Increased plasma levels of non-sugar sweeteners in patients with symptomatic carotid atherosclerosis

Abstract

Purpose. The non-sugar sweeteners acesulfame K and saccharin are considered safe, but there is conflicting evidence on their effects on cardiovascular health. Materials and methods. In this explorative pilot study, we measured plasma levels of acesulfame K and saccharin in 15 patients with symptomatic carotid atherosclerosis, 18 asymptomatic patients and 15 control subjects. Fecal microbiota and short-chain fatty acids were analyzed. Dietary and medical history was assessed. Results. Symptomatic patients had higher levels of acesulfame K and saccharin compared to controls. Acesulfame K was associated with increased leukocyte count. Saccharin was associated with more severe carotid stenosis, as well as lower fecal butyric acid.

Keywords:

• Non-sugar sweeteners
• acesulfame K
• saccharin
• carotid atherosclerosis
• ischemic stroke
• microbiota

Introduction

Health effects of non-sugar sweeteners (NSSs) are conflicting, including the associations to type 2 diabetes mellitus and overweight [1]. Some large observational studies have shown artificial sweetened beverages (ASB) to be associated with stroke, cardiovascular events and all-cause mortality, whereas others showed an association between coronary heart disease (CHD) and sweetened beverages, but not ASB [2]. A current study showed an association between cerebrovascular events and the NSS aspartame [3]. In total, however, the evidence of possible links between intake of NSSs and carotid atherosclerosis and related stroke is scarce.

The NSSs acesulfame K and saccharin are approved for dietary consumption and current evidence supports safety [1]. Acesulfame K [4] is almost completely absorbed and excreted within 24 h in the kidneys [5]. For saccharin [4], up to 95% is absorbed and excreted in urine and the rest in feces, with rapid elimination during the first 10 h, followed by a slower eliminated fraction [5] that may affect the gut microbiota [6].

Atherosclerosis is a chronic inflammatory disease, potentially linked to disturbed gut microbiota [7]. One could hypothesize a potential role of NSS-disrupted gut microbiota in eliciting inflammation and plaque instability. The main aim of this pilot study was to investigate if there is an association between NSSs and symptomatic carotid atherosclerosis. Plasma levels of NSSs were measured in patients with symptomatic and asymptomatic carotid atherosclerosis and in healthy control subjects without carotid atherosclerosis. Levels were correlated to cardiovascular risk factors, dietary intake of sugars and snacks, and gut microbiota composition and their derived metabolites short-chain fatty acids (SCFAs).

Material and methods

Please find this section in Supplementary file 1.

Results

Demographic, clinical and biochemical characteristics

Fifteen patients with symptomatic carotid atherosclerosis (ipsilateral stroke or TIA within the last 60 days, median 7 days, 1–60 days), 18 patients with asymptomatic carotid atherosclerosis and 15 healthy control subjects were included (Table 1). Both symptomatic and asymptomatic patients had a more profound cardiometabolic risk profile and used more antibiotics as compared with healthy controls. Importantly, however, except for HDL-cholesterol and waist-hip ratio there were no significant differences in demographic, clinical, dietary or biochemical characteristics between symptomatic and asymptomatic patients.

Table 1. Demographic, clinical and biochemical characteristics of the study group.

	Symptomatic carotid atherosclerosis N = 15	Asymptomatic carotid atherosclerosis N = 18	Healthy controls N = 15	p-value Symptomatic vs. healthy control	p-value Asymptomatic vs. healthy control	p-value Symptomatic vs. asymptomatic
Age (years)^a	73.3 (6.9)	71.6 (6.3)	70.9 (2.7)	0.115	0.704	0.464
Male sex	60 (9)	33.3 (6)	46.7 (7)	0.464	0.435	0.126
C-reactive protein, (mg/L)^a	4.1 (5.3)	3.3 (6.2)	1.3 (1.1)	0.067	0.197	0.741
Leukocyte count, (10^9/L)^a	8.7 (1.7)	7.4 (2.0)	5.1 (1.2)	<0.001	<0.001	0.066
HbA1c (%)^a	6.0 (1.2)	5.9 (0.86)	5.3 (0.26)	0.072	0.022	0.818
LDL, (mmol/L)^a	2.1 (0.81)	2.1 (0.80)	3.4 (1.0)	<0.001	<0.001	0.927
HDL, (mmol/L)^a	1.1 (0.40)	1.5 (0.37)	2.0 (0.57)	<0.001	0.009	0.011
Creatinine level, (mmol/L)^a	84.4 (21.1)	92.9 (26.9)	77.5 (7.8)	0.256	0.056	0.333
Body Mass Index (kg/m^2)^a	26.7 (3.7)	25.8 (4.3)	24.4 (3.5)	0.045	0.256	0.556
Waist-hip ratio (cm/cm)^a	1.0 (0.07)	0.94 (0.07)	0.91 (0.06)	<0.001	0.246	0.011
Hypertension	86.7 (13)	83.3 (15)	20 (3)	<0.001	<0.001	0.790
Type 2 Diabetes mellitus	40 (6)	16.7 (3)	0 (0)	0.006	0.097	0.134
Statin treatment	86.7 (13)	72.2 (13)	0 (0)	<0.001	<0.001	0.312
Antibiotics during last 3 months	26.7 (4)	27.8 (5)	0 (0)	0.032	0.027	0.943
Snacks per week^a	1.9 (0.92)	1.5 (0.79)	1.8 (0.83)	0.772	0.366	0.225
High-degree stenosis (>70%)	86.7 (13)	88.9 (16)	0 (0)			0.803
Low-echogenic or heterogenic plaque	80 (12)	33.3 (6)	0 (0)			0.007
Acesulfame K (Peak Area Data)	4.58x10^7 (0.00942x10^7– 22.3x10^7)	0.238x10^7 (0.0118x10^7– 3.82x10^7)	0.138x10^7 (0.00683x10^7– 0.201x10^7)	0.022	1.000	0.040
Saccharin (Peak Area Data)	1.93x10^6 (0.0453x10^6– 15.2x10^6)	0.393x10^6 (0.0256x10^6– 2.34x10^6)	0.0610x10^6 (0.0474x10^6– 0.115x10^6)	0.014	0.303	0.649

Values are given as % (n).

^amean (SD) or median (min–max).

The degree of carotid stenosis was similar in symptomatic and asymptomatic patients, but symptomatic patients had a higher number of plaques classified as low-echogenic/heterogenic compared to asymptomatic patients.

Plasma levels of acesulfame K and saccharin in symptomatic patients, asymptomatic patients and controls

In the symptomatic group, 93.3% (14 of 15) had detectable levels of one or both NSSs, compared to 77.8% (14 of 18) in the asymptomatic group and 46.7% (7 of 15) in the control group (p = 0.005 and p = 0.066 versus healthy controls and asymptomatic patients, respectively, Supplemental Table 1). Symptomatic patients had significantly higher plasma levels of acesulfame K and saccharin compared to healthy controls without atherosclerosis and for acesulfame K, also compared with asymptomatic patients (Figure 1). Regression analysis between NSS levels and groups confirmed a strong relationship (p < 0.001).

Figure 1. The violin plots show plasma levels of acesulfame K (left) and saccharin in controls, asymptomatic and symptomatic patients.

Associations of acesulfame K and saccharin to cerebrovascular risk factors

For the total population, a higher level of acesulfame K was associated with a higher leukocyte count. High saccharin level was associated to an increased degree of carotid stenosis, higher weekly consumption of snacks and to statin treatment (Supplementary Table 2).

Associations of saccharin to microbiota

Higher saccharin level was associated to decreased concentration of the SCFA butyric acid in feces and to decreased abundance of a major butyric acid producer; Faecalibacterium (Supplemental table 2), but not for other typical butyric acid producers, nor for other SCFAs. We found no associations to TMAO or TMAO-related metabolites or TMAO producers (data not shown).

Association of acesulfame K and saccharin to potential confounders

We found no differences for acesulfame K or saccharin between the two sexes, between diabetics versus non-diabetics, or between the users versus non-users of antibiotics. Acesulfame K and saccharin levels were not associated with neither creatinine level, age, nor BMI (Supplemental Table 3).

Discussion

The main findings were significantly higher plasma levels of acesulfame K and saccharin in patients with symptomatic carotid atherosclerosis compared to healthy controls and for acesulfame K, also as compared with asymptomatic patients. Acesulfame K level was higher with increased leukocyte counts and, saccharin with an increased degree of stenosis, intake of snacks and statin treatment, as well as decreased concentration of fecal butyric acid.

Previously, an association between cerebrovascular events (stroke or TIA) and aspartame was reported [3]. As aspartame is broken down to other compounds, we lack such data in our population. While in the same cohort [3], higher consumption of acesulfame K was associated with CHD events, we herein show an association between higher plasma level of acesulfame K and symptomatic carotid atherosclerosis.

As the harmful effects of sugars are well recognized as risk factors for cardiovascular disease (CVD), and NSSs are considered safe, patients with carotid atherosclerosis and other risk factors for stroke might be likely to consume more NSSs. In fact, higher consumers of artificial sweeteners were reported to have increased CVD risk [3].Thus, whereas the association of NSSs to higher leukocyte count (acesulfame K) and to increased degree of carotid stenosis (saccharin) as well as to plaque symptomatology (acesulfame K and saccharin) may indicate a potential role of these NSSs in plaque destabilization and inflammation, as also have been suggested in some preclinical studies [8], we cannot exclude that having atherosclerosis could lead to increase consumption of artificial sweeteners instead of vice versa.

We found associations to decreased total cholesterol levels and statins for acesulfame K and saccharin, respectively. According to Lin [9] supplement of acesulfame K to a high-cholesterol diet in mice, worsened dyslipidemia, and increased aorta atherosclerosis. Our finding is probably caused by the widespread use of statins in this group.

We found increased level of saccharin to be associated to decreased concentration of fecal butyric acid and one of its main producers, Faecalibacterium. Although other human studies have showed conflicting results [10,11], our findings may suggest that the link between saccharin and plaque destabilization could involve altered microbiota composition. We found no correlations for acesulfame K to these parameters as the amount reaching the gut is negligible [4].

The main limitation of this cross-sectional pilot study is the sample size which is too small to make any firm conclusion. The lack of registration of NSSs, especially in the last 24 h, but also last intake and habits over time are also limitations. Given the design and sample size, data were presented without adjustments which is also an important limitation. In addition, both patient groups differed from the healthy controls in regard to traditional risk factors for stroke. Moreover, correlations do not prove any causal relationship and a reverse causality bias exists as higher plasma levels of NSSs could rather be due to a stricter diet due to more severe atherosclerosis. The strengths were the fasting samples, the well-clinically defined group, including carotid ultrasound examination also of healthy controls, detailed data on general characteristics including diets and microbiota data.

Our findings may potentially provide new insights to the interplay between diet, microbiota, inflammation, and atherosclerosis including plaque progression. However, this explorative pilot study needs to be confirmed in larger studies that also include validation cohorts, investigating links between NSSs and plaque instability.

Author contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Acknowledgments

We would like to thank Gunn-Helen Malmstrøm and Jennifer T. Fiennes for analysis of SCFAs, as well as, Johannes Hov, Beate Vestad and Kristian Holm for 16S sequencing of feces.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Due to data protection regulations in Norway and lack of consent, the data are not deposited in public repositories. However, data will be available upon request to the corresponding author, pending a material and data transfer agreement and an amendment to the Regional committee for medical and health research ethics.

Additional information

Funding

The study was funded by grants from the South Eastern Norway Regional Health Authority.