Effect of nutritive and non-nutritive sweeteners on hemodynamic responses to acute stress: a randomized crossover trial in healthy women

The study investigates whether consumption of sweetened beverages (nutritive sucrose or non-nutritive sweeteners) modulates hemodynamic responses to acute stress in humans. Psychosocial stress contributes to cardiometabolic risk through mechanisms such as increased systemic vascular resistance and blood pressure, potential myocardial ischemia, and stress-induced eating leading to visceral adiposity and insulin resistance. Prior reports suggest sucrose can reduce sympatho-adrenal activation and stress responses in animals and humans, which raises the hypothesis that sweet taste or caloric content may attenuate hemodynamic stress responses. The authors aim to test, in healthy women, whether sucrose- or NNS-sweetened drinks alter SVR and other hemodynamic parameters during mental stress (predominantly β-adrenergic) and CPT (predominantly α-adrenergic).

Evidence links stress to increased preference for energy-dense, sweet foods and to glucocorticoid-driven visceral fat deposition. Animal studies showed sweetened drink access attenuates HPA axis responses and downregulates CRF expression; proposed mechanisms include non-metabolic sweet taste effects via opioid pathways, though some data indicate energy-dependent metabolic effects of sucrose are also involved. In humans, a high-sucrose diet reduced salivary cortisol response to mental stress, whereas an NNS-containing diet increased it. Acute stress can elicit myocardial ischemia through hemodynamic and neurohumoral pathways. Mental stress typically increases heart rate, mean arterial pressure, and cardiac output with no change or slight decrease in SVR, attributed partly to β-adrenergic, nitric oxide–mediated skeletal muscle vasodilation. CPT reliably increases SVR, heart rate, and mean arterial pressure without changing cardiac output. Pediatric literature shows intraoral sucrose reduces pain responses and heart rate in newborns, implicating taste mechanisms; effects differ with non-sweet caloric solutions, suggesting sweetness per se. Developmental differences in taste sensitivity and reward processing may explain discrepancies between pediatric and adult responses.

Ethics and registration: Conducted Sept 2015–Oct 2016 in accordance with the Declaration of Helsinki; approved by the Human Research Ethics Committee of Canton de Vaud; registered at ClinicalTrials.gov (NCT02973334). Participants provided written informed consent. Participants: Recruited by advertisement; Caucasian females aged 18–40 years, BMI 18.5–25 kg/m^2, using monophasic oral contraceptives. Exclusions included hypertension (>140/90 mmHg), psychological/cardiovascular disorders, anemia, high daily intake of caloric or NNS beverages (>5 dL), caffeine (>400 mg), alcohol (>10 g), color blindness, recent weight change (>3 kg in 4 weeks), physical activity >4 h/week, current medications, smoking, illicit drug use. Sixteen screened in familiarization; twelve completed (mean ± SD: age 21.8 ± 2.4 y; BMI 21.6 ± 1.7 kg/m^2). Design: Randomized, crossover with three conditions: Water, Sucrose, and NNS. Each participant completed three functional evaluations separated by 1–3-week washouts. Pre-evaluation standardization included a 2-day energy-balanced weight-maintenance diet; avoidance of sugar-rich and NNS-rich foods, caffeine, and alcohol; ensuring ≥8 h sleep; and refraining from structured physical activity. Compliance assessed via diet, activity, and sleep records. Procedures: Participants arrived at 0700 after overnight fast (since 2200). After voiding and weighing, thoracic electrical bioimpedance sensors were placed (BioZ AdvaSense). At 0800 (time 0 min), a 90-min resting period began, followed by 30-min mental stress (time 90–120 min) alternating 5-min Stroop color–word interference and 5-min complex mental arithmetic, then 30-min recovery, then a 3-min cold pressor test (CPT) at time 150 min (right hand immersion in ice water). Intervention drinks: From time 60 min and every 15 min thereafter to time 150 min (times 60, 75, 90, 105, 120, 135, 150 min), participants ingested 25 mL of test drink after a 10-s mouth rinse with the same liquid. Conditions: Water; Sucrose 106 g·L^-1; NNS mixture reflecting commercial beverages: sodium cyclamate 392 mg·L^-1, acesulfame K 181 mg·L^-1, aspartame 116 mg·L^-1. Participants were instructed to rinse for 10 s prior to swallowing to activate oral sweet taste receptors. Measurements: Mean arterial pressure (MAP = 1/3 systolic + 2/3 diastolic) measured at times 30, 45, 75, 85, 95, 100, 105, 110, 115, 120, 125, 130, 135, 145, 153, 158, 163 min using an automatic sphygmomanometer (BioZ ICG Monitor) on the left arm; triplicate readings averaged. Cardiac output continuously recorded by thoracic electrical bioimpedance (BioZ impedance cardiograph model BZ-4110-221). Systemic vascular resistance (SVR) calculated as (MAP – central venous pressure)/cardiac output, with default central venous pressure = 6 mmHg. Outcomes: Primary outcome was SVR changes across drink conditions during stress protocols. Secondary outcomes included MAP, cardiac output, and heart rate responses. Data handling and statistics: Hemodynamic data averaged by predefined periods: pre-drink (0–60 min), baseline pre–mental stress (60–85 min), mental stress (90–120 min), baseline pre-CPT (125–145 min), and CPT (150–153 min). Data distribution assessed via Shapiro–Wilk; homoscedasticity via Bartlett. Non-normal data (SVR pre-drink and baseline pre-CPT) transformed using Box–Cox. Mixed-model analyses assessed effects of stress (mental stress or CPT), drink condition, and stress × condition interactions, with condition as fixed factor and participant-specific intercepts and slopes as random effects. Significance threshold P < 0.05. Sample size (n = 12) provided 90% power (α = 0.05) to detect a 6 ± 4% difference in postprandial SVR after sucrose versus water, based on prior data. Analyses performed in R 3.3.1. Data reported as mean ± SD.

- Baseline comparability: No significant differences among conditions in body weight or pre-drink hemodynamic parameters. Table 1 (mean ± SD): Body weight ~59.2–59.3 kg (P = 0.678); resting SVR 1135–1208 U (P = 0.505); resting MAP 65–66 mmHg (P = 0.322); resting cardiac output 4.7–5.0 L/min (P = 0.416); resting heart rate 66–69 bpm (P = 0.223). - Mental stress (30 min): During baseline pre–mental stress (60–85 min), parameters similar between conditions (SVR P = 0.422; MAP P = 0.185; CO P = 0.805; HR P = 0.932). Mental stress effects: SVR unchanged (Stress P = 0.437); MAP increased (P < 0.001); cardiac output increased (P < 0.001); heart rate increased (P < 0.001). No stress × condition interactions for any parameter (all P > 0.05), indicating no differences among water, sucrose, and NNS. - Cold pressor test (3 min): During baseline pre-CPT (125–145 min), parameters similar between conditions (SVR P = 0.202; MAP P = 0.448; CO P = 0.235; HR P = 0.230). CPT effects: SVR increased (Stress P = 0.045 or <0.001 per figure; text indicates significant); MAP increased (P < 0.001); heart rate increased (P < 0.001); cardiac output unchanged (P = 0.252). No stress × condition interactions (all P > 0.05). - Overall: Repeated small boluses of sucrose- or NNS-sweetened drinks did not alter hemodynamic responses to either mental stress or CPT compared with water.

Both stress tasks elicited expected hemodynamic patterns: mental stress increased heart rate, MAP, and cardiac output with no change in SVR, consistent with β-adrenergic activation and nitric oxide–mediated vasodilation; CPT increased SVR, MAP, and heart rate without changing cardiac output, consistent with α-adrenergic vasoconstriction. Across conditions, sweeteners—whether caloric (sucrose) or non-caloric (NNS)—did not modify SVR, MAP, cardiac output, or heart rate responses, suggesting no modulation of sympathetic-driven hemodynamic responses by these beverages under the tested protocol. This contrasts with pediatric findings where intraoral sucrose reduces pain indicators and heart rate, implying taste-mediated analgesia; developmental differences in sweet taste sensitivity and central processing may underlie discrepancies between infants and adults. Practically, limited, repeated exposure to sucrose or NNS during acute standardized stress does not appear to blunt hemodynamic stress reactivity; however, this does not negate other pathways by which stress may influence eating behavior or long-term cardiometabolic risk.

In healthy young women, consuming drinks sweetened with sucrose or non-nutritive sweeteners does not alter hemodynamic responses to standardized acute stress (mental stress and cold pressor test) compared with water. Future research should assess direct measures of sympathetic and HPA axis activity, evaluate different doses/routes or purely oral (non-swallowed) stimulation, and test broader populations (including men and individuals with obesity or cardiometabolic risk) to determine generalizability and mechanistic pathways.

- Autonomic and endocrine measures: No direct assessment of sympathetic nervous system activity (e.g., microneurography, norepinephrine spillover) or HPA axis hormones. - Ingestion vs. taste-only: Participants swallowed the test drinks after mouth rinse; potential insulin-mediated vasodilatory effects from carbohydrate metabolism cannot be excluded for sucrose. - Procedural factors: Possible unintentional Valsalva maneuver during CPT could have influenced hemodynamic responses. - Population: Young, healthy women only; findings may not generalize to men or individuals with obesity, who can exhibit different hemodynamic stress responses.