Acute Effects of Nutritive and Non-Nutritive Sweeteners on Postprandial Blood Pressure

Postprandial hypotension (PPH) is under-recognised, but common, particularly in the elderly, and is of clear clinical importance due to both the independent association between PPH and an increase in mortality and lack of effective management for this condition. Following health concerns surrounding excessive consumption of sugar, there has been a trend in the use of low- or non-nutritive sweeteners as an alternative. Due to the lack of literature in this area, we conducted a systematic search to identify studies relevant to the effects of different types of sweeteners on postprandial blood pressure (BP). The BP response to ingestion of sweeteners is generally unaffected in healthy young subjects, however in elderly subjects, glucose induces the greatest decrease in postprandial BP, while the response to sucrose is less pronounced. The limited studies investigating other nutritive and non-nutritive sweeteners have demonstrated minimal or no effect on postprandial BP. Dietary modification by replacing high nutritive sweeteners (glucose, fructose, and sucrose) with low nutritive (d-xylose, xylitol, erythritol, maltose, maltodextrin, and tagatose) and non-nutritive sweeteners may be a simple and effective management strategy for PPH.


Introduction
The consumption of nutritive sweeteners has dramatically increased worldwide due to increasing urbanisation and beverage promotion [1]. Nutritive sweeteners commonly found in foods and beverages include glucose, fructose and sucrose [2,3]. More recently, some other sweeteners like d-xylose, xylitol, erythritol, maltose, maltodextrin, stevia and tagatose with lower energy have been used [2,3]. There has been a long history of debate as to the detrimental health effects of sucrose, glucose, and more recently, fructose, particularly when consumed in excess. The effects of both nutritive and non-nutritive sweeteners on cardiovascular risk factors, including blood pressure (BP), have been extensively investigated, however, less is known about their specific effects in the postprandial state. Postprandial hypotension (PPH), a fall in systolic BP (SBP) ≥20 mmHg after a meal, is now recognized as a frequent and important clinical problem, with a reported prevalence of 24-48% in the healthy elderly [4]. PPH is particularly common in conditions associated with autonomic dysfunction e.g., in type 2 diabetes mellitus (T2DM) (~40%) [4,5] and Parkinson's disease (PD) (40-100%) [4,5]. Patients with PPH are at increased risk of syncope and falls [6], and importantly the presence of PPH has been established as an independent predictor of increased mortality [7]. In a cohort study of 401 elderly ambulatory hypertensive subjects, 292 (72.8%) were found to have PPH, and over a 4 year period, 34 died from cardiovascular disease; the post-breakfast BP fall was the strongest predictor of mortality in this cohort [8]. In a 29-month prospective study of 499 nursing home residents, the maximum subsequent falls, syncope, new cardiovascular events (myocardial infarction and stroke) and total mortality [9].
The pathophysiology of PPH is multifactorial and incompletely understood, however, it involves the interplay between autonomic and neural mechanisms, including the release of gut hormones, which are influenced by meal composition, gastric distension, and small intestinal nutrient delivery [4]. After a meal, there is a doubling of the superior mesenteric artery (SMA) blood flow [5] and in healthy young individuals with intact baroreflex function, the increase in splanchnic blood flow is accompanied by concomitant increases in heart rate (HR), peripheral vascular resistance, stroke volume and cardiac output [10]. In patients with PPH, these compensatory responses are inadequate [5]. The postprandial fall in BP is greater when gastric emptying is more rapid [11][12][13], whereas gastric distension attenuates the fall in BP in both healthy young and older participants [11][12][13]. The macronutrient composition of the meal is also an important determinant of postprandial BP, with carbohydrate inducing an earlier fall in BP compared with protein and fat [14] ( Figure 1). The interaction between nutrients and the small intestine stimulates the secretion of the incretin hormones, glucagon-like peptide-1 (GLP-1) from L-cells and glucose-dependent insulinotropic polypeptide (GIP) from K-cells [15]. GLP-1 stimulates insulin secretion, inhibits glucagon secretion and slows gastric emptying [16,17]. Acute infusion of GLP-1 has been shown to attenuate the fall in BP and increases SMA blood flow following oral [16] or intraduodenal (ID) glucose [16,18]. In contrast, exogenous GIP has been shown to increase the rate of gastric emptying [19], although results have been conflicting [20] and there have been no studies directly investigating the effect of GIP on postprandial BP. A reduction in BP is not observed following intravenous administration of glucose [21], which is a substantial stimulus to insulin secretion, suggesting that insulin by itself if not an important mediator of PPH. Furthermore, PPH is seen in the absence of insulin secretion i.e., in those with type 1 diabetes [22]. Values are means ± SEM. * p < 0.05 for glucose compared with saline; † p < 0.05 for glucose compared with fat; ‡ p < 0.0001 for glucose compared with protein; ^ p < 0.05 for fat compared with saline; # p < 0.05 for protein compared with saline; + p < 0.05 for fat compared with protein [14]. BPM = beats per minute.
In recent years, due to several negative health effects from the excessive consumption of added sugar, a variety of non-nutritive artificial sweeteners have increasingly been used in the food industry as an alternative to sugar among people of all ages, particularly directed to the obese and those with diabetes, with the aim of reducing energy intake and minimising the risk of obesity, diabetes and cardiovascular disease [3,23]. Beverages account for the highest percentage of non-nutritive sweetener utilisation in the world [24,25]. In the United States, non-nutritive sweeteners were used in up to 32% and 19% of beverages consumed by adults and children respectively from 2007 to 2010 ‡ p < 0.0001 for glucose compared with protein;ˆp < 0.05 for fat compared with saline; # p < 0.05 for protein compared with saline; + p < 0.05 for fat compared with protein [14]. BPM = beats per minute.
In recent years, due to several negative health effects from the excessive consumption of added sugar, a variety of non-nutritive artificial sweeteners have increasingly been used in the food industry as an alternative to sugar among people of all ages, particularly directed to the obese and those with diabetes, with the aim of reducing energy intake and minimising the risk of obesity, diabetes and cardiovascular disease [3,23]. Beverages account for the highest percentage of non-nutritive sweetener utilisation in the world [24,25]. In the United States, non-nutritive sweeteners were used in up to 32% and 19% of beverages consumed by adults and children respectively from 2007 to 2010 [26]. In 2018, the American Heart Association suggested that short-term use of "non-nutritive" sweeteners instead of "nutritive sweeteners" can reduce caloric intake in the management of excess weight and obesity [27].
There are some differences between regulatory bodies regarding approval of nutritive and non-nutritive sweeteners. The six non-nutritive sweeteners approved by Food and Drug Administration (FDA) as food additives in the United States include sucralose, Acesulfame-K, aspartame, saccharin, neotame and advantame [2,28,29]. Stevia [29,30] has not received official FDA approval but has been rated as generally recognized as safe (GRAS) and is being increasingly used as a non-nutritive sweetener [31]. Maltitol similarly has not received official FDA approval, but may be used in food manufacture [31]. The nutritive sweeteners currently approved by the FDA include mannitol, xylitol, sorbitol and erythritol [32]. The European Food Safety Authority (EFSA) has approved the non-nutritive sweeteners Acesulfame-K, aspartame, cyclamate, neohesperidin dihydrochalcone, saccharin, sucralose, neotame, thaumatin and stevioside, while approved nutritive sweeteners include erythritol, isomalt, lactitol, maltitol, mannitol, sorbitol and xylitol [31].
There have been a number of short-term studies evaluating the physiological responses to nutritive and non-nutritive sweeteners that have included assessment of BP. With the increasing widespread use of non-nutritive sweeteners and the documented severe clinical sequelae of PPH, it is clearly important to ascertain the extent to which these agents decrease postprandial BP. To date, a published systematic summary of the impact of nutritive and non-nutritive sweeteners on postprandial BP has not been performed. For this systematic review, we aim to establish current scientific evidence of the effects of some of the most commonly used FDA-regulated nutritive and non-nutritive sweeteners on postprandial BP, as well as their potential applications in the management of PPH.

Approach
Based on the pathophysiology of PPH, a systematic review of the PubMed database was conducted up to March 2019 to identify articles related to the effects of different types of sweeteners on postprandial BP. The following keywords were used: "sweeteners" or "nutritive sweeteners" or "caloric sweeteners" or "sucrose" or "fructose" or "glucose" or "xylose" or "xylitol" or "erythritol" or "maltose" or "maltodextrin" or "tagatose" or "non-nutritive sweeteners" or "non-caloric sweeteners" or "sucralose" or "Acesulfame-K" or "aspartame" or "saccharin" or "stevia" or "steviol glycoside" or "neotame" or "advantame") and ("postprandial blood pressure" or "postprandial hypotension" or "hypotensive response").
Screening of studies was performed initially by assessment of the relevance of the abstract by two independent reviewers. The reference lists of relevant articles were also reviewed. Study inclusion was based on the guidelines for preferred reporting items for systematic reviews and meta-analyses (PRISMA) [33].

Results
A total of 150 papers were identified from the database search. There were no duplicate papers. Of the 150 papers, papers without full text (n = 2), irrelevant papers (n = 55), animal studies (n = 12), reviews (n = 9), papers that were unavailable in the English language (n = 7), letters to the editor or comments (n = 1), and case reports (n = 2) were removed, leaving a total of 62 papers which were reviewed. Figure 2 summarizes the selection process.
Oral fructose, but not glucose, significantly increased SBP and DBP. The maximum rise in SBP after fructose was 6.2 ± 0.8 mmHg.
Small and sometimes significant increases in BP occurred during the sessions. There were falls in SBP and DBP and a rise in HR during ID glucose; granisetron had no effect on these responses.
In patients with progressive autonomic failure and MSA, glucose ingestion caused a decrease in BP which was attenuated by SMS-201-995.
Randomized crossover study 300 mL drink of either 75 g glucose or 75 g fructose.
Glucose decreased MAP significantly in the EH, EN and YH group. After fructose, BP remained unchanged in 4 groups. Non-randomized study 75 g glucose in 350 mL water The fall in MAP was significantly greater in the T2DM than in older healthy participants with no change in young healthy participants. The magnitude of the fall in BP was related to the rate of gastric emptying.
Jones et al. [11] 2005 10 healthy participants, 6M:6F, aged 73.9 ± 1.2 years Randomized crossover study 25 g glucose in 200 mL (12.5%), 75 g glucose in 200 mL (37.5%), 25 g glucose in 600 mL (4%), and 75 g glucose in 600 mL (12.5%) Increased drink volume attenuates the fall in BP with no effect of glucose concentration.  16 participants had PPH. In PPH, there was a significant fall in SBP (26.69 ± 8.43 mmHg) on the "G" day and no change on "GW" or "WW" days. In those without PPH, there were no changes in SBP on the "G" or "GW" days, with an increase in SBP on the "WW" day.  Between t = 0-60 min, SBP was lower during the glucose-only infusion than during the glucose and guar infusion. The maximum fall in SBP on the glucose-only study was 10 ± 4 mmHg. Between t = 0-30 min, DBP fell during the glucose-only infusion, but did not change with the glucose and guar infusion.
MAP decreased during glucose but not during sucralose or saline. By t = 60 min, MAP was lower after glucose (85.9 ± 2.8 mmHg) than after sucralose (93.1 ± 2.2 mmHg) infusions without significant difference between sucralose and saline infusions.
Pham et al. [   There was a fall in SBP and DBP during 2 and 3 kcal/min glucose infusions, but not during saline or 1 kcal/min glucose infusion. There was no difference in the maximum falls in SBP during 2 kcal/min (15 ± 2 mmHg) and 3 kcal/min (12 ± 2 mmHg) loads.
SBP and DBP fell during G, but not during S or GD; and increased during SD. The maximum changes in SBP during G, GD, S and SD were −14 ± 5, −3 ± 4, +11 ± 2, and +15 ± 3 mmHg respectively.
SBP and DBP fell during V0, but did not change significantly during V100, V300, V500.

Intraduodenal Glucose Infusion
When the protective mechanism of gastric distension [86,87] is bypassed by infusing glucose directly into the duodenum, the fall in BP is greater compared to oral glucose in healthy older participants [54]. Vanis et al. showed that gastric distension, by as little as 100 mL water, could mitigate the fall in BP induced by ID glucose in healthy older participants, supporting the potential use of non-nutritive gastric distension in the management of PPH [87]. ID glucose-induced reductions in BP are observed in both older people [54,75,81] and patients with T2DM [76,85,88]. The hypotensive response to ID glucose depends primarily on the small intestinal glucose load, but solution concentration does not appear to be important [74]. Within the normal physiological range of gastric emptying (1-4 kcal/min) [90], the rate of small intestinal delivery correlates with the magnitude of the fall in BP [76,77,79]. Trahair et al. reported that SBP decreased significantly during 2 kcal/min and 3 kcal/min glucose infusions, but not during 1 kcal/min glucose infusion; indicating that a threshold for the fall occurs between 1-2 kcal/min [91]. On the contrary, young healthy participants with sufficient cardiovascular responses remain normotensive [79] or become only slightly hypotensive [81] after ID glucose infusion. There is no difference in postprandial BP following ID glucose infusion between young healthy and obese groups [80].

Management of PPH
Due to its ability to induce a prominent hypotensive response, several studies investigating non-pharmacological [63,67,68,84,86,87] and pharmacological [18,55,[57][58][59][60][61]64,65,83,85,88] approaches to the management of PPH have used glucose as test meals. Interventions based on slowing gastric emptying [63,84] and small intestinal absorption of nutrients [82] attenuate the postprandial fall in BP. Guar gum, a vegetable-based, gel-forming non-absorbable carbohydrate, commonly used as a bulking agent, delays gastric emptying, slows the intestinal absorption of glucose and, accordingly, attenuates the magnitude of the fall in BP after oral and ID glucose [13,63,84]. Intermittent walking may also be effective in preventing PPH induced by glucose ingestion [67,68]. In terms of the pharmacologic management of PPH, numerous medications have been tested. One study has shown that maximal postprandial fall in sitting SBP after a standardized 400 kcal glucose drink is attenuated by caffeine [55], although the benefits of caffeine as a treatment for PPH are inconsistent and largely empirical. The effects of the somatostatin analogue, octreotide, in attenuating the fall in BP after oral glucose has been demonstrated in three studies [58,60,61]. These effects are not mediated via the inhibition of insulin secretion [61]. Voglibose, an alpha-glucosidase inhibitor, is effective in attenuating PPH in neurologic patients [65]. More recently, GLP-1 based therapy has drawn more attention. In a placebo-controlled study, intravenous GLP-1 infusion (0.9 pmol kg −1 .min −1 ) in 14 healthy older individuals and 10 patients with T2DM, resulted in the slowing of gastric emptying and attenuation in the fall in BP following ingestion of a radiolabelled 75 g glucose drink [16]. GLP-1 also reduces the maximum fall in SBP in response to ID glucose [18]. Both short-acting GLP-1 receptor agonists, exenatide [85] and lixisenatide [64], known to slow gastric emptying, markedly attenuate the decrease in BP compared to placebo. In contrast, the dipeptidyl peptidase-4 inhibitor (DPP-4), vildagliptin, known to block the degradation of the incretin hormones GLP-1 and GIP, did not attenuate the fall in SBP and DBP during ID glucose administration compared with placebo in T2DM [88]. Recent evidence suggests that GIP may have detrimental effects to lower postprandial BP and it is possible that the beneficial effects of GLP-1 by DPP-4 inhibition were nullified by GIP. The effect of the anti-diabetic medication, metformin, has also been evaluated in two studies [38,88], but the outcomes are inconsistent and more studies are warranted. Other investigators have reported that the antihypertensive, nitrendipine [59], or a combination of selective α1 and β1-adrenergic agonists [57], might attenuate the postprandial fall in BP after glucose ingestion, while the antiemetic drug, granisetron, a 5-HT3 antagonist, [83] has no benefit in the management of PPH.

Fructose
Fructose, a monosaccharide found in fruit, honey, and some vegetables [2], may be preferred over glucose by consumers and cooks due to its intrinsically greater sweetness and ability to improve the appearance and texture of baked goods [92]. Fructose has been used increasingly as an alternative to glucose or sucrose [93][94][95] in several processed foods and beverages, especially in diets targeting patients with T2DM, as it provides fewer calories for the same level of sweetness as glucose [96,97]. In contrast to glucose, the metabolism of fructose primarily occurs in the liver, entering cells in an insulin-independent fashion [98]. Fructose is not an insulin secretagogue [98], inducing a substantially lower glycaemic response compared with glucose [95]. However, although there is no clear consensus in the literature, the potential for increased de novo lipogenesis from excessive fructose intake, particularly in those with insulin resistance, has raised the concern regarding the widespread use of this monosaccharide [99]. Compared with glucose, fructose is absorbed more slowly from the intestine [100,101]. There has been a total of five studies relating to the effects of oral fructose on postprandial BP in a range of cohorts including in healthy participants [39][40][41]50,89] and hypertensive patients [41] (Table 2).
Some studies have demonstrated no change in BP in healthy young [41] and older participants [50] as well as those with hypertension [41] following ingestion of fructose. In healthy older people, SBP decreased significantly from baseline following glucose: −3.96 ± 1.38 mmHg but not after the fructose drink: 2.59 ± 1.62 mmHg [41] (Figure 3). Two other studies reported an increase in BP in healthy young participants following ingestion of fructose [39,40]. This may be due to the slower intestinal absorption rate of fructose and lower glycaemic response compared with glucose [102,103], which may also result in a limited increase in SMA blood flow. However, there have been no studies comparing the difference in postprandial intestinal vascular perfusion following oral glucose and fructose. In addition, due to the lower insulinotropic effects of fructose compared with glucose and sucrose [104], effects on total vascular peripheral resistance may be mitigated [39,89].

Fructose
Fructose, a monosaccharide found in fruit, honey, and some vegetables [2], may be preferred over glucose by consumers and cooks due to its intrinsically greater sweetness and ability to improve the appearance and texture of baked goods [92]. Fructose has been used increasingly as an alternative to glucose or sucrose [93][94][95] in several processed foods and beverages, especially in diets targeting patients with T2DM, as it provides fewer calories for the same level of sweetness as glucose [96,97]. In contrast to glucose, the metabolism of fructose primarily occurs in the liver, entering cells in an insulin-independent fashion [98]. Fructose is not an insulin secretagogue [98], inducing a substantially lower glycaemic response compared with glucose [95]. However, although there is no clear consensus in the literature, the potential for increased de novo lipogenesis from excessive fructose intake, particularly in those with insulin resistance, has raised the concern regarding the widespread use of this monosaccharide [99]. Compared with glucose, fructose is absorbed more slowly from the intestine [100,101]. There has been a total of five studies relating to the effects of oral fructose on postprandial BP in a range of cohorts including in healthy participants [39][40][41]50,89] and hypertensive patients [41] (Table 2).
Some studies have demonstrated no change in BP in healthy young [41] and older participants [50] as well as those with hypertension [41] following ingestion of fructose. In healthy older people, SBP decreased significantly from baseline following glucose: −3.96 ± 1.38 mmHg but not after the fructose drink: 2.59 ± 1.62 mmHg [41] (Figure 3). Two other studies reported an increase in BP in healthy young participants following ingestion of fructose [39,40]. This may be due to the slower intestinal absorption rate of fructose and lower glycaemic response compared with glucose [102,103], which may also result in a limited increase in SMA blood flow. However, there have been no studies comparing the difference in postprandial intestinal vascular perfusion following oral glucose and fructose. In addition, due to the lower insulinotropic effects of fructose compared with glucose and sucrose [104], effects on total vascular peripheral resistance may be mitigated [39,89].   Fructose 500 mL of either water, 60 g glucose, or 60 g fructose.
Oral fructose, but not glucose, significantly increased SBP and DBP. The maximum rise in SBP after fructose was 6.2 ± 0.8 mmHg.
Ingestion of fructose (60 or 30 g) elevated SBP, DBP and MAP. Ingestion of glucose elevated SBP. Ingestion of sucrose showed no BP changes. The increases in DBP and MAP were significantly higher for fructose (60 or 30 g) than for either glucose or sucrose. The increase in SBP was significantly higher for fructose than for sucrose. An iso-osmotic solution of glucose (1 g/kg body weight) or xylose (0.83 g/kg body weight) in 250 mL water.
Xylose caused a lower and more transient fall in BP than glucose in patients with CAF (15 ± 6% vs. 34 ± 7%). After glucose, there was a substantial fall in 6 age-matched participants but a minimal change in 8 male normal participants. T2DM = type 2 diabetes; BP = blood pressure; SBP = systolic blood pressure; DBP = diastolic blood pressure; ID = intraduodenal; MAP = mean arterial pressure; MSA = multi-system atrophy; PPH = postprandial hypotension; OH = orthostatic hypotension.

Sucrose
Sucrose is a common table sugar produced naturally in fruit and vegetables and often added to many processed foods [108]. Sucrose tastes sweeter than glucose, but not as sweet as fructose. The digestion of sucrose is initiated in the mouth and stomach, but the majority of this disaccharide is hydrolysed and digested in the small intestine by the α-glucosidase sucrase to release an equimolar mixture of glucose and fructose [109]. There have been five studies relating to effects of sucrose on postprandial BP in healthy participants [40,50,[105][106][107] (Table 2): four used an oral load [40,50,105,107] while one study used an ID infusion [106].
The cardiovascular effects of sucrose are more similar to those observed following glucose than fructose, despite its equimolar constitution [110]. Visvanathan et al. reported that ingestion of 50 g sucrose or 50 g glucose induced a comparable fall in SBP in healthy elderly individuals, but the decrease in SBP occurred earlier after glucose, than sucrose, ingestion [50] (Figure 3). This may be due to the fact that sucrose is a disaccharide, requiring hydrolysis prior to mediating cardiovascular effects [50]. In comparison, a 60 g sucrose drink had no substantial impact on BP in healthy young participants [40], which may be due to the more effective postprandial haemodynamic responses compared with an older population. The α-glucosidase inhibitor, acarbose, is approved for the treatment of T2DM and is known to reduce postprandial glycaemia [111] through stimulation of GLP-1 [112], and slowing of gastric emptying [105,113]. In healthy older participants, acarbose substantially decreases falls in SBP and DBP induced by not both oral [105], and ID [106] sucrose, loads.

d-Xylose and Xylitol
Xylose, a monosaccharide of the aldopentose type consists of five carbon atoms and is derived from wood. The free aldehyde group reduces its sweetness and makes it a widely used alternative for glucose in the diabetic diet. Xylose is primarily absorbed by passive penetration through the wall of human duodenum and jejunum [114], with the remainder delivered to the ileum and the colon. Xylose is emptied from the stomach at a similar rate to glucose [49]. There have been three studies related to effects of oral xylose on postprandial BP in healthy older participants [43,45,49], orthostatic hypotension (OH) [45] and autonomic failure [43,45] patients (Table 2). While Mathias et al. [43] reported xylose induced a milder and more transient fall in postprandial BP compared with glucose in those with chronic autonomic failure, Robinson et al. reported that glucose and xylose ingestion induced a comparable fall in SBP at 60-90 min in supine patients diagnosed with OH and autonomic failure [45]. In a previous study by our group, designed to compare effects of glucose and xylose on BP, gastric emptying and incretin hormones, xylose emptied from the stomach at a comparable rate to glucose, with a greater and more prolonged effect on GLP-1 secretion in healthy older participants [49]. However, there was no change in postprandial BP observed, potentially due to the effects of GLP-1 in delaying the nutrient absorption in the distal small intestine. These findings support the replacement of glucose by xylose as a simple management to decrease the postprandial fall in BP [49].
Reduction of xylose by catalytic hydrogenation produces xylitol [115]. Xylitol has a comparable sweetness to sucrose and contains less than one-third of calories in conventional sugars. As xylitol is metabolized in an insulin-independent manner, its use has been widely advocated in the prevention and control of hyperglycaemia, obesity, and related metabolic disorders [116,117]. However, xylitol is often associated with diarrhoea due to its stimulatory effect on the intestinal transit and the release of plasma motilin [118]. Salminen et al. [118] reported that 30 g xylitol solution was emptied from the stomach at a significantly slower rate than a 30 g of glucose, suggesting that xylitol might protect against PPH, although further studies are required.

Erythritol
Erythritol, a four-carbon pentose, is made through food fermentation but may occur naturally in a wide variety of fruits and mushrooms. Erythritol has a sweetness of~70% compared with sucrose and has minimal insulinotropic effects [119,120]. Nevertheless, the effect erythritol on BP has not been studied. In patients with T2DM, acute consumption of erythritol has vasoprotective effects by decreasing oxidative stress and endothelial dysfunction [121].

Maltose and Maltodextrin
Maltose, or malt sugar, a disaccharide composed of two glucose units, is an intermediate in the intestinal breakdown of glycogen and starch and is found in germinating grains [122]. Maltodextrin is a nutritive saccharide polymer consisting of D-glucose units with a low sweetness (dextrose equivalency of less than 20). It is often used in the production of soft drinks, candy, and processed food [123]. Both maltose [124] and maltodextrin [125] are emptied more slowly from the stomach compared with sucrose, although there is no significant difference in gastric emptying between maltodextrin and glucose [125]. To our knowledge, there have been no studies investigating the effects of maltose on postprandial BP. There has been only one study related to effects of oral maltodextrin on postprandial BP in healthy older participants [107] (Table 2) in which postprandial MAP and DBP were shown to be greater after sucrose compared to after maltodextrin [107]. The underlying mechanisms remain unclear [107].

Tagatose
Tagatose has a very similar structure to sucrose, is commonly found in dairy products [126] and is low calorie; only 20 to 25% of ingested D-tagatose is absorbed through the small intestine to provide 1.5 kcal/g [127]. The remaining ingested tagatose is fermented into short-chain fatty acids by colonic bacteria [2,128]. Tagatose has a substantial hypoglycaemic effect without major adverse effects, contributing to its potential therapeutic benefits in obese people and patients with T2DM [129][130][131][132]. While tagatose has been shown to slow gastric emptying [133,134], to our knowledge there is no information regarding the effect of tagatose on BP.

Sucralose
Sucralose, a noncaloric artificial sweetener with a sweetness approximately 600 times higher than that of sucrose [135], is increasingly being used as a sugar substitute to reduce calorie intake, especially in obese people and those with diabetes [3]. It is almost unabsorbed in the small intestine and excreted in the faeces in both animal [136,137] and human [138] experiments. In healthy participants, sucralose empties at a similar rate to saline [139].
There have been two studies investigating effects of sucralose on postprandial BP [78,140] (Table 3). Kazmi et al. reported that there was no change in BP following oral sucralose in 200 young healthy participants compared to saline. In a recent study by our group, while ID infusion of glucose at a rate of 3 kcal/min induced a substantial fall in postprandial BP and increase in SMA blood flow in 12 healthy older participants, there were no significant effects on BP or splanchnic blood flow following an ID infusion of sucralose [78] (Figure 4). Studies by others have shown that sucralose has no effect on insulin, GLP-1 or GIP secretion or SMA blood flow [139,141], supporting a therapeutic role for sucralose in the dietary management of PPH. ID infusion of either glucose (25%, 1400 mOsmol/L), sucralose (4 mmol/L,~300 mOsmol/L) or saline (0.9%,~300 mOsmol/L) at a rate of 3 mL/min for 60 min followed by ID saline for a further 60 min.
MAP decreased during glucose but not during sucralose or saline. By t = 60 min, MAP was lower after glucose (85.9 ± 2.8 mmHg) than after sucralose (93.1 ± 2.2 mmHg) infusions without significant difference between sucralose and saline infusions.

Acesulfame-K
Acesulfame-K, a non-nutritive sweetener, was invented in 1967 by the pharmaceutical company, Hoechst AG. It is estimated to be 200 times sweeter than sucrose and is often mixed with sucralose or aspartame in foods and beverages [28]. In 2009, Brown et al. reported that the GLP-1 release increased by more than 30% after drinking "diet soda", comprising sucralose, aspartame and Acesulfame-K compared to carbonated water [142]. However, this finding was not replicated in subsequent studies [141,143]. Acesulfame-K alone or combined has no effect on GLP-1 secretion, insulin, or blood glucose. There has been only one study investigating the effect of Acesulfame-K on BP [140] (Table 3). Kazmi et al. reported that SBP was lower at 60 min following the ingestion of 3.24 gm (45 mg/kg) of Acesulfame-K in 50 healthy young participants compared to controls. The design of that study was, however, suboptimal, as SBP and DBP was only measured at 30, 60, 90 and 120 min and the baseline measurement was not taken into account.

Aspartame
Aspartame, a methyl ester of aspartic acid and phenylalanine dipeptide, was approved as a nonnutritive sweetener in 1996 by the FDA. It is 160 to 220 times sweeter than sucrose and the only nonnutritive sweetener providing energy (4 kcal/g). However, due to the intense sweetness, only a tiny quantity is used in foods and soft drinks and the number of calories is negligible, hence, it is considered 'non-nutitive' [2,3]. Little et al. [144] demonstrated that the gastric emptying of aspartame does not differ from that of water. There has been only one study investigating the effect of aspartame on BP [140] (Table 3). In this study, young healthy participants experienced falls in SBP at 60, 90 and 120 min following the ingestion of 10.8 g (150 mg/kg) aspartame compared to 10 g of cellulose (control) ( [140]. Again, the design of that study was, suboptimal, as SBP and DBP were only measured at 30, 60, 90 and 120 min and the baseline measurement was not taken into account.

Saccharin
Saccharin is the oldest non-nutritive sweetener that is currently approved, is not metabolized in humans and is considered non-carcinogenic [2,3]. To date, there are no studies investigating the acute effects of saccharin on BP. Values are means ± SEM. MAP: p < 0.05 for glucose compared with sucralose or saline, and p = 1.0 for sucralose compared with saline. Heart rate: p < 0.05 for glucose compared with sucralose, p = 0.07 for glucose compared with saline, and p = 1.0 for sucralose compared with saline [78].

Acesulfame-K
Acesulfame-K, a non-nutritive sweetener, was invented in 1967 by the pharmaceutical company, Hoechst AG. It is estimated to be 200 times sweeter than sucrose and is often mixed with sucralose or aspartame in foods and beverages [28]. In 2009, Brown et al. reported that the GLP-1 release increased by more than 30% after drinking "diet soda", comprising sucralose, aspartame and Acesulfame-K compared to carbonated water [142]. However, this finding was not replicated in subsequent studies [141,143]. Acesulfame-K alone or combined has no effect on GLP-1 secretion, insulin, or blood glucose. There has been only one study investigating the effect of Acesulfame-K on BP [140] (Table 3). Kazmi et al. reported that SBP was lower at 60 min following the ingestion of 3.24 gm (45 mg/kg) of Acesulfame-K in 50 healthy young participants compared to controls. The design of that study was, however, suboptimal, as SBP and DBP were only measured at 30, 60, 90 and 120 min and the baseline measurement was not taken into account.

Aspartame
Aspartame, a methyl ester of aspartic acid and phenylalanine dipeptide, was approved as a non-nutritive sweetener in 1996 by the FDA. It is 160 to 220 times sweeter than sucrose and the only non-nutritive sweetener providing energy (4 kcal/g). However, due to the intense sweetness, only a tiny quantity is used in foods and soft drinks and the number of calories is negligible, hence, it is considered 'non-nutritive' [2,3]. Little et al. [144] demonstrated that the gastric emptying of aspartame does not differ from that of water. There has been only one study investigating the effect of aspartame on BP [140] (Table 3). In this study, young healthy participants experienced falls in SBP at 60, 90 and 120 min following the ingestion of 10.8 g (150 mg/kg) aspartame compared to 10 g of cellulose (control) ( [140]. Again, the design of that study was, suboptimal, as SBP and DBP were only measured at 30, 60, 90 and 120 min and the baseline measurement was not taken into account.

Saccharin
Saccharin is the oldest non-nutritive sweetener that is currently approved, is not metabolized in humans and is considered non-carcinogenic [2,3]. To date, there are no studies investigating the acute effects of saccharin on BP.

Steviol Glycoside
Steviol glycoside sweeteners are extracted and purified from the leaves of the Stevia rebaudiana Bertoni plant, which is native to South America. Its sweet potency is reported to be 200 to 400 times sweeter than table sugar [145,146]. Steviol glycoside has been categorized as a non-caloric sweetener and widely used in the food and soft drinks consumed by millions of people and is well tolerated [126]. In one study, postprandial blood glucose and insulin levels were significantly lower when healthy lean and obese participants consumed a steviol glycoside preload before meals compared with an isocaloric sucrose preload [147]. Another study in participants with T2DM reported that steviol glycoside reduced postprandial blood glucose without a significant change in insulin levels [148]. Furthermore, steviol glycoside was also shown to significantly reduce food intake [147]. Peirera et al. [149] did not demonstrate any effect of steviol glycoside on gastric emptying of a liquid test meal in 10 male healthy volunteers. Most of the studies in the literature have evaluated the effect of long-term steviol glycoside intake on BP and demonstrated that it may be effective in decreasing SBP and DBP compared with placebo in hypertensive [150,151], but not in normotensive or hypotensive, participants [152,153]. However, there have been no publications evaluating the acute effect of this sweetener on postprandial BP.

Neotame and Advantame
Neotame is closely related to aspartame in terms of the chemical constitution. It is 7000-13,000 times sweeter than sucrose [154]. Advantame is considered the most potent non-caloric sweetener with sweetness approximately 20,000 times greater than sucrose [154]. Advantame is derived from isovanillin and aspartame [155]. To our knowledge there have been no studies evaluating the effects of these sweeteners on BP.

Conclusions
PPH is associated with an increased incidence of falls, syncope, angina and transient ischaemic attacks, particularly in older people and patients with autonomic failure that is frequently secondary to diabetes mellitus and PD. For nutritive sweeteners, glucose induces the greatest fall in postprandial BP and should be limited in people with PPH. Most of the currently available literature indicates that non-nutritive sweeteners have little effect on BP. While pharmacological and dietary approaches are being explored, current management of PPH remains suboptimal. Our literature review supports the view that low nutritive (d-xylose, xylitol, erythritol, maltose, maltodextrin, and tagatose) and non-nutritive sweeteners could be used to replace high nutritive sweeteners (glucose, fructose, and sucrose) in the dietary management of PPH.