed that when mothers expressed their breastmilk at the end of each breastfeed for 2 weeks, milk production increased by an average of 124 mL/24 hours above their infants' breastmilk intake. Furthermore, Daly et al.⁸ found that the short-term rate of milk synthesis was significantly associated with the degree of fullness of individual breasts. Thus, it has been concluded that milk synthesis in mothers is a continuous process¹⁰ and is, in part, controlled by a mechanism that is sensitive to the degree of fullness of the breast.

In this connection, it is of interest that the average volumes of milk removed from the 2^nd to 7^th hour were not significantly different and furthermore that the mean volumes of milk expressed from the 2^nd to 7^th hour on three different days for the same mother (M4) were not significantly different (Fig. 2). These findings suggested that, on the one hand, it takes more than 7 hours to obtain a response in milk synthesis to increased milk removal and, on the other hand, that hourly milk removal prevented the down-regulation of milk secretion by local inhibitory factors. However, further studies are required to determine the reproducibility of the amount of milk removal during hourly breast expression within and between days.

The average volumes of milk removed from the 2^nd to 7^th hour were not significantly different from the average hourly rate of milk production calculated from the 24-hour measurement of milk production using the test weighing procedure¹¹ in the mothers' homes (Fig. 3). Thus, it is apparent that the hourly rate of milk production from the 2^nd to the 7^th hour could represent the longer-term physiological rate of milk synthesis in lactating mothers. However, further studies are required to confirm this finding.

Conclusions

One of the major concerns of breastfeeding mothers is the perception of insufficient milk supply,^12–15 and currently the lactating breasts are the only major organs in the human body without clinical tests to determine their normal function. Our findings suggest that the volume of milk removed at the second hour of an hourly expression regime may be a relatively rapid estimate of the functional capacity of the lactating breast. However, further studies are required to fully validate this method.

Footnotes

Acknowledgments

This study was funded by Medela AG, Switzerland. The authors would like to thank the participating mothers and babies, without whom this research would not have been possible. We thank Anna Hepworth for statistical advice.

Disclosure Statement

No competing financial interests exist for all authors. Postgraduate scholarship for C.T. Lai provided by Medela AG, Switzerland. Research grant provided by Medela AG, Switzerland.

References

Lollivier

, Guinard-Flament

, Ollivier-Bousquet

et al. Oxytocin and milk removal: Two important sources of variation in milk production and milk quality during and between milkings. Reprod Nutr Dev, 2002; 42:173–186.

Linzell

. The effect of very frequent milking and of oxytocin on the yield and composition of milk in fed and fasted goats. J Physiol, 1967; 190:333–346.

Kent

, Mitoulas

, Cregan

et al. Importance of vacuum for breastmilk expression. Breastfeed Med, 2008; 3:11–19.

Team RDC. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing: Vienna, 2009.

Pinheiro

, Bates

, DebRoy

et al. nlme: Linear and Nonlinear Mixed Effects Models, 3.1-90 2008. http://cran.r-project.org/web/packages/nlme/index.html.

Ramsay

, Mitoulas

, Kent

et al. Milk flow rates can be used to identify and investigate milk ejection in women expressing breast milk using an electric breast pump. Breastfeed Med, 2006; 1:14–23.

Ramsay

, Kent

, Owens

et al. Ultrasound imaging of milk ejection in the breast of lactating women. Pediatrics, 2004; 113:361–367.

Daly

, Kent

, Owens

et al. Frequency and degree of milk removal and the short-term control of human milk synthesis. Exp Physiol, 1996; 81:861–875.

Dewey

, Lonnerdal

. Infant self-regulation of breast milk intake. Acta Paediatr Scand, 1986; 75:893–898.

10.

Prime

, Geddes

, Hartmann

. Oxytocin: Milk ejection and maternal-infant well-being. Textbook of Human Lactation. Hale

, Hartmann

. Hale Publishing, L.P.: Amarillo, TX, 2007; 114–153.

11.

Arthur

, Hartmann

, Smith

. Measurement of the milk intake of breast-fed infants. JPEN J Pediatr Gastroenterol Nutr, 1987; 6:758–763.

12.

Amir

, Cwikel

. Why do women stop breastfeeding? A closer look at ‘not enough milk’ among Israeli women in the Negev Region. Breastfeed Rev, 2005; 13:7–13.

13.

Hill

, Humenick

. Insufficient milk supply. Image J Nurs Sch, 1989; 21:145–148.

14.

Otsuka

, Dennis

, Tatsuoka

et al. The relationship between breastfeeding self-efficacy and perceived insufficient milk among Japanese mothers. J Obstet Gynecol Neonatal Nurs, 2008; 37:546–555.

15.

Sacco

, Caulfield

, Gittelsohn

et al. The conceptualization of perceived insufficient milk among Mexican mothers. J Hum Lact, 2006; 22:277–286.