Although there is growing data on the acute and chronic health benefits of cow milk, albeit not yet conclusive, whether milk from alternative (non-bovine) sources could provide comparable or superior cardiometabolic protection has not yet been comprehensively reviewed. This narrative review outlines the marked differences in macronutrient composition, particularly protein and lipid content, and discusses how whole milk product (and individual milk ingredients) from different species could impact cardiometabolic health.
The consumption of cow dairy products is a dominant feature in the diet of many cultures globally, particularly among Western communities. There is some evidence from epidemiological studies and systematic reviews alike that dairy intake is inversely linked with the risk of developing metabolic syndrome [1–3]. More pertinently, a body of data supports a negative association between milk intake and the risk of developing dysglycaemia, dyslipidaemia, and hypertension [1,4]. However, with gold-standard data from long-term randomised controlled trials (RCTs) featuring type II diabetes (T2D) and cardiovascular disease (CVD) incidence as primary endpoints not currently available, the causality of these findings remains to be confirmed [5]. Nonetheless, putative explanations for a possible metabolic syndrome risk reduction include a direct modulation of the glycaemic response [2,6], and an indirect modulation of body weight through upregulation of postprandial thermogenesis [6–8] and/or suppression of appetite [9–11]. Features of, or responses to, milk that might contribute to any cardiometabolic protection include the bioactive peptide content [12,13]; fatty acid (FA) content [14], e.g., conjugated linoleic acid (CLA) [15]; glycaemic index (GI) [16,17]; promotion of satiety [18]; mineral content, particularly calcium, magnesium, and potassium [19–22]; and folate bioavailability [23].
Although there is growing data on the acute and chronic health benefits of cow milk, albeit not yet conclusive, whether milk from alternative (non-bovine) sources could provide comparable or superior cardiometabolic protection has not yet been comprehensively reviewed.
The worldwide commercial production of cow milk decisively eclipses the relatively minor contributions from alternative animal species (Table 1). Nonetheless, these milks remain valuable primary sources for many countries and communities globally.
Table 1. Mean contribution of individual species’ milks towards global production [24].
Milk Origin |
Global Milk Production (%) |
Global Milk Production (kg) |
Cow |
81.3 |
714,400,000,000 |
Buffalo |
14.8 |
130,300,000,000 |
Goat |
2.2 |
18,900,000,000 |
Sheep |
1.3 |
11,800,000,000 |
Camel |
0.4 |
3,200,000,000 |
Values rounded to nearest 0.1 percent or 109 kg.
Owing to the specific make-up of proteins (e.g., β-lactoglobulin; β-lg) and sugars (e.g., lactose) within cow milk, the global prevalence of cow milk allergy and intolerance is notably high. Approximately 65% of adults worldwide have a suboptimal capacity to digest and absorb lactose [25]. In Asian and American Indian populations, the reported prevalence of lactose intolerance is closer to 100% [26,27]. However, with marked compositional differences, hypoallergenicity and improved tolerability have been indicated following the ingestion of goat [28], sheep [29], camel [30], buffalo [31], and donkey [32] milk, as compared to cow milk. It should be noted that throughout this review buffalo milk refers to the produce of animals of the Bubalus genus.
Lastly, non-dairy substitutes for milk, including soy, oat, rice, and nut ‘milk beverages’ have received growing attention. These plant-based alternatives are formulated through the disintegration of plant material, extraction in water, and subsequent homogenisation, which produces a ‘milk’ reminiscent of the consistency and appearance of animal milk [33]. Despite a typically substandard macronutrient profile relative to mammalian milk, plant-based ‘milks’ possess distinct functional ingredients, lower allergenicity and greater affordability, which have impelled a noticeable surge in demand and production.
With knowledge of varying nutritional profiles, beyond any implications of a reduced allergenic potential of non-bovine milk, the ingestion of milk from different species could also engender distinct health benefits. Marked differences have been documented in macronutrient composition between multiple milk sources, particularly in terms of protein and lipid content. For instance, 100 g of sheep milk provides a markedly greater amount of protein (P: 5.5 g) and fat (F: 5.9 g) compared to cow (P: 3.4 g; F: 3.3 g), goat (P: 3.7 g; F: 3.8 g), and camel (P: 3.3 g; F: 4.0 g) milk [22,29]. Buffalo and reindeer milks also have a notably high lipid content (7.4 g/100 g and 16.1 g/100 g, respectively) [34]. In addition, mean lactose content varies modestly across ruminant milks at 4.51%, 4.75%, 4.79%, and 4.82% for 100 g of goat, sheep, buffalo, and cow milk, respectively [35]. See Figure 1 for a comparison of the differing macronutrient profiles of common animal milks, and Table 2 for a more detailed examination of the nutritional composition between different animal milks and plant-derived milk alternatives. It is noted that these cited values, and those in the following sections, should merely serve as typical examples of a milk’s nutritional composition. This caveat is raised with knowledge that milk composition can greatly vary under different conditions (i.e., protein composition is largely determined by genetics, and thus varies with herds; whilst lipid content is largely determined by environment, and thus varies with forage and season). Multiple studies with similar designs yet discordant findings have epitomised this variability of milk composition with animal breed, age, health status, diet, and lactation stage, or even milking yield/time of day [36,37]. For instance, the proportion of lipid content ascribed to either unsaturated or saturated FAs in mare (i.e., equine) milk can vary considerably across breed and lactation stage (unsaturated: 39–62%; saturated: 38–61%) [36]. Environmental pollutants may also alter the properties of milk fat [38]. Hence, great caution should be taken when drawing conclusions from a single study’s findings, especially when this is in the form of low-level evidence from analytical studies such as those cited above.
Figure 1. The composition of different species’ milk by fat, protein, and lactose content per 100 mL [22,29,34,35]. Equine milk values represent the mean nutrient content in mare and donkey milks.
Besides gross protein, fat, and carbohydrate quantity, macronutrient quality can also greatly differ between milks of different origin, further altering implications for human health. Relative to monogastric mammals such as humans and mares, casein forms a far greater portion of the protein content in ruminant milk [22]. Moreover, comparisons across ruminant milks have documented that the casein fraction of sheep, goat, buffalo, and yak milk largely comprises β-casein, whilst αS1-casein predominates in cow milk [22,39]. Alongside αS1-casein and β-casein, αS2-casein and κ-casein complete the group of different casein phosphoproteins in mammalian milk. These four phosphoproteins are defined by their distinct primary amino acid (AA) sequence, micellar position, and subsequently, function (e.g., calcium and phosphate transportation, stability and solubility) [40]. Although whey, a by-product of the cheese-making process, is also a family of (five) heterogeneous and polymorphic protein fractions, the composition of whey is more consistent across species, with a common preponderance of β-lg, except in camel milk—in which serum albumin dominates, and β-lg is largely absent [22].
Table 2. Nutritional composition per 100 mL milk of different animal-derived milk and plant-based milk alternatives.
|
Milk Origin |
|||||||||
Cow |
Buffalo |
Sheep |
Goat |
Equine |
Camel |
Soy |
Oat |
Rice |
Almond |
|
Total fat (%) |
3.3 |
7.4 |
5.9 |
3.8 |
1.1 |
4.0 |
2.0 |
2.2 |
1.0 |
1.1 |
MCT (% of |
10.5 |
7.1 |
21.8 |
23.0 |
15.2 |
1.5 |
n.d. |
n.d. |
n.d. |
0.2 |
CLA (% of |
0.7 |
0.5 |
1.2 |
0.6 |
0.1 |
0.9 |
n/a |
n/a |
n/a |
n/a |
SFA (% of |
68.4 |
70.8 |
65.0–75.0 |
65.0–73.8 |
38.0–61.0 |
66.1 |
14.3 |
18.9 |
12.0 |
22.6 |
MFG diameter (µm) |
3.8 |
8.7 |
3.8 |
3.2 |
2.8 |
3.0 |
n/a |
n/a |
n/a |
n/a |
Total protein (%) |
3.4 |
4.4 |
5.5 |
3.7 |
1.8 |
3.3 |
2.6 |
1.0 |
0.5 |
0.6 |
Casein:whey |
82:18 |
82:18 |
76:24 |
78:22 |
52:48 |
73:27–76:24 |
n/a |
n/a |
n/a |
n/a |
Lactose (%) |
4.8 |
4.8 |
4.8 |
4.5 |
6.9 |
4.3 |
n/a |
n/a |
n/a |
n/a |
Galactose (%) |
4.0 |
3.3 |
0.3 |
0.6 |
<0.1 |
<0.1 |
n/a |
n/a |
n/a |
n/a |
GI (0–100) |
27–37 |
- |
- |
- |
89.3 (donkey) |
- |
31–37 |
69 |
79–92 |
49–64 |
Energy (kJ) |
316.9–373.0 |
345.0 |
593.2 |
301.8 |
184.2–205.1 |
328.3 |
179.9 |
195.8 |
225.9 |
126.8 |
Calcium (mg) |
119.8 |
183.9 |
181.7 |
130.4 |
92.9 |
106.0 |
113.0 |
120.0 |
118.0 |
160.0 |
Potassium (mg) |
145.0 |
101.6 |
120.0 |
181.0 |
50.5 |
156.0 |
122.0 |
162.0 |
27.0 |
67.0 |
Data obtained and collated from a range of supermarket product labels and/or the following sources in the literature [17,22,28,29,34,35,51–58]. Equine milk refers to mare milk unless otherwise specified. n.d., not detected.
The unique classification of AAs and FAs composing individual milks has also sparked interest in the literature. Total and essential AA content are the highest in sheep and reindeer milk, whilst goat, buffalo, and yak milk still possess a higher composition than cow milk [22]. Regarding lipid profile, goat, sheep, and camel milk marginally surpass cow milk concentrations of monounsaturated (MUFA) and polyunsaturated (PUFA) FAs [41]. More notably, goat and sheep milk are considerable sources of short- and medium-chain triglycerides (MCTs), compared to long-chain triglyceride (LCT)-rich cow milk [42,43]. In exemplification of this, three MCTs (caproic acid, caprylic acid, and capric acid) even owe their names to the caprine species (i.e., goats). MCT content (C6:0–C12:0) (as a percentage of total fat content) is the highest in goat (23.0%) and sheep (21.8%) milk, followed by mare (15.2%), cow (10.5%) and human (7.3%) milk [37]. Potentially of further bioactive utility, the content of CLA (a PUFA found in animal milk) and vaccenic acid (a CLA precursor) is also higher in sheep milk (1.2% of total FAs) than cow (0.7%) and goat (0.6%) milk [44]. Finally, the oligosaccharide content of goat milk is approximately five and ten times greater than that of cow and sheep milk, respectively [45].
Beyond differences between mammalian milks, disparities between animal- and plant-sourced milks are more prominent still. The gross protein content of non-dairy milk beverages is typically only half that of cow milk [46]. Nonetheless, soy protein possesses a greater content of branched chain amino acids (BCAAs) than whey protein [47]. However, when using a reference evaluative means for assessing dietary protein quality in humans, the digestible indispensable amino acid score (DIAAS) [48] for milk protein is greater than that of soy protein [49]. Specifically, whilst cow milk protein possesses a DIAAS value ≥118% for all indispensable AAs, soy protein is limited by methionine and cysteine content with a lowest DIAAS of 90.6% despite having a DIAAS value >100% for most individual indispensable AAs [49,50]. Aside from protein content and quality, plant-based milk substitutes, except coconut milk, have lower levels of saturated fatty acids (SFAs) and greater levels of PUFAs and MUFAs than cow milk [46].
The literature has detailed the potential value of, inter alia, protein [59], particularly functional AAs [60]; MUFAs [61]; PUFAs [62]; MCTs [63]; and CLA [15] for improved cardiometabolic health. Hence, given the compositional differences outlined above, a robust assessment of the therapeutic potential of cow milk alternatives is required. The relationship between cow milk and glycaemia, blood pressure, and lipidaemia has been well-researched and reviewed in the literature [4], albeit with variable findings, yet investigations into how alternatively-derived milks influence acute postprandial and chronic fasting metabolism are limited. With a newfound importance emphasising the functional implications of individual foods and beverages for health, a review of non-bovine milk consumption is overdue.
Before progressing to any disparate effects of milk origin on cardiometabolic health, it is imperative to acknowledge that the protein and fat composition of different milks might also be variably digested. For example, the extent of gastric casein coagulation (or curd formation) alters the absorption of AAs, hence a higher ingested protein load may not necessarily translate to a higher delivered protein load. The acidification of αS1-casein in cow milk forms rigid and durable curds which are difficult to digest, whereas the coagula from goat, camel and mare milk are far more assimilable [22,64]. The degradation of β-lg also varies between species’ milk [22]. Goat [65] and sheep [66] β-lg is digested more efficiently than cow β-lg, but not as rapidly as mare β-lg [67]. Regarding lipid digestion, lipid globule size and FA chain length are two factors that may be inversely related to digestibility, thus influencing fat assimila[1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27][28][29][30][31][32][33][34][35][36][37][38][39][40][41][42][43][44][45][46][47][48][49][50][51][52][53][54][55][56][57][58][59][60][61][62][63][64][65][66][67][68][69][70][71][72][73][74][75][76][77][78][79][80][81][82][83][84][85][86][87][88][89][90][91][92][93][94][95][96][97][98][99][100][101][102][103][104][105][106][107][108][109][110][111][112][113][114][115][116][117][118][119][120][121][122][123][124][125][126][127][128][129][130][131][132][133][134][135][136][137][138][139][140][141][142][143][144][145][146][147][148][149][150][151][152][153][154][155][156][157][158][159][160][161][162][163][164][165][166][167][168][169][170][171][172][173][174][175][176][177][178][179][180][181][182][183][184][185][186][187][188][189][190][191][192][193][194][195][196][197][198][199][200][201][202][203]tion. Camel and goat milk are noted for small milk fat globules (MFG) compared to sheep, cow, and, most largely, buffalo milk with a mean diameter of 2.99 µm, 3.20 µm, 3.76 µm, 3.78 µm, and 8.7 µm, respectively [68]. However, the direct clinical consequences of MFG size for human health is still largely unknown, with a lack of long-term RCTs having been conducted [69]. Moreover, significant variation in mean MFG size with breed, herd, days in milk, season, and milking period has been reported [70,71]. Hence, this variability within each individual species’ milk devalues current comparisons of MFG size between different species’ milk. Finally, as the ester bonds of short- and medium-chain FAs are more readily hydrolysed than long-chain FAs, a greater fraction of the former, as found in goat and sheep milk, may contribute to higher digestibility [72].
The effect of milk origin on cardiometabolic health is an emerging area of research. There is some data, although primarily from compositional analyses [35,37], in vitro studies [83], animal studies [80], and acute clinical RCTs [77,78,187], that milk from non-bovine origin (notably sheep and goat milk) could prove to be a viable substitute to cow milk for the maintenance, or even enhancement, of cardiometabolic health. However, a collation of the compositional differences and postulated therapeutic utility, as presented in this review, indicate that the level of evidence required to form nutritional recommendations surrounding milk origin is currently lacking. Nonetheless, there are some interesting results, albeit largely from preliminary studies, that have generated excitement around sheep milk consumption for the possible attenuation of cardiometabolic risk. This interest is largely based upon its favourable profile of lipids (e.g., MCTs, CLA), protein (e.g., leucine), and minerals (e.g., calcium). In theory, these compounds could provide protection from obesity, T2D, and CVD through the modulation of postprandial glycaemia, lipidaemia and aminoacidaemia; nutrient processing; postprandial thermogenesis; and/or appetite. Comparably, with desirable nutritional compositions and some promising early findings, goat and buffalo milk may also prove to be robust alternatives to cow milk. However, as with sheep milk, there is currently a stark absence of high-quality research in humans. Hence, as remains pertinent for cow milk, to substantiate any claims that the consumption of cow-milk alternatives can improve cardiometabolic health, causal data from long-term clinical RCTs, ideally with T2D and/or CVD events as the primary endpoint, are required. Evidence from large-scale studies that support the conjectures formed in this review could not only be of value to individuals allergic or intolerant to cow milk, but potentially also to those at an increased risk of cardiometabolic disease. Thus, this review concludes that further exploration into the therapeutic potential of milk beyond the realms of cow dairy is warranted.
This entry is adapted from the peer-reviewed paper 10.3390/nu14020290