The Glycemic Impact of Protein Ingestion in People With Type 1 Diabetes Free

In individuals with type 1 diabetes, carbohydrates are perceived as the primary macronutrient influencing postprandial glucose levels. However, accumulating evidence indicates that protein ingestion substantially influences postprandial blood glucose levels (1,2), adding a layer of complexity to the dietary management of glycemia in people with type 1 diabetes. Dietary and insulin dosing recommendations to optimize glycemic outcomes in type 1 diabetes usually only account for the carbohydrate content of the meal, while the influence of protein on postprandial glycemic excursions is often overlooked—and is not well understood. While the contribution of protein in the context of mixed meals was previously reviewed for people with type 1 diabetes (1,3), a comprehensive review of the isolated impact of protein or amino acid ingestion is needed. Importantly, protein ingestion provides a unique physiological challenge for people with type 1 diabetes, distinct from that in people without diabetes or with type 2 diabetes. Accordingly, this article provides a physiological understanding to highlight the unique metabolic challenges and therapeutic opportunities that protein ingestion presents in people with type 1 diabetes.

That the endocrine and metabolic responses to carbohydrate and protein differ is well established. In people without diabetes, carbohydrate ingestion alone increases insulin concentrations and inhibits glucagon secretion (4). These responses inhibit endogenous glucose production (EGP) and stimulate tissue glucose uptake, ensuring that circulating glucose returns to postabsorptive levels (4). However, in people without diabetes protein ingestion (without carbohydrate) raises both insulin and glucagon levels (5–7), which synergistically stimulate plasma amino acid clearance via skeletal muscle and the liver, respectively (8,9), while glycemia remains stable as the glucose-lowering effects of insulin are offset by glucagon (7,10). In contrast, protein ingestion in people with type 1 diabetes is characterized by an absent endogenous insulin response and an unopposed and enhanced glucagon excursion, causing a rise in circulating glucose (6,11–13). That ingested protein can potently stimulate glucagon secretion has been consistently demonstrated in people with type 1 diabetes (12,14,15), despite an impaired glucagon response to other stimuli, such as exercise and hypoglycemia (16–18).

The glycemic effect of protein ingestion in people with type 1 diabetes is often misunderstood and is unique from the effects seen in people without diabetes or with type 2 diabetes. Specifically, adding protein to carbohydrate-containing meals often reduces postprandial glucose excursions in people without diabetes or with type 2 diabetes. This is typically due to enhanced insulin response and a slowing of gastric emptying, which in part are mediated by heightened incretin hormone secretion (19). As such, mealtime protein ingestion is generally considered “glucose lowering” in people without type 1 diabetes. In those with type 1 diabetes, however, the absolute deficiency in the endogenous insulin response means that protein ingestion, either alone or with carbohydrates, ultimately elevates glucose levels, likely due to a glucagon-stimulated rise in EGP (Fig. 1).

In comparison with carbohydrates, which are typically digested and absorbed within 1–2 h, protein digestion is slow, with most amino acid absorption occurring within ∼4 h (20). Protein-derived amino acids are absorbed via the small intestine, with a substantial amount extracted by the liver and intestine (21). Due to its unique metabolic capabilities, the liver is the main site for amino acid disposal (22), where amino acids are converted to urea and glutamine, while their carbon skeleton is converted to glucose via hepatic gluconeogenesis (23) (Fig. 2). Hepatic conversion of amino acids to glucose usually occurs in both the postabsorptive and postprandial states and is responsible for a large portion of total EGP (22). Glucagon is the key regulator of hepatic amino acid metabolism, and the net effect on glucose output by the liver depends on the insulin–to–glucagon concentration ratio (24). Despite an impaired glucagon response to exercise and hypoglycemia, several studies have reported that the glucagon response to amino acids is preserved in people with type 1 diabetes (12,14,15,25,26). Thus, due to the absent endogenous insulin response, amino acid–induced hyperglucagonemia is expected to result in a marked stimulation of hepatic gluconeogenesis and glycogenolysis and a subsequent increase in circulating glucose levels.

However, observations regarding protein’s glycemic effect in people with type 1 diabetes are equivocal, with the glycemic response seemingly dependent on the type of protein ingested. Similar to carbohydrates, protein digestion and amino acid absorption rate is a key determinant of the resultant glycemic response (27). Additionally, amino acid content and composition affect the digestive properties of the ingested protein and influence the resultant glucoregulatory hormonal response (28–30), where potent glycemic responses may be seen with some proteins, while others have little effect. Thus, the apparent inconsistencies surrounding the glycemic effects of protein ingestion in type 1 diabetes may be attributed to the misconception that all protein sources have a similar impact on glucose metabolism. Protein composition is, therefore, a key consideration in investigating the glycemic responses to protein ingestion in people with type 1 diabetes.

An additional consideration regarding protein’s glycemic effect in people with type 1 diabetes is the effect of glycemic control on this response. While there are little detailed data on the relationship between glycemic control and protein digestion/absorption, available evidence supports the notion that glycemia can influence gastric emptying. Specifically, it is suggested that long-term hyperglycemia is associated with a delayed gastric emptying rate (31) and that up to half of individuals with type 1 and type 2 diabetes have delayed gastric emptying (32). Acutely, even mild hyperglycemia (∼8 mmol/L) can slow gastric emptying in both people with and people without type 1 diabetes (33)—while hypoglycemia seems to speed up gastric emptying, such that it is positively correlated with the degree of hypoglycemia (34). Accordingly, it is expected that acute and long-term glycemic control could influence the rate of protein digestion and absorption, which in turn can affect the overall glucoregulatory response. Indeed, Raskin et al. (35) demonstrated that the glucagon and glycemic response to protein ingestion is impaired during hyperglycemia compared with euglycemia, although further data are needed to explore this relationship.

Amino acids can stimulate glucagon secretion via direct and indirect mechanisms, and the high sensitivity of the α-cell to amino acids makes it the main sensor and regulator of plasma amino acid levels (36). Amino acids can be transported into the α-cell, where they stimulate glucagon secretion through direct cell depolarization or by fueling the glucagon secretory pathway (37). This has been shown in people with type 1 diabetes, where intravenous infusion of individual glucogenic amino acids such as alanine, arginine, and glutamine, or an amino acid mixture, stimulated glucagon secretion (26,38–40). Additionally, ingested amino acids can indirectly stimulate glucagon secretion via gut-derived hormones (37).

Importantly, more “complete” protein sources (e.g., lean beef or whey protein) tend to elicit a greater response compared with ingestion of a single amino acid (e.g., alanine) alone (6,11,12,15). The magnitude of the glucagon response is also dependent on the absorption rate of ingested protein. For example, in comparing the ingestion of 50 g rapid-absorbing whey protein (a natural byproduct of milk) with 50 g beef protein, whey protein induced a greater and more rapid glucagon response (11,12). Interestingly, the kinetics and magnitude of the amino acid–induced glucagon response in individuals with type 1 diabetes appear to be similar to the response of those with type 2 diabetes and without diabetes (6). Also noteworthy, glucagon stimulation by oral amino acids in people with type 1 diabetes—similar to that in people without diabetes, where protein ingestion stimulates glucagon secretion in the presence of an insulin secretory response—still occurs under experimentally induced hyperinsulinemia, reinforcing the fact that amino acid–induced stimulation of glucagon secretion overrides insulin’s inhibitory actions on α-cell glucagon secretion (14,40).

Interestingly, Rossetti et al. (14) found that while the glucagon response during a hyperinsulinemic-hypoglycemic clamp was completely attenuated in people with type 1 diabetes, oral amino acids restored this response. Moreover, the glucagon response to amino acids during hypoglycemia was greater than that during euglycemia, suggesting a synergistic effect of amino acids with hypoglycemia, where amino acids may “awaken” the α-cells’ counterregulatory ability to respond to hypoglycemia (14). A similar synergistic effect in people with type 1 diabetes was seen by Kristensen et al. (41) in the context of exercise. The plasma glucagon response to a 45-min aerobic moderate-intensity exercise session was absent when a high-carbohydrate meal (52% carbohydrates, 21% protein, 27% fat) was consumed 90 min before moderate-intensity exercise, yet glucagon increased during exercise when exercise was preceded by a high-protein meal (21% carbohydrates, 52% protein, 27% fat). The mechanism by which amino acids may amplify glucagon responses to hypoglycemia or exercise remains unclear and could be an area worthy of future investigation.

The gut-derived incretin hormones, glucose-dependent insulinotropic polypeptide (GIP) and glucagon-like peptide 1 (GLP-1), are secreted when nutrients enter the gastrointestinal tract (42). These hormones act as a sensor to communicate incoming nutrient levels to the pancreatic islets, thus modulating the insulin and glucagon secretory response (37). In people without diabetes, both GIP and GLP-1 potentiate the postmeal insulin response (42). In contrast, the effects of GIP and GLP-1 on postprandial glucagon secretion vary and appear to depend on the prevailing plasma glucose levels. Specifically, GLP-1 inhibits glucagon secretion, whereas GIP enhances glucagon secretion during hypoglycemia and euglycemia, but not hyperglycemia (43,44). Protein ingestion stimulates a robust GIP response (45) and, like glucagon, also depends on protein digestion and amino acid absorption rates. Calbet and Holst (46) demonstrated that ingesting fast-absorbing whey and casein hydrolysate results in ∼20% greater and earlier GIP response than with whole food alone.

Unlike glucagon, individuals with type 1 diabetes have normal basal levels of GLP-1 and GIP, and their incretin response appears intact (47,48). Vilsbøll et al. (48) showed that on ingestion of a set of mixed meals varying in calorie content (48% carbohydrate, 33% fat, and 19% protein), GLP-1 and GIP responses were comparable between individuals with and individuals without type 1 diabetes, regardless of the meal size. Despite the available evidence with mixed meals, data on the incretin effect in response to protein ingestion alone in people with type 1 diabetes are limited. Paramalingam et al. (11) found an increase in plasma GLP-1 and GIP after ingestion of 50 g whey protein in individuals with type 1 diabetes. Interestingly, as in people without diabetes, the stimulation of glucagon secretion occurred despite the presence of elevated GLP-1, suggesting that amino acids can override the inhibitory effects of GLP-1 on glucagon secretion (44). Additionally, the antagonistic actions of GLP-1 on glucagon secretion appear to be glucose dependent, with the inhibitory effect stronger at higher glucose concentrations (49). Considering that the rise in glucose after protein ingestion in people with type 1 diabetes is relatively slow, with blood glucose peaking at ∼2 h after whey protein ingestion, this may provide a favorable setting to promote glucagon secretion (11). Indeed, in this early postprandial period following protein ingestion when glucose levels have yet to rise substantially, the stimulation of glucagon secretion by amino acids would occur independent of the inhibitory actions of GLP-1 or hyperglycemia itself.

Despite the interest in the glycemic response following protein ingestion in people with type 1 diabetes, surprisingly, limited studies have quantified the influence of protein ingestion on EGP. Since protein ingestion stimulates glucagon secretion and directly increases gluconeogenic amino acid availability, EGP stimulation would be expected in people with type 1 diabetes after protein consumption (50). In studies in individuals without diabetes, varying effects of protein ingestion on EGP have been reported, with some studies demonstrating unaltered EGP following protein ingestion; however, this is likely due to the relatively slow-absorbing protein sources used (i.e., meat and eggs) (12,51). Since the glucagon response depends on the rate of amino acid absorption (52), it is plausible that the magnitude and duration of the glucagon response produced by slow-absorbing proteins are insufficient to stimulate EGP. In contrast, recent studies have shown that ingestion of fast-absorbing whey protein stimulates EGP in people without diabetes (7,10). With a multidose whey protein ingestion protocol (3 × 25 g whey protein boluses ingested over 60 min) a marked and sustained increase in glucagon was demonstrated, resulting in an ∼25% increase in EGP (7). Interestingly, this occurred in the face of a significant insulin response, demonstrating that glucagon can overcome insulin’s ability to suppress EGP. Moreover, blood glucose remained at basal levels as the insulin response caused by protein ingestion increased glucose disposal rates by ∼25%, matching the increment in EGP (7). A subsequent study showed that as little as 25 g whey protein can stimulate EGP (∼20% increase) (10). Notably, the rise in EGP was gradual, peaking at ∼1 h postingestion (10), suggesting that gluconeogenesis was the main pathway contributing to the increased EGP, as opposed to glycogenolysis. This is consistent with the delayed action of glucagon on gluconeogenesis compared with the more rapid action on glycogenolysis, which typically peaks within 15 min (50,53).

From the limited evidence available in type 1 diabetes, protein ingestion in isolation appears to increase blood glucose within 30 min, and the effect persists for several hours. Müller et al. (6) demonstrated a gradual rise in plasma glucose levels following ingestion of lean beef containing ∼75 g protein (1.2 g/kg body wt) in adults and youth with type 1 diabetes, beginning from 30 min postingestion and peaking at ∼2 mmol/L above basal levels at 3.5 h. Similar kinetics were reported by Wahren et al. (12), who found that glucose increased within 30 min of ingestion of 3 g/kg lean beef, peaked at ∼1.5 mmol/L above fasting baseline levels at ∼60 min, and therein remained elevated for 3 h. While these early data demonstrate that protein ingestion causes a gradual, modest, and sustained increase in blood glucose in individuals with type 1 diabetes, it is important to highlight that in these studies, 1) slow-absorbing whole proteins in the form of beef were used and 2) insulin was withheld, resulting in fasting glucose levels of ∼12 mmol/L before protein ingestion, which may have reduced the glycemic response to protein ingestion (35).

While it is known that the glycemic responses to protein ingestion vary among individuals with type 1 diabetes, accumulating evidence suggests that fast-absorbing proteins tend to elicit stronger responses. With use of a constant basal insulin infusion rate, a single dose of 50 g whey protein increased plasma glucose by 3.5 mmol/L at 2.0–2.5 h postingestion (11,54). Notably, glucose responses began to rise within the first hour after ingestion and were sustained for up to 8 h thereafter (11). As noted earlier, in individuals without diabetes, the concurrent endogenous insulin secretory response prevents any significant rise in glycemia following protein ingestion, as any stimulation of EGP is counterbalanced by a precise stimulation of glucose uptake (7). This supports the above findings in type 1 diabetes, where EGP would be expected to rise more significantly given the absence of an endogenous insulin response (26). In contrast, some studies where protein ingestion in type 1 diabetes was explored demonstrated no glycemic impact (55). For example, Klupa et al. (55) demonstrated no significant glycemic response following casein-based protein ingestion (0.3 g/kg body wt) in adults with type 1 diabetes. However, it is important to note that casein is recognized as a slow-digesting protein (Table 1). Thus, while these factors were not measured in the study, the resultant plasma amino acid excursion was likely insufficient to stimulate an adequate glucoregulatory hormonal response.

The absorption rate and amino acid composition of ingested protein are key determinants of the postmeal glycemic response. The protein composition influences the gastric emptying rate, intestinal transit time, and protein accessibility to digestive enzymes (28). For example, due to its acid-soluble properties, whey protein is easily soluble in the stomach, while casein clots in this acidic environment, delaying gastric emptying (56). The distinct physiological effect of “fast”- versus “slow”-digesting protein was first demonstrated by Gannon et al. (29) in individuals with type 2 diabetes, where 25 g gelatin or whey protein powder induced a rapid increase in plasma amino acids, while whole foods such as cottage cheese, soy, and turkey induced moderate increases and egg whites, beef, and fish induced slow and modest increases in amino acids (Table 1). Importantly, absorption rates of common dietary protein sources, especially in the context of type 1 diabetes, are not well characterized. As described by He and Giuseppin (28), available data suggest that most common protein sources have slower absorption rates than whey protein, and although heterogeneity in study design and methodology precludes clear delineation of absorption rates of common proteins, pooled data indicate that casein is slower than whey protein, and free amino acids produce more rapid absorption than whole proteins (28,30,57). Given the well-characterized absorption rate of whey protein, this likely explains why a bulk of data in type 1 diabetes (3,11,13,54,58) and type 2 diabetes is from use of whey protein as an intervention (59). Future data on common (or investigational/bespoke) dietary protein sources, their absorption, and resultant glucoregulatory effects would be beneficial.

It has been observed that, in general, glucagon responses are greatest for the fastest-absorbing proteins (i.e., gelatin), while the slowest-absorbing proteins (i.e., egg white) induce the lowest glucagon responses (29). Calbet and MacLean (52) demonstrated that the glucagon response to protein ingestion was linearly related to the changes in plasma amino acid levels, regardless of the nature of the protein or the degree of protein fractionation (Fig. 3). Thus, an important clinical consideration is not only the total amount of protein ingested but also the rapidity of its absorption, as the glycemic response to protein ingestion in people with type 1 diabetes is related to the rate of absorption and magnitude of the increase in circulating amino acids (6,11,12,54,55). Given the ability of protein to affect glycemia in people with type 1 diabetes, the Food Insulin Index (FII) has been developed, which provides a more holistic approach for insulin dosing (60,61). Briefly, the FII represents the insulin demand in people without diabetes after ingestion of a serving of food (60,61). Compared with traditional carbohydrate counting, FII use significantly mitigates postprandial hyperglycemia in people with type 1 diabetes, particularly after ingestion of high-protein food (61). To this extent, it may be beneficial in future research to develop a “glycemic index” for protein (i.e., an “aminoglucogenic index”) that is specific to people with type 1 diabetes that represents the ability of a set amount of dietary protein to raise blood glucose levels in this population.

In addition to the absorption rate, the amino acid composition may also impact the endocrine response to protein ingestion (52), where certain amino acids can exert a greater glucagon response than others (Table 2). Rocha et al. (62) demonstrated that glucagon secretion was stimulated most strongly by amino acids that most contribute to gluconeogenesis, particularly alanine, serine, glycine (via serine), cysteine, and arginine (63). This may explain the findings from Gannon et al. (29) that beef induces a large glucagon response despite being slow absorbing, which could be due to the high arginine and glycine content (64). Several studies confirm that dietary proteins differ in their capacity to modulate glucagon secretion (29,52,65,66). For example, plasma methionine resultant from pea peptide hydrolysate ingestion and plasma tyrosine from whey hydrolysate or milk ingestion are highly correlated with plasma glucagon (52). However, to our knowledge, no studies have been published with investigation of the effects of protein composition on the glucagon and glycemic responses in people with type 1 diabetes.

The literature on glucoregulatory actions of protein in people with type 1 diabetes is rapidly expanding and a topic of significant debate. As a result, clarification is still needed regarding strategies to manage protein ingestion and the resultant glycemia in people with type 1 diabetes. However, in recent systematic reviews investigators have extensively discussed the research on mealtime insulin dosing for macronutrients other than carbohydrates (1,67). On the other hand, given the clear glycemic stimulatory capacity of fast-absorbing proteins described above, protein ingestion can be viewed not only as a factor to account for postprandially to prevent hyperglycemia but also as a possible intervention to mitigate hypoglycemia. Surprisingly, the utility of protein ingestion to mitigate hypoglycemia in type 1 diabetes has not been extensively explored (68,69). As rapid-absorbing proteins (e.g., protein isolates such as whey, soy, or pea protein) are readily available and largely free of side effects, they could easily be implemented into the lives of people with type 1 diabetes for glycemic management.

Studies among children, adolescents, and adults with type 1 diabetes have shown consistent trends in postprandial glycemia following high-protein mixed meals, with glucose peaking at ∼3.5 h following ingestion and remaining elevated for as long as 12 h for very-high-protein meals (40–75 g protein) (13,70,71). In children with type 1 diabetes, a 30-g carbohydrate meal with additional supplementation with 40 g protein increased postprandial blood glucose concentrations for 3–5 h postmeal, which was found to be protective against hypoglycemia during the subsequent 5-h period (70). Additionally, compared with a low-protein meal (28 g protein with 70 g carbohydrate), a very-high-protein meal (110 g protein with 70 g carbohydrate) in adolescents with type 1 diabetes increased glucose excursions for 12 h, highlighting the sustained glycemic actions of protein (71). Furthermore, Paterson et al. (13) investigated the dose-response relationship between protein intake and glycemia, finding that even small doses of protein (12.5 g), added to a 30-g carbohydrate drink, could enhance the glycemic response for 3–5 h. Importantly, the effect of protein on glycemia (especially in the early postprandial period) is likely masked by the variability in response to the fat and carbohydrate content. To this effect, the American Diabetes Association and International Society for Pediatric and Adolescent Diabetes consensus reports for nutrition therapy in diabetes recommend monitoring glucose levels for 3 h following high-protein meals to determine whether additional insulin adjustments are necessary (72,73).

Hypoglycemia is a significant issue for people living with type 1 diabetes, despite advancements in technology such as continuous glucose monitoring and automated insulin delivery (AID) systems; on average, a person with type 1 diabetes has two episodes of symptomatic hypoglycemia per week (74,75). This is especially evident in situations involving rapid and acute changes in insulin requirements, such as during and after exercise (76). As mentioned briefly, there is a blunted glucagon response to hypoglycemia (16,17) and exercise (18) in people with type 1 diabetes. This likely occurs because α-cell glucagon secretion is critically dependent on upstream paracrine signals from β-cells (i.e., the “intra-islet hypothesis”), which are absent in people with type 1 diabetes (77,78). In contrast, the glucagon response to amino acid appears to be intact in people with type 1 diabetes, as this occurs independent of intra-islet communication (78). Thus, the physiological impact of protein ingestion in people with type 1 diabetes may represent an untapped utility for hypoglycemia prevention, given that the resultant physiological response causes a sustained and moderate increase in glucose concentrations. This gradual and sustained glycemic effect of protein may also be beneficial for those using AID and is in sharp contrast to the large and rapid elevations in glycemia that can result from the currently recommended protocols for exercise-induced hypoglycemia prevention, which typically involve preexercise carbohydrate feeding or reducing mealtime insulin boluses (76). These rapid rises in glucose levels would be expected to elicit a response that increases insulin delivery from commonly used AID systems, which could in some cases paradoxically increase hypoglycemia risk, as the action of subcutaneously delivered insulin cannot be rapidly offset. Accordingly, the modest and gradual glucose-raising actions of protein ingestion may counteract declining glucose concentrations (such as during exercise) without triggering the AID algorithm to significantly increase subcutaneous insulin infusion. Additionally, while the mechanism is unclear, amino acids may be able to influence the central nervous system and possibly improve cognitive functions during hypoglycemia (14) and could potentially work synergistically to restore some counterregulatory glucagon responses to exercise and hypoglycemia, which is typically impaired in type 1 diabetes (14,41).

Nocturnal hypoglycemia is a significant concern for people with type 1 diabetes, as sleep is generally associated with blunting of hypoglycemia awareness, a reduction in the counterregulatory response, and reduced ability to treat hypoglycemia (79). Exercise, as well as alcohol consumption, can result in a delayed and sustained increase in insulin sensitivity and thereby increase the risk of nocturnal hypoglycemia (80). Prior alcohol consumption also has inhibitory actions on hepatic glucose production (by inhibiting gluconeogenesis), which further enhances overnight hypoglycemia risk (81). Overnight hypoglycemia is associated with impaired glycemic control the following day (82); hence, reducing episodes of overnight hypoglycemia can have several physical and psychological benefits for people with type 1 diabetes. Due to protein’s sustained and moderate effect on raising blood glucose levels, use of protein has previously been suggested as a potential prebedtime intervention to minimize hypoglycemia risk (83). Indeed, protein ingestion has been demonstrated to be protective against overnight hypoglycemia (11,84). In adolescents with type 1 diabetes, a high-protein evening meal (25% energy from protein) resulted in significantly fewer (32% vs. 5%) hypoglycemic events in the overnight period in comparison with a low-protein meal (15% energy from protein) (84). Additionally, Paramalingam et al. (11) showed that under experimentally induced euglycemic conditions, consumption of 50 g whey protein reduced the need for exogenous glucose administration by nearly 2.5-fold overnight in adolescents with type 1 diabetes. While results are promising, further free-living studies in people with type 1 diabetes are warranted to determine whether prebedtime protein ingestion is an effective intervention for preventing overnight hypoglycemia.

Despite the health benefits of exercise, many people with type 1 diabetes do not meet physical activity guidelines and are less active than people without diabetes (85). One of the leading barriers to exercise in people with type 1 diabetes is fear of hypoglycemia (85–88). Regardless of the exercise regimen (endurance or resistance training), exercise-induced hypoglycemia in type 1 diabetes occurs due to the nature of subcutaneously delivered insulin, where circulating insulin cannot be decreased rapidly enough to compensate for rapid increases in exercise-induced glucose utilization (89). Current exercise guidelines for type 1 diabetes recommend ingestion of carbohydrates to help mitigate hypoglycemia risk (76), though fast-absorbing protein ingestion may also improve glycemic regulation during exercise.

For the general population, current guidelines by the International Society of Sports Nutrition recommend protein ingestion in ranges between 1.4 and 2.0 g protein/kg body wt/day to facilitate recovery and enhance the adaptive response to exercise (90). Thus, using protein ingestion for glycemic management in people with type 1 diabetes would not represent a significant deviation from current exercise nutrition guidelines. Consistent with the theory that stimulation of glucagon secretion by amino acids would have efficacy in preventing exercise-induced hypoglycemia, recent evidence demonstrates that subcutaneous injection of low-dose recombinant human glucagon is more effective at preventing exercise-induced hypoglycemia than insulin dose reduction (91). In the context of protein ingestion, the stimulation of endogenous glucagon secretion and elevated plasma amino acid levels may have similar EGP-stimulating effects, given the direct secretion of glucagon into the hepatic portal vein and the increased abundance of gluconeogenic substrates. Thus, protein ingestion could also be beneficial in preventing exercise-induced hypoglycemia.

Although direct evidence for protein use with exercise is limited, there is evidence that increased dietary protein is associated with improved glycemia around exercise in adults with type 1 diabetes (92). Increased protein intake was associated with a modest reduction (relative reduction −1.4% per 0.25 g/kg protein) in percent time spent with glucose below target range (<3.9 mmol/L [<70 mg/dL]) for female individuals following exercise (93). Additionally, preexercise protein intake of approximately ≥0.125 g/kg was associated with significantly reduced percent time spent with glucose below target range (>4% absolute reduction) during exercise in adolescents with type 1 diabetes (94). However, it is important to note in these studies only total dietary protein was quantified and a formal protein ingestion intervention was not used. While there is a clear rationale for using protein ingestion as a tool for the prevention of exercise-induced hypoglycemia in people with type 1 diabetes, only two studies in this population exist with examination of the direct effect of pure protein ingestion on the glycemic response to exercise (11,58). Paramalingam et al. (11) demonstrated that in adults with type 1 diabetes, ingestion of 50 g whey protein in the evening following exercise nearly doubled average overnight glucagon levels and produced a sustained overnight increase in glycemia (∼1 mmol/L), a response that may be beneficial for preventing postexercise nocturnal hypoglycemia. Dubé et al. (58) included a ∼0.48 g/kg body wt dose of whey protein in a breakfast meal that was ingested 2 h before 60 min of continuous moderate-intensity exercise in adolescents with type 1 diabetes, with plasma glucose sampled during and after exercise. Compared with results in the control trial, protein ingestion resulted in an overall smaller decrease in plasma glucose during exercise (−4.6 ± 1.9 vs. −6.0 ± 1.9 mmol/L) and significantly reduced the incidence of hypoglycemia, both during exercise (40% vs. 0%) and postexercise (50% vs. 10%) (58). These studies demonstrate the potential efficacy of protein ingestion as a strategy for mitigating exercise-induced hypoglycemia in people with type 1 diabetes. However, more detailed research is clearly needed to fully understand protein ingestion as a targeted intervention in this context.

As far back as the late 1960s it was demonstrated that amino acids have a potent role in stimulating glucagon release, a physiological response that remains intact in people with type 1 diabetes. Combined with the absence of an endogenous insulin response in type 1 diabetes, protein ingestion provides a powerful amino acid– and perhaps GIP-driven stimulus for glucagon secretion. This provides the liver with a synergistic physiological signal to stimulate EGP and thereby increases postprandial blood glucose concentrations, an effect that is moderate in magnitude and sustained for several hours. Amino acid composition and digestibility of the ingested protein are important considerations for this metabolic response, where the most rapid-absorbing proteins appear to have the most potent effects. Although glycemic index is well characterized and understood, no such index is widely accepted for protein’s effect on glycemia, and more research quantifying the impact of protein composition on glycemia in type 1 diabetes would be clinically valuable. Finally, although there is a clear rationale and emerging evidence for using protein ingestion as a tool to prevent overnight or exercise-induced hypoglycemia in people with type 1 diabetes, there is a dearth of well-controlled studies performed in this area. With future research on the utility of fast-absorbing protein to mitigate the risk of hypoglycemia, a clinically useful tool could be provided that could be readily implemented into the lives of people with type 1 diabetes.

Funding. D.P.Z. and D.J.M. have received research support from The Leona M. and Harry B. Helmsley Charitable Trust.

Duality of Interest. D.N.O. has received honoraria from Medtronic, Insulet, Abbott, Novo Nordisk, and Sanofi; has received research support from Medtronic, Insulet, Dexcom, Roche, GlySens, BioCapillary, and Endogenex; and is on advisory boards for Medtronic, Insulet, Abbott, Ypsomed, Novo Nordisk, and Sanofi. C.E.S. has received speaker honoraria from Medtronic, Sanofi, and Eli Lilly and served on advisory boards for Abbott and Medtronic. D.P.Z. has received honoraria for speaking engagements from Ascensia Diabetes Care, Insulet, Medtronic, and Dexcom. D.P.Z. is on an advisory board for Dexcom. No other potential conflicts of interest relevant to this article were reported.

Author Contributions. G.M.D., G.M.K., C.R.B., D.N.O., C.E.M., D.P.Z., D.T.H., S.Z., and D.J.M. were involved in all aspects of this work including conceptualization, design, manuscript preparation and draft reviews. All authors approved the final draft for submission.

Handling Editors. The journal editors responsible for overseeing the review of the manuscript were Steven E. Kahn and Adrian Vella.