Journal Logo

Basic Sciences: Original Investigations

Does exercise intensity or diet influence lactic acid accumulation in breast milk?


Author Information
Medicine & Science in Sports & Exercise: January 1999 - Volume 31 - Issue 1 - p 105-110
  • Free


Lactic acid (LA) has been previously shown to increase in the breast milk following maximal (30) and moderate intensity (29) exercise. The increase in milk LA following maximal exercise correlates with reduced infant acceptance of this postexercise breast milk (28). This has caused some health-care professionals to recommend altering exercise patterns in an effort to avoid infant rejection of postexercise breast milk. We have recently studied the impact of exercise intensity on LA accumulation in breast milk (5). Our findings demonstrated that a 30-min treadmill exercise bout conducted at 50 or 75% of maximal oxygen uptake (V˙O2max) resulted in nonsignificant differences in milk LA as compared with a 30-min seated control session. This suggests that exercise conducted up to 75% of maximum does not alter milk LA levels, and we recommended that women pursue moderate physical activity if desired. During this study, however, we became aware of two variables that could affect our findings. The first of these was dietary carbohydrate (CHO) intake. Dietary CHO intake can increase blood LA levels (14,24), and although our subjects consumed a calorically adequate diet, the macronutrient content was not controlled.

The second of these was exercise intensity, especially near lactate or "anaerobic" threshold. While performing the 75% intensity exercise, some of our subjects were more physiologically stressed than others, suggesting that some of the women were very close to, or at, their anaerobic threshold; this was verified by ventilatory data.

Therefore, the objective of this study was to determine blood and milk LA concentrations under controlled exercise and diet conditions. Exercise intensity was established at maximal, the LA threshold (LAT), and 20% below LAT (LAT-20). A nonexercise control session was also used. In addition, women were assigned to either a high or a moderate CHO diet group. These diet groups contained 63% or 52% of total kcal as CHO, respectively, representing a practical range of CHO intakes that most postpartum women might consume.


Subjects. Twelve, lactating women (3-7 months postpartum) were recruited to participate in this study (Table 1). Subjects were mild to moderately active (walking, aerobic dance, jogging) two to six times per week. None of the subjects smoked or were on medication known to affect milk quality or quantity. All the women were either exclusively breast-feeding or providing >80% of the feedings via the breast. All clinical indices of health and fitness including blood pressure (BP), body mass index, V˙O2max, blood lipids, and % body fat were within normal ranges. Subjects were randomly assigned to either a high-CHO or a moderate-CHO (mod-CHO) diet group, with one exception. One subject's dietary assessment performed at the first meeting revealed that she was accustomed to consuming a diet high in CHO, and further discussion revealed that she would have difficulty increasing the fat content of her diet. This subject was placed in the high-CHO group.

Subject characteristics for each dietary group (values are mean ± SD).

Procedures. Subjects reported to the UNH Exercise Physiology Laboratory on five separate occasions. At the first session, subjects read and signed an informed consent form that had been approved by the University of New Hampshire Institutional Review Board for the Protection of Human Subjects. They were also familiarized with the nutrition regimen and the testing environment, including equipment, facilities, and personnel involved in the study. The nutrition regimen, Healthy Eating, is modeled after the Diabetic Exchange List and is an effective tool for monitoring and guiding food choices (4). Each page of the guidebook lists foods and their serving size that belong to one of six food categories: fruit, vegetable, fat, milk, meat, and high CHO. The number of servings from each of the six food categories constituted each subject's six "daily goals." These goals were a function of the daily caloric requirements of each subject (estimated by summing resting metabolic needs using the Harris-Benedict equation (11), normal and exercise energy needs (16), and the average energy cost of lactation (500 kcal·d−1) (21)) and the diet group: daily goals for the high-CHO group had a macronutrient composition of 62% CHO, 17% protein, and 20% fat; goals for the mod-CHO group were 45% CHO, 17% protein, and 37% fat.

Subjects were counseled for 45 min on using the guidebook, measuring and recording food choices to the nearest 1/4 of a serving using food models, and targeting their daily goals. Each subject was contacted by phone 2 d later to review their diet compliance and to answer questions. Subjects followed the nutrition regimen for an average of 4 (range = 3-6) days before the study began. In addition, each subject was weighed on all remaining visits to ensure that caloric needs were being met. On each of the remaining four visits, the subject's guidebook was reviewed and, if needed, suggestions were made for improving compliance (for example, if a subject tended to chronically underconsume their daily goals from the fruit category, their daily goal from the high-CHO category was increased to insure that calorie intake and macronutrient profile remained stable). Between visits, subjects were contacted by phone to monitor progress and to answer questions.

The second session was to assess body composition via skin-fold analysis and to perform a treadmill test to determine LAT and V˙O2max. Skin-fold sites included the tricep, subscapula, suprailiac, abdominal, and thigh. Percent body fat was determined by the equation developed by Jackson et al. (13). A discontinuous, modified Astrand (2) protocol was used, and blood LA, heart rate (HR), and rhythm, as well as expired respiratory gases, were monitored and measured. Each woman self-selected a comfortable speed and started at a 0% incline, and then the grade was increased by 2.5% every 3 min. At the end of each 3-min workload, the subject straddled the moving treadmill belt, and a blood sample was taken from the finger-tip. Lactic acid was determined (duplicate) from the blood samples, and the mean was graphed and subsequently analyzed, in blind fashion, by three investigators. The LAT was determined as the first breakpoint from linearity. Two of the three investigators had to agree on the breakpoint for acceptance. This method of determining LAT has been used previously (23,26) and has been shown to be not significantly different from the log-log transformation model (26). A metabolic measurement cart was used to assess V˙O2. Analyzers were calibrated before each test. Milk samples were collected via a dual-chamber electric breast pump at the following intervals: 30 min and immediately preexercise, immediately postexercise, and 30, 60, and 90 min postexercise. Both breasts were completely emptied of milk at each collection interval. Blood samples were also taken, via fingerstick, just before milk expression immediately preexercise and immediately, 30, 60, and 90 min postexercise. Care was taken to avoid sweat contamination in both the blood and milk samples.

Following the maximal test, subjects reported to the laboratory three additional times to perform two, 30-min exercise bouts (conducted at the LAT and LAT-20) as well as a control session. These three sessions were conducted in random order and no sooner than 3 days apart. During the control session, subjects sat quietly for 30 min. During the exercise sessions, oxygen uptake was periodically monitored to ensure correct intensity and treadmill speed was adjusted if necessary. In addition, perceived exertion was monitored using Borg's 6-20 rating of perceived exertion (RPE) scale (3). Blood and milk samples were collected at the same intervals described for the maximal test. After each exercise bout, subjects were encouraged to drink water to replace fluids lost due to exercise and milk expression.

Sample analyses. Lactic acid in whole milk and blood was measured immediately after sample collection with a portable LA analyzer that measures hydrogen peroxide production via a platinum electrode (Yellow Springs Incorporated, Yellow Springs, OH). Milk LA values were confirmed in six samples using an independent, enzyme-coupled assay that measures NADH production via spectrophotometer (12). Immediately after expression, milk volume was measured, and aliquots were removed for LA analysis and pH measurement on a Corning pH meter (model no. 345, Corning Inc., Corning, NY), which automatically compensated for temperature variation. The remaining milk was returned to the mother for infant feeding.

Statistical analyses. Differences between main effects of exercise and diet treatments were analyzed using an ANOVA with repeated-measures design, followed by a Tukey post hoc analysis, if necessary. Dependent variables included milk and blood LA, milk pH, and milk volume. Significance was set at P < 0.05.


The dietary macronutrient composition and energy content for subjects in both diet groups is shown in Table 2. Subjects in both groups did not differ in their daily calorie intake (1952 kcal and 2095 kcal for the high-CHO and mod-CHO groups, respectively), and all subjects were weight stable throughout the study. The macronutrient composition of the diets in the two groups differed significantly. Subjects in the high-CHO group consumed 11% more of their calories as CHO and 10% fewer calories as fat, compared with the mod-CHO group (P < 0.05). Normalizing CHO intake for body mass demonstrates that the high-CHO group consumed 5.03 g CHO·kg BM−1, which was nearly one-third greater (P < 0.05) than that consumed by the mod-CHO group (3.89 g CHO·kg BM−1). The g of CHO consumed by each group was also significantly different (305 g vs 269 g, P < 0.05).

Dietary characteristics per group (values are mean ± SD).

The LAT and LAT-20 exercise bouts were monitored with our metabolic gas analysis system, and all subjects maintained the desired intensity for both 30-min sessions. The LAT sessions averaged 71.1% of V˙O2max (range = 60.2-80.3%), whereas the LAT-20 sessions averaged 57% of V˙O2max (range = 48.2-64.2%). The women reported the LAT exercise bout to be significantly more strenuous than the LAT-20 session (RPE = 15 ± 1 for LAT vs 12 ± 2 for LAT-20, P < 0.05).

Milk and blood lactic acid. In both the high-CHO and mod-CHO groups, milk LA was significantly elevated at 0 and 30 min after maximal exercise compared with the same time points after their respective control sessions (Fig. 1). In addition, the LAT exercise bout in both groups resulted in significantly elevated milk LA immediately postexercise. This elevation returned to baseline within 30 min. No elevations were observed following the LAT-20 exercise bout in either group. There was no significant difference in milk LA between the high-CHO and mod-CHO groups at the same time point and exercise intensity.

Figure 1
Figure 1:
Milk lactic acid content before and after exercise of different intensities. Values are means ± SD (N = 6 per group). Pre-ex values represent the average of the 30-min pre and 0-pre samples. * Significantly different from control value at same time point, P < 0.05. These include: high-CHO, V˙O2max 0-post; mod-CHO, V˙O2max 0-post; high-CHO, LAT 0-post; mod-CHO, LAT 0-post; high-CHO, V˙O2max 30-post; mod-CHO, V˙O2max 30-post.

The blood LA responses were similar but longer-lasting than those observed in the milk. Both the max and LAT exercise bouts resulted in significant elevations in blood LA immediately postexercise. These blood LA values remained elevated through 60 min for max exercise while levels returned to normal by 60 min for LAT (Fig. 2). In addition, there was a slight, but significant, rise in blood LA during the LAT-20 intensity immediately postexercise, which returned to normal within 30 min. There was no significant difference in blood LA between the high-CHO and mod-CHO groups at the same time point and exercise intensity.

Figure 2-B
Figure 2-B:
lood lactic acid content before and after exercise of different intensities. Values are means ± SD (N = 6 per group). * Significantly different from control value at same time point, P < 0.05. These include: high-CHO, V˙O2max 0-post; mod-CHO, V˙O2max 0-post; high-CHO, LAT 0-post; mod-CHO, LAT 0-post; high-CHO, LAT-20 0-post; mod-CHO, LAT-20 0-post; high-CHO, V˙O2max 30-post; mod-CHO, V˙O2max 30-post; high-CHO, LAT 30-post; mod-CHO, LAT 30-post; high-CHO, V˙O2max 60-post; mod-CHO, V˙O2max 60-post.

Milk pH and volume. Milk pH tended to increase during the postexercise period: This did not differ between diet groups, so the data were pooled (Table 3). There were no significant differences in milk pH at any of the time points before and after LAT-20, LAT, and maximal exercise intensity compared with the same time points of the control session (with one exception immediately after exercising at V˙O2max). Milk volume tended to drop during the recovery period (pooled data), but no significant differences between time points before and after exercise versus control (except two data points during the LAT session) were noted (Table 4).

Milk pH data as expressed per exercise intensity (mean ± SD;N = 12).
Milk volume (mL) data as expressed per exercise intensity (mean ± SD;N = 12).


This study refines our previous work (5) by examining the influence of nutrition and relative exercise intensity on breast milk LA levels. It also contributes practical information for developing exercise guidelines for breastfeeding mothers. Our results demonstrate that when women are trained in the use of the 6-20 RPE scale, exercise intensity can be accurately scaled, and when the RPE ≤ 12, no LA accumulates in breast milk. This agrees with Dewey et al. (8) and Lovelady et al. (15), who have also reported no changes in breast milk composition or infant acceptance of milk when women exercise to moderate levels of exertion. When women exercised to the "hard" level on this scale, significant but apparently short-term elevations in LA occurred in the milk. We would recommend that women who desire to exercise during the early postpartum (3-4 months) period become familiar with this scale and maintain intensity below the level of "hard." Health care professionals could also distribute and encourage the use of this activity scale.

This study provided a range of CHO intake that women might typically ingest to determine whether practical changes in diet affect milk LA. The CHO composition of the diet did not influence blood or milk LA. One explanation for this is that the dietary CHO content may not have been high enough to cause blood changes in LA. In a study by Rennie and Johnson (25), male subjects consumed a high-CHO diet containing 520 g of CHO over a 3-d period. This was compared with their normal diet, which contained approximately 47% of the energy from CHO. Significant elevations in blood LA during and following a strenuous 90-min run were seen following this high-CHO diet, but not after the subject's normal diet. In another study of male subjects by Prusaczyk et al. (23), 3 d of consuming 744 or 416 g CHO significantly elevated plasma LA during running compared to consuming 168 g CHO. These wide ranges in CHO intake are only achieved by radically altering and supplementing the diet. Moreover, due to their smaller size, it is unlikely that women would consume such large amounts of CHO under usual living conditions. Our results suggest that dietary alterations in CHO intake that women would practically encounter do not affect milk LA in lactating women.

Although not statistically significant, there was a tendency for milk and blood LA to be slightly higher in the mod-CHO group compared with the high-CHO group, at many of the collection points. This may have a biological basis: adipose tissue releases LA (7), and plasma LA levels are higher in obese subjects than in healthy controls (6). In the present study, the mod-CHO group was not obese but did have a higher % body fat (29.2 ± 5.1%) than the high-CHO group (21.9 ± 6.2%) (P = 0.05). Whether the slight elevations in milk and blood LA in the mod-CHO group originate from their higher adipose mass remains to be determined.

Subjects in this study consumed an average of 2024 kcal·d−1. This is approximately 275 kcal under our targeted value but is above the minimum level that lactating women are encouraged to consume per day (10). Dusdieker et al. (9) showed that consuming 1765 kcal·d−1 did not affect quality or quantity of milk or infant growth.

The milk LA values from the present study, verified by an independent enzymatic assay, are similar to values in our prior study (5). In the present study, subjects had peak milk LA values of 1.35 mM immediately after maximal exercise, which was similar to our previous study (0.98 mM). However, this value is one-half of the mean value of 2.88 mM following maximal exercise previously reported (29). Similarly, our milk LA values immediately after the LAT-20 intensity averaged 0.11 mM, whereas a milk LA value of 1.06 mM following a "typical workout" (calculated as 55% intensity) in postpartum women has been reported (29). This "typical workout" value is similar to the LA value after the V˙O2max exercise in the present study. It is possible that the "typical workout" intensity was higher than originally calculated (5), since the authors used a modification of the Karvonen formula, which requires accurate HR monitoring during exercise. Because it was not stated if and how the HR was monitored during activity (i.e., was HR recorded during the actual exercise or immediately after), exercise intensity may have been underestimated, and the women may have been exerting at a higher level than anticipated. No RPE data were collected so comparisons with the present study are difficult. Methodological differences may also be responsible for the discrepancy in LA values. In our studies, both breasts were completely emptied of milk at each collection time, whereas previously published work reports that 1 mL (29) to 3 mL (28,30) were collected; milk composition can vary greatly within a feeding (22). Finally, we analyzed whole milk in duplicate immediately following collection in both of our studies, whereas in previous studies deproteinized milk was assayed after having been frozen.

Milk pH remained fairly stable but had a tendency to rise slightly across the measurement time periods, suggesting that the mammary gland has an inherent buffering capacity even as blood and milk LA levels rise. These data are similar to our previous work in which milk pH was inversely correlated with milk volume (5). This slight increase in milk pH may be due to the fact that expressed milk gradually loses carbon dioxide (1) and the smaller the milk sample, the greater the surface area relative to volume. Therefore, there is a greater opportunity for carbon dioxide to be released and for pH to rise as sample volume decreases.

A variety of factors are known to influence breast milk taste and, potentially, infant acceptance of that milk. Alcohol consumption has an immediate effect on the odor of milk, causing a reduction in infant feeding (19). Maternal smoking raises breast milk nicotine levels and causes physiological effects (changes in respiratory frequency and oxygen saturation) in breast-feeding infants (27). In contrast, maternal ingestion of garlic causes infants to consume more milk and attach to the breast for longer periods of time (18), and a variety of flavors find their way into the breast milk including mint and vanilla (17). Mental and physical stress can also influence taste perception (20). Thus, it is possible that factors other than LA were responsible for reduced infant acceptance of postexercise breast milk previously reported (28). Precisely what factors influenced infant acceptance of postexercise milk (28) remains to be determined.

In conclusion, these data support the notion that altering dietary CHO intake from 3.9 to 5.0 CHO·kg BM−1 will not increase milk LA levels before or after exercise, regardless of exercise intensity. It also supports the hypothesis that moderate exercise performed by lactating women will not increase the milk LA content (5). We recommend that postpartum women who desire to exercise do so at a moderate intensity level using a perceptual scale such as Borg's 6-20 RPE scale. Exercising above this moderate level may affect LA content in the breast milk; however, the effect this would have on the nursing infant is unknown. Further work should address this issue.


1. Allen, J. C., and M. C. Neville. Ionized calcium in human milk determined with a calcium sensitive electrode. Clin. Chem. 29: 858-861, 1983.
2. Astrand, I. Aerobic work capacity in men and women with special reference to age. Acta Physiol. Scand. 49(Suppl. 169):1-92, 1960.
3. Borg, G. Perceived exertion as an indicator of somatic stress. Scand. J. Rehabil. Med. 2:92-98, 1970.
4. Carey, G. B. A nutrition guidebook to improve the diet of male and female collegiate athletes. J. Swimming Res. 8:24-29, 1992.
5. Carey, G. B., T. J. Quinn, and S. E. Goodwin. Breast milk composition after exercise of different intensities. J. Hum. Lact. 13:115-120, 1997.
6. Chen, Y. D. I., B. B. Varasteh, and G. M. Reaven. Plasma lactate concentration in obesity and type 2 diabetes. Diabetes Metab. 19:348-354, 1993.
7. Coppack, S. W., K. N. Frayn, S. M. Humphreys, P. L. White, and T. D. R. Hockaday. Arteriovenous differences across human adipose and forearm tissues after overnight fast. Metabolism 39:384-390, 1990.
8. Dewey, K. G., C. A. Lovelady, L. A. Nommsen-Rivers, M. A. McCrory, and B. Lonnerdal. A randomized study of the effects of aerobic exercise by lactating women on breast-milk volume and composition. N. Engl. J. Med. 330:449-453, 1994.
9. Dusdieker L. B., D. L. Hemingway, and P. J. Stumbo. Is milk production impaired by dieting during lactation? Am. J. Clin. Nutr. 59:833-840, 1994.
10. Hamosh, M. Nutrition during lactation. In: Nutrition in Pregnancy and Growth, M. Porrini and P. Walter (Eds.). Basel: Karger, 1996, pp. 23-36.
11. Harris, J., and F. Benedict. Publication #279. In: A Biometric Study of Basal Metabolism in Man. Washington, DC: Carnegie Institution, 1919, pp. 1-266.
12. Hohorst, H. J. Lactate dehydrogenase. In: Methods in Enzymatic Analysis, H. Bergmeyer (Ed.). New York: Academic Press, 1963, pg. 266.
13. Jackson, A. S., M. L. Pollock, and A. Ward. Generalized equations for predicting body density in women. Med. Sci. Sports Exerc. 12:175-182, 1980.
14. Kelman, G. R., R. J. Maughan, and C. Williams. The effect of dietary modifications on blood lactate during exercise. J. Physiol. 251:34-35, 1975.
15. Lovelady, C. A., B. Lonnerdal, and K. G. Dewey. Lactation performance of exercising women. Am. J. Clin. Nutrition. 52:103-109, 1990.
16. McArdle, W. D., F. I. Katch, and V. L. Katch. Exercise Physiology: Energy, Nutrition, and Human Performance. Baltimore: Williams & Wilkins, 1996, pp. 769-781.
17. Menella, J. A. Mother's milk: a medium for early flavor experiences. J. Hum Lact. 11:39-45, 1995.
18. Mennella, J. A., and G. K. Beauchamp. Maternal diet alters the sensory qualities of human milk and the nursling's behavior. Pediatrics 88:737-744, 1991.
19. Mennella, J. A., and G. K. Beauchamp. The transfer of alcohol to human milk: effects on flavor and the infant's behavior. N. Engl. J. Med. 325:981-985, 1991.
20. Nakagawa, M., K. Mizuma, and T. Inui. Changes in taste perception following mental or physical stress. Chem. Senses 21:195-200, 1996.
21. National Research Council, Food, and Nutrition Board. Recommended Dietary Allowances. Washington, DC: National Academy Press, 1989, pg. 35.
22. Neville, M. C., R. P. Keller, J. Seacat, C. E. Casey, J. C. Allen, and P. Archer. Studies on human lactation. I. Within-feed and between-breast variation in selected components of human milk. Am. J. Clin. Nutr. 40:635-646, 1984.
23. Prusaczyk, W. K., K. J. Cureton, R. E. Graham, and C. A. Ray. Differential effects of dietary carbohydrate on RPE at the lactate and ventilatory thresholds. Med. Sci. Sports Exerc. 24:568-575, 1992.
24. Quirion, A., G. R. Brisson, L. Laurencelle, et al. Lactate threshold and onset of blood lactate accumulation during incremental exercise after dietary modifications. Eur. J. Appl. Physiol. 57:192-197, 1988.
25. Rennie, M. J., and R. H. Johnson. Effects of an exercise-diet program on metabolic changes with exercise in runners. J. Appl. Physiol. 37:821-825, 1974.
26. Smith E. W., M. S. Skelton, D. E. Kremer, D. D. Pascoe, and L. B. Gladden. Lactate distribution in the blood during progressive exercise. Med. Sci. Sports Exerc. 29:654-660, 1997.
27. Stepans, M. B., and N. Wilkerson. Physiologic effects of maternal smoking on breast-feeding infants. J. Am. Acad. Nurse Pract. 5:105-113, 1993.
28. Wallace, J. P., G. Inbar, and K. Ernsthausen. Infant acceptance of postexercise breast milk. Pediatrics 89:1245-1247, 1992.
29. Wallace, J. P., G. Inbar, and K. Ernsthausen. Lactate concentrations in breast milk following maximal exercise and a typical workout. J. Women's Health 3:91-96, 1994.
30. Wallace, J. P., and J. Rabin. The concentration of lactic acid in breast milk following maximal exercise. Int. J. Sports Med. 12:328-331, 1991.


© 1999 Lippincott Williams & Wilkins, Inc.