- Email: [email protected]
Repeated dose study of sucralose tolerance in human subjects
Food and Chemical Toxicology 38 (Suppl. 2) (2000) S123±S129
www.elsevier.com/locate/foodchemtox
Repeated Dose Study of Sucralose Tolerance in Human Subjects I. McLEAN BAIRD1, N. W. SHEPHARD2, R. J. MERRITT3* and G. HILDICK-SMITH4 Pine Court, Fairbourne, Cobham, Surrey KT11 2BT, UK, 2Medical Science Research, Brambles, Granborough, Bucks, MK18 3NT, UK, 3McNeil Specialty Products Company, 501 George St, New Brunswick, NJ 08903, USA and 4Johnson & Johnson, World Tower, One Plaza, New Brunswick, NJ 08903, USA 1
AbstractÐTwo tolerance studies were conducted in healthy human adult volunteers. The ®rst study was an ascending dose study conducted in eight subjects, in which sucralose was administered at doses of 1, 2.5, 5 and 10 mg/kg at 48-hour intervals and followed by daily dosing at 2 mg/kg for 3 days and 5 mg/kg for 4 days. In the second study, subjects consumed either sucralose (n = 77) or fructose (50 g/day) (n = 31) twice daily in single blind fashion. Sucralose dosage levels were 125 mg/day for weeks 1±3, 250 mg/day during weeks 4±7, and 500 mg/day during weeks 8±12. No adverse experiences or clinically detectable eects were attributable to sucralose in either study. Similarly, haematology, serum biochemistry, urinalysis and EKG tracings were unaected by sucralose administration. In the 13-week study, serial slit lamp ophthalmologic examination performed in a random subset of the study groups revealed no changes. Fasting and 2-hour post-dosing blood sucralose concentrations obtained daily during week 12 of the study revealed no rising trend for blood sucralose. Sucralose was well tolerated by human volunteers in single doses up to 10 mg/kg/day and repeated doses increasing to 5 mg/kg/day for 13 weeks. Based on these studies and the extensive animal safety database, there is no indication that adverse eects on human health would occur from frequent or long-term exposure to sucralose at the maximum anticipated levels of intake. # 2000 Elsevier Science Ltd. All rights reserved Keywords: sucralose; arti®cial sweetener; repeated dose study; man. Abbreviations: ADI = acceptable daily intake; EDI = estimated daily intake; HNEL = highest-noeect level.
INTRODUCTION
Evaluation of a potential new food additive such as sucralose requires extensive testing in both animals and man (Borzelleca, 1992). Studies on the fate of the compound in man are essential for the precise interpretation of animal toxicity data and are important in the human safety evaluation process. Absorption, distribution, metabolism and elimination (ADME) studies in several animal species and man have shown sucralose to exhibit limited absorption, rapid urinary excretion, faecal excretion of unchanged sucralose and minimal metabolism of the absorbed material with no bioaccumulation (Goldsmith and Grice, 2000; John et al., 2000; Roberts et al., 2000; Sims et al., 2000; Wood et al., *Present address: Ross Labs, 151 South Parkview, Columbus, OH 43209, USA.
2000). Following an oral dose of sucralose, approximately 10±35% is absorbed in those laboratory animals tested and about 15% in humans. Unlike sucrose, sucralose is not metabolized or otherwise used for energy in mammalian systems (Kille et al., 2000; McNeil, 1987). Metabolism is limited to glucuronidation products which are excreted in the urine and account for about 2±3% of the dose in humans. Overall, the fate of sucralose in humans was found to be similar to that found in rats, dogs and mice, suggesting that these species are good models for assessing the safety of sucralose in man. To further complement the safety evaluation process of sucralose, two supplementary human tolerance studies were conducted. This paper reviews the results of two ascending dose human tolerance studies on sucralose, the ®rst lasting 17 days and the second 13 wk.
0278-6915/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved. Printed in Great Britain PII S0278-6915(00)00035-1
S124
I. McLean Baird et al. MATERIAL AND METHODS
Methods Study 1: 17-day ascending dose study Clinical protocol. The ®rst ascending dose study was divided into two separate phases. Phase 1 involved single ascending doses on days 1, 3, 5, 7 and 9 at doses up to 10 mg/kg, while Phase 2 involved 3 days of continuous doses of 2 mg/kg followed by 4 days of continuous doses of 5 mg/kg up to day 17. Eight normal subjects (four male and four female) aged between 18 and 48 yr (mean 32 yr) and weighing between 57 and 80 kg (mean 70 kg), were recruited into the study at Princess Margaret Hospital, Windsor, UK. An Ethical Committee approved this study, and each subject gave written, informed consent before the study commenced. Volunteers, many of whom were scienti®cally oriented, fully understood the trial. The inclusion criteria for participation included good physical health, normal electrocardiogram (ECG, CardioMini, Fukuda Denshi Co., Japan), and normal values for the haematologic, biochemical and urinary parameters monitored. Study subjects were inpatients on metabolic wards at Princess Margaret Hospital during the course of this investigation. Sucralose was supplied by Tate & Lyle Group Research & Development as an aqueous solution (5 mg/ml). The dosage for each subject (mg/kg) was determined and independently veri®ed before administration on each occasion. The dose was supplied with 100 ml water. Throughout the study, subjects were authorized to fast from 21.00 hr the night prior to dosing. At 1 and 2 hr following each dosing, all subjects received 200 ml water. There were no other restrictions on food intake apart from the consumption of alcohol, which was limited to the equivalent of one glass of wine each day. Phase 1Ðsingle ascending doses Dosing was 0 mg/kg on day 1, 1.0 mg/kg on day 3, 2.5 mg/kg on day 5, 5 mg/kg on day 7 and 10 mg/kg on day 9. No doses were given on days 2, 4, 6 and 8. Each dose was given at approximately 09.00 hr followed by physical observations, including vital signs, at times 0 (dosing), 2, 4, 8, 12, 16, 20 and 24 hr post-dosing. A blood sample was taken for blood insulin assay 0.5 hr after each dose (Randle et al., 1963). 24 hr after dosing, ECGs were repeated and blood samples were taken for haematological (haemoglobin, RBC, PCV, MCV, MCH, MCHC, WBC, ESR and platelets) and biochemical (thyroxin, total protein, albumin, globulin, calcium, phosphorus, urea, uric acid, creatinine, total bilirubin, alkaline phosphatase, SGOT, SGPT, gamma GT, glucose, cholesterol and triglyceride) analyses.
Urine was collected daily in 24-hr aliquots during the 9-day period and was examined for pH, ketones, blood, glucose, bilirubin, protein, urobilinogen, speci®c gravity, white cells, red cells, squamous cells, crystals and organisms. After the 10 mg/kg dose, serial blood samples were taken with heparinized tubes at times 0, 0.5, 1, 2, 3, 4, 6, 8 and 12 hr post-dosing, stored at 48C and then analysed for sucralose. Phase 2Ðcontinuous dosage On day 11, when all the biochemical screening test results were available (within 48 hr), the second phase commenced. The same subjects as in phase 1 received a single dose of sucralose each morning for 7 days. On days 11, 12 and 13, a dose of 2 mg/kg was administered, and on days 14, 15, 16 and 17 a dose of 5 mg/kg was given. Again, after an overnight fast, each dose was given with 100 ml water. Physical observations were made at times 0, 2, 4, 8, 12, 16 and 24 hr post-dosing, and ECGs were done daily. Blood samples for insulin assays were taken 0.5 hr after each sucralose dose, and blood samples were taken on days 14, 18 and 25 for haematological and biochemical testing. Urine was collected in 24-hr aliquots and examined daily. On day 25 of the study, blood insulin levels were measured 0.5 hr after administration of 50 g standard sucrose, as a 10% solution. Haematology, urinalysis and biochemical analyses were performed throughout by J. S. Pathology Laboratory using standard clinical practice methods. The normal values were taken from established values. A trend analysis was performed on the obtained variables. ECG tracings were scanned by two physicians independently for any abnormality. Blood sucralose analysis Blood sucralose assays were performed by Tate & Lyle Research & Development, Whiteknights, Reading, UK. Blood (5 ml) was centrifuged at 5000 rpm for 10 min and the plasma retained. The cells were resuspended in isotonic saline (5 ml) and centrifuged, and the supernatant was added to the plasma. The solution was injected, using a 10 ml Luer syringe, into a Sep-Pak cartridge (Waters Associates, UK) previously activated with methanol (2 ml) and water (1 ml). The column was washed with water (5 ml) and the washings discarded. The retained material was then eluted with methanol (2 ml) into a silylation tube. The solvent was removed in a stream of air, and a solution (250 ml) of the internal standard, 6,1',6'-trichloro-6,1',6'-trideoxy-sucrose, in pyridine (Fisons, UK), was added followed by 250 ml trimethylsilyllimidazole (Sigma, UK). The concentration of the internal standard used depended on the expected level of sucralose in the sample. The vial was sealed and heated at 708C for 30 min. A range of external sucralose standards
Dose study of sucralose in humans
to cover the predicted level of sucralose in the silylated samples was prepared. The samples and standards were analysed by gas chromatography±mass spectrometry (Kratos MS 80 Rf mass spectrometer linked to a Carlo Erba capillary GC) using the following conditions: column, 2 m glass 3% OV-1; injection temperature, 2808C, helium ¯ow rate, 40 ml/min; temperature program, 2758C for 1 min increasing to 3008C at 88C/min and holding for 10 min; injection volume, 1 ml. Jet separator temperature, 2708C; source temperature, 2508C; single ion monitoring, m/z 308. Duplicate injections of each sample and standard were made. The average recovery of sucralose added to control blood has been shown to be 68 with a variation of 10%. The response factor of the external sucralose standard was obtained using the expression: Response factor Rf
S125
aqueous solution (pH 7.67) was stored at 308C for approximately 1 yr (McNeil, 1987). The control group received 5 g fructose (Finn Sugar Company) daily throughout the study. Each sweetener was administered twice daily in equal doses at approximately 10.00 hr and 16.00 hr. Each subject was provided weekly with a supply of sweetener and a diary card to complete. Each dose was recorded and signed by an independent witness to con®rm compliance. Blood and urine analyses were carried out at the end of wk 3, 7 and 13. A repeat physical examination including an ECG was conducted at the end of the study. Blood and urine samples were analysed at J.S. Pathology Services, London, using standard automated analytical procedures utilizing the Technician SMA II analyser for chemistries and
Peak area sucralose ppm internal standard in silylated soln Peak area internal standard ppm sucralose in silylated soln
The concentration of sucralose in blood is then given by: Peak area sucralose ppm internal standard in silylated soln 1 100 Peak area internal standard Rf concn factor 68
Study 2: 13-wk repeat dose study Clinical protocol. The study was a single-blind randomized controlled study conducted over a period of 13 wk at two sites in the Greater London area. The objective of the study was to compare the eects of sucralose in human volunteers with that of fructose administered daily. The study was approved by a local Ethical Committee, and written informed consent was obtained from each subject. 118 volunteers ful®lling the inclusion and exclusion criteria as follows were recruited into the study. Eligibility criteria included a normal physical examination, a 12-lead electrocardiogram, a complete blood count including prothrombin time and partial thromboplastin time, normal serum biochemistry and urinalysis. Subjects with a history of serious illness, metabolic disease, including diabetes, chronic active gastrointestinal disease, a history of drug or alcohol abuse or reported sensitivity to sugars or sweeteners were excluded. The recruited subjects were randomly assigned to sucralose or fructose (control sweetener) in a 2:1 ratio. The sucralose, obtained from Tate & Lyle, Whiteknights, Reading, as an aqueous solution (35 g/litre), was administered daily for 90 days at doses of 125, 250 and 500 mg/day during wk 1±3, 4±7 and 8±12, respectively. Fresh aqueous solutions were prepared on six separate occasions and forwarded to the lab. Aqueous solutions of sucralose have been found to be stable almost inde®nitely. No detectable loss of sucralose occurred when an
the Technician H6000 for haematology. Urine samples were analysed using standard clinical laboratory procedures. Prothrombin time was performed by the method of Quick (1966) and partial thromboplastin time by the method of Procter and Rapaport (1961). ECG recordings were analysed for heart rate, PR interval, QRS duration and QTc interval using Bazett's formula. In addition, sucralose blood level determinations were performed on 10 of the subjects (®ve males and ®ve females) receiving 250 mg sucralose twice daily during wk 12. Blood samples (5 ml) were obtained on 5 consecutive days, both immediately before and 2 hr after the morning sucralose dose. Blood was analysed for sucralose as described above in the 17-day ascending dose study with modi®cation for multiple ion monitoring at M/Z 308 and 310. Also a detailed ophthalmological examination, including slit-lamp examination using a Hamblin SL 202, was carried out before and after the study in a randomly selected group of 24 subjects of whom 18 were taking sucralose (15 male and three female) and six fructose (®ve male and one female). RESULTS
17-Day ascending dose study All eight subjects who were entered completed the 17-day ascending dose study. During the study there were no abnormal variations in body tempera-
S126
I. McLean Baird et al.
Fig. 1. Sucralose blood levels after 10 mg/kg dose. On day 9 of study one, a single dose of 10 mg/kg sucralose was administered orally in the fasting state to the eight subjects. Blood was obtained for sucralose assay at times 0.5, 1, 2, 3, 4, 6, 8 and 24 hr post-dosing. Peak blood concentrations were detected at 1 hr; sucralose was barely detectable at 24 hr. (N = 8).
ture, pulse rate, blood pressure or respiratory rate following either single ascending or continuous doses. In addition, no adverse comments attributed to sucralose were made. The general physical examinations at the end of each phase of the study were normal. During the study, mean haematological values remained stable. There were occasional results outside the normal range for the laboratory, but no trends were apparent and these variations were, therefore, not clinically meaningful. The 17 biochemical values monitored also remained stable throughout the study. No treatment eect or trends were evident, and values remained in the normal range. Two females taking oral contraceptives had marginal elevations in the pre-study thyroxin values, which were unaected by sucralose administration. All urinalysis results taken in Phase 1 and 2 were found to be normal. No eect on urine volume, urine chemistries or sediment was observed. Similarly, 12- lead ECG readings were unaected by sucralose ingestion over the 17-day period. The absorption study of sucralose indicated a general pattern of peak blood levels 1 hr post-dosing which then decreased thereafter. Levels were
barely detectable 24 hr post-dosing (Fig. 1). Sucralose had no eect on fasting serum insulin levels as no variation from the fasting range was detected for each individual (McNeil, 1987). 13-Week repeat dose study A total of 118 subjects were recruited, of which 108 completed the study. Of these, 77 received sucralose (47 males, 30 females) and 31 received fructose (17 males, 14 females) (Table 1). 10 subjects withdrew from the study, three of these taking sucralose and seven taking fructose. None of the withdrawals in the sucralose group was due to adverse experiences. Withdrawals in the sucralose group were due to a loss of appetite during wk 2, a concern about consuming low-calorie sweeteners and non-compliance with the dosing schedule. One subject, taking fructose, exhibited an apparent sensitivity to the compound, while most other dropouts in this group were related to the high sweetness of the test material (Table 2). Overall, a common complaint throughout the study was the intense sweetness of the test beverages. The vast majority of subjects were able to adjust to this, but most indicated it was far in excess of their normal sweetener intake.
Table 1. Composition of sucralose and control groups
Sexes Mean age in yr (SD) Mean weight in kg (SD)
Male Female Total Male Female Overall Male Female Overall
Sucralose group
Control group
47 30 77 33.7 (9.73) 35.9 (11.71) 34.6 (10.52) 78.7 (10.70) 60.2 (6.69) 71.5 (12.97)
17 14 31 34.0 (10.76) 33.7 (11.72) 33.9 (11.02) 73.1 (9.33) 64.6 (7.00) 69.3 (9.28)
Dose study of sucralose in humans
S127
Table 2. Withdrawals from study Sucralose (three subjects) Male (No. 58) Male (No. 110)
Wk 2 Wk 5
Female (No. 32)
Wk 8
Fructose (seven subjects) Male (No. 113) Male (No. 24) Male (No. 98)
Day 2 Day 3 Wk 2
Male (No. 107) Female (No. 12) Female (No. 38) Female (No. 16)
Wk Wk Wk Wk
2 2 8 5
``Loss of appetite'' ``Did not like the idea of taking an arti®cial sweetener for any length of time. OtherwiseÐno complaints or side-eects Not taking sucient ``doses'', therefore, considered an unreliable subject and she was asked to withdraw. No side-eects ``Too sweet'' ``Too sweet'' Skin reactionÐcleared on cessation of fructose intake and reappeared when subject challenged again with two ``doses'' of fructose Persistent sore throat which resolved when fructose intake stopped ``Too sweet'' ``Too sweet'' Decided (reluctantly) to stop as she had to have a minor operation unrelated to the study
There were no changes between initial and ®nal physical examinations in those completing the study, and there was no evidence of weight loss. Mean BP recordings before and after 13 wk showed no changes. There were no signi®cant changes observed in any of the haematologic parameters measured. There was an upward trend in the prothrombin time measurements in both the sucralose and control groups, but there was no statistical dierence between the two groups. There were no signi®cant changes in any of the biochemical parameters or urinalysis. Tracings from the 12-lead ECGs were compared before and after the 13-wk period. Heart rates, PR, QRS and QTc intervals showed no signi®cant changes after sucralose administration. Of the 24 individuals examined by an ophthalmologic consultant, 18 were taking sucralose (15 male and three female) and six fructose (®ve male and one female). Ophthalmoscopic and slit-lamp examination revealed no abnormalities attributable to either sweetener. The daily blood samples taken from 10 subjects each morning during wk 12 immediately before and 2 hr following the administration of the morning dose of sucralose (250 mg) revealed no trend towards increasing fasting or post-dose concentrations of sucralose over time (Table 3). DISCUSSION
The levels of sucralose used in these two studies were in excess of expected human intakes. To understand how much sucralose can be safely con-
sumed by an individual, it is useful to consider three related concepts: highest-no-adverse-eect level (HNEL), acceptable daily intake (ADI) and estimated daily intake (EDI). In order to assess the safety of a new food additive, long-term animal studies are conducted to determine the highest daily dose that can be given to test animals without adverse eects. In the sucralose safety studies, no adverse eects were seen in rodents given up to approximately 1500 mg/kg/day for 104 wk (Goldsmith et al., 2000; Mann et al., 2000a,b). This amount, therefore, was considered the HNEL of sucralose. An ADI represents the amount of a food additive that, if consumed daily throughout a person's life, would be considered safe by a wide margin. This amount is established by dividing the product's HNEL by a large safety factor, usually 100. Dividing the HNEL of sucralose (1500 mg/kg/day) by this 100-fold safety factor yields an ADI of 15 mg/kg/day. The Joint Expert Committee on Food Additives (JECFA) has allocated sucralose an ADI of 15 mg/kg/day (WHO, 1991). Also, the Australian National Food Authority has similarly approved an ADI of 15 mg/kg/day for sucralose. The EDI of a food additive is the amount that is expected to be consumed daily by an individual. To further ensure a product's safety, this projected consumption level should average less than the acceptable daily intake level established for the product. To estimate the EDI of sucralose, an independent research organization, Market Research Corporation of America (MRCA), asked members of more than 2000 households to keep diaries listing all the foods and beverages consumed over a 2-wk
Table 3. Mean and standard deviation (SD) of daily concentration of sucralose in whole-blood during wk 12 of study (ng/ml) while subjects received 5 mg/kg/day Day Time after dosing 0 hr 2 hr
1
2
3
4
5
30 (20) 150 (60)
40 (30) 190 (120)
30 (20) 100 (60)
20 (3) 120 (70)
50 (50) 180 (90)
S128
I. McLean Baird et al. Table 4. EDI, ADI and HNEL of sucralose for individuals of dierent age groups EDI* Mean
90th percentile
Age
mg
mg/kg/day
mg
mg/kg/day
0±23 months 2±5 yr 6±12 yr 13±17 yr 18±44 yr 45±64 yr 65+ yr All ages
22 46 58 57 51 55 51 52
2.0 2.8 2.0 1.0 0.8 0.8 0.8 1.1
49 77 100 110 100 100 93 98
4.3 5.1 3.9 1.9 1.5 1.5 1.5 2.3
ADI** (mg/kg/day)
HNEL*** (mg/kg/day)
15
1500
*EDI = estimated daily intake. **ADI = acceptable daily intake as allocated by JECFA, June 1990. ***HNEL = highest-no-adverseeect level.
period. In their analysis, the researchers assumed that sucralose would replace virtually all sweeteners including sugar in all the food and beverage categories in which it would be approved for use. As it is extremely unlikely that this would ever be the case, the EDI provides an excessive estimate of how much sucralose might be consumed. Despite this very exaggerated set of assumptions, the mean EDI of sucralose for people of all ages is 1.1 mg/kg/day (McNeil, 1987). The EDI was also estimated for individuals of all ages who might consume sucralose-sweetened foods and beverages in great amounts on a continuous basis throughout their lifetime (consumption at the 90th percentile). For these individuals, the EDI is 2.3 mg/kg/day (Table 4). The EDI with its overestimation of consumption has a built-in margin of safety and, consequently, does not remotely approximate a toxic threshold. The overestimation of intake provided by the EDI can be quite large. For example, for another widely used low-calorie sweetener, the EDI is 5 mg/kg, but actual Canadian intake levels are below 1.3 mg/kg (Heybach and Ross, 1989). In terms of human consumption, a 160-lb (73 kg) adult would have to drink 1500 12-oz (360 ml) soft drinks sweetened with sucralose every day to consume an amount comparable to the HNEL. Similarly, a 50-lb (23 kg) child would have to drink 480 12-oz (360 ml) sucralose-sweetened soft drinks daily to reach the HNEL. To reach the ADI of 15 mg/kg/day, a 160-lb (73 kg) adult would have to ingest an amount of sucralose equivalent in sweetness to approximately 1 lb of sugar a day. Even if an individual occasionally consumed sucralose in amounts greater than the ADI, this would pose absolutely no health risk. For instance, if an individual on occasion consumed 10 mg/kg, that level would still be 1/150th the dose which caused no adverse eects in animals. In the 17-day ascending dose study, sucralose doses were administered which were as high as nine times the projected EDI for users of low-calorie
sweeteners. Even given this much higher than expected dietary intake, none of the study subjects reported adverse eects, and their physical, chemical and haematological data were unaected by sucralose ingestion. Sucralose was found to be well tolerated at a single dose of up to 10 mg/kg and then when administered daily in doses of 2 mg/kg for 3 days, then 5 mg/kg for 4 days. Fasting insulin values were also unaected by repeated sucralose ingestion, and the insulin response to sucrose 1 wk after the last dose of sucralose followed the expected pattern. These results indicate that humans, like other mammalian species tested, do not biochemically recognize sucralose as a disaccharide. To assess the eects of chronic sucralose exposure, a single blind sweetener-controlled 13-wk study was undertaken with fructose as the control. During the dosing period, the daily intakes of sucralose varied between 4.8 and 8.0 mg/kg for males and 6.4 and 10.1 mg/kg for females. Many subjects disliked the intense sweet taste of both beverages used to achieve the desired intake level of sucralose by the end of the study, approximately ®ve times the EDI. Several subjects elected to drop out in both groups. However, of those three subjects who dropped out of the sucralose group, none did so due to an adverse experience. During the study, no group dierences in physical, haematological or biochemical data were detected between the sucralose and control groups. Neither were there changes in these parameters over the 13wk period. Similarly, the ECG tracings were unaffected by chronic sucralose ingestion, and ophthalmologic examination remained normal after 13 wk of ingestion in the subjects examined. Consistent with the results from the 17-day study, sucralose was found to be well tolerated in normal volunteers, even when dosed subchronically at high intakes. Analysis of the blood samples taken following the 10 mg/kg single oral dose showed that the peak sucralose concentrations occurred after 1 hr (0.62 0.30 ppm) and decreased over the time period
Dose study of sucralose in humans
with virtually complete elimination after 24 hr (0.02 2 0.02 ppm). These data support the human pharmacokinetic observations which showed that peak plasma sucralose concentrations occurred 1.5 to 3 hr after an oral dose and that negligible accumulation would occur during consumption consistent with the normal pattern of use of a lowcalorie sweetener. (Roberts et al., 2000). The lack of sucralose blood bioaccumulation was shown following the analysis of blood samples after chronic ingestion of 5 mg/kg in wk 12 of the 13-wk study. In this study, there was no indication of rising blood sucralose levels with prolonged high-dose exposure. Both of the human studies show that sucralose is well tolerated in healthy volunteers, even when dosed subchronically at levels greater than would be consumed during normal patterns of use. REFERENCES
Borzelleca J. F. (1992) Macronutrient substitutes: safety evaluation. Regulatory Toxicology and Pharmacology 16, 253±264. Goldsmith L. A. (2000) Acute and subchronic toxicity of sucralose. Food and Chemical Toxicology 38 (Suppl. 2), S53±S69. Goldsmith L. A. and Grice H. C. (2000) SucraloseÐan overview of the toxicity data. Food and Chemical Toxicology 38 (Suppl. 2), S1±S6. Heybach J. P. and Ross C. (1989) Aspartame consumption in a representative sample of canadians. Journal of the Canadian Dietetic Association 50, 166±170. International Food Information Council (1990) Aspartame consumption found safe in new studies. Food Insight, Current Topics in Food Safety and Nutrition. Vol. 6. John B. A., Wood S. G. and Hawkins D. R. (2000) The pharmacokinetics and metabolism of sucralose in the rabbit. Food and Chemical Toxicology 38 (Suppl. 2), S111±S113.
S129
Kille J. W., Ford W. C. L., McAnulty P., Tesh J. M., Ross F. W. and Willoughby C. R. (2000) Sucralose: lack of eects on sperm glycolysis and reproduction in the rat. Food and Chemical Toxicology 38 (Suppl. 2), S19±S29. McNeil Specialty Products Food Additive Petition 7A3987 (1987) (Sucralose). Mann S. W., Yuschak M. M., Amyes S. J. G., Aughton P. and Finn J. P. (2000a) A combined chronic toxicity/ carcinogenicity study of sucralose in Sprague±Dawley rats. Food and Chemical Toxicology 38 (Suppl. 2), S71± S89. Mann S. W., Yuschak M. M., Amyes S. J. G., Aughton P. and Finn J. P. (2000b) A carcinogenicity study of sucralose in the CD-1 mouse. Food and Chemical Toxicology 38 (Suppl. 2), S91±S98. Market Research Corporation of America. (1989) Unpublished report on sucralose intake. Vols I±V. Procter R. R. and Rapaport S. I. (1961) The partial thromboplastic time with kaolin. American Journal of Clinical Pathology 36, 212±219. Quick A. J. (1966) The prothrombin consumption time test. American Journal of Clinical Pathology 45, 105± 109. Randle P. J., Hales C. N., Garland P. B. and Newsholme E. A. (1963) The glucose fatty acid cycle, its role in insulin sensitivity and the metabolic disturbances of diabetes mellitus. Lancet 785±789. Roberts A., Renwick A. G., Sims J. and Snodin D. J. (2000) Sucralose metabolism and pharmacokinetics in man. Food and Chemical Toxicology 38 (Suppl. 2), S31±S41. Sims J., Roberts A., Daniel J. W. and Renwick A. G. (2000) The metabolic fate of sucralose in rats. Food and Chemical Toxicology 38 (Suppl. 2), S115± S121. WHO (1991) Evaluation of Certain Food Additives and Contaminants. Food Additives Series, No. 37, World Health Organization, Geneva. Wood S. G., John B. A. and Hawkins D. R. (2000) The pharmacokinetics and metabolism of sucralose in the dog. Food and Chemical Toxicology 38 (Suppl. 2), S99± S106.
www.elsevier.com/locate/foodchemtox
Repeated Dose Study of Sucralose Tolerance in Human Subjects I. McLEAN BAIRD1, N. W. SHEPHARD2, R. J. MERRITT3* and G. HILDICK-SMITH4 Pine Court, Fairbourne, Cobham, Surrey KT11 2BT, UK, 2Medical Science Research, Brambles, Granborough, Bucks, MK18 3NT, UK, 3McNeil Specialty Products Company, 501 George St, New Brunswick, NJ 08903, USA and 4Johnson & Johnson, World Tower, One Plaza, New Brunswick, NJ 08903, USA 1
AbstractÐTwo tolerance studies were conducted in healthy human adult volunteers. The ®rst study was an ascending dose study conducted in eight subjects, in which sucralose was administered at doses of 1, 2.5, 5 and 10 mg/kg at 48-hour intervals and followed by daily dosing at 2 mg/kg for 3 days and 5 mg/kg for 4 days. In the second study, subjects consumed either sucralose (n = 77) or fructose (50 g/day) (n = 31) twice daily in single blind fashion. Sucralose dosage levels were 125 mg/day for weeks 1±3, 250 mg/day during weeks 4±7, and 500 mg/day during weeks 8±12. No adverse experiences or clinically detectable eects were attributable to sucralose in either study. Similarly, haematology, serum biochemistry, urinalysis and EKG tracings were unaected by sucralose administration. In the 13-week study, serial slit lamp ophthalmologic examination performed in a random subset of the study groups revealed no changes. Fasting and 2-hour post-dosing blood sucralose concentrations obtained daily during week 12 of the study revealed no rising trend for blood sucralose. Sucralose was well tolerated by human volunteers in single doses up to 10 mg/kg/day and repeated doses increasing to 5 mg/kg/day for 13 weeks. Based on these studies and the extensive animal safety database, there is no indication that adverse eects on human health would occur from frequent or long-term exposure to sucralose at the maximum anticipated levels of intake. # 2000 Elsevier Science Ltd. All rights reserved Keywords: sucralose; arti®cial sweetener; repeated dose study; man. Abbreviations: ADI = acceptable daily intake; EDI = estimated daily intake; HNEL = highest-noeect level.
INTRODUCTION
Evaluation of a potential new food additive such as sucralose requires extensive testing in both animals and man (Borzelleca, 1992). Studies on the fate of the compound in man are essential for the precise interpretation of animal toxicity data and are important in the human safety evaluation process. Absorption, distribution, metabolism and elimination (ADME) studies in several animal species and man have shown sucralose to exhibit limited absorption, rapid urinary excretion, faecal excretion of unchanged sucralose and minimal metabolism of the absorbed material with no bioaccumulation (Goldsmith and Grice, 2000; John et al., 2000; Roberts et al., 2000; Sims et al., 2000; Wood et al., *Present address: Ross Labs, 151 South Parkview, Columbus, OH 43209, USA.
2000). Following an oral dose of sucralose, approximately 10±35% is absorbed in those laboratory animals tested and about 15% in humans. Unlike sucrose, sucralose is not metabolized or otherwise used for energy in mammalian systems (Kille et al., 2000; McNeil, 1987). Metabolism is limited to glucuronidation products which are excreted in the urine and account for about 2±3% of the dose in humans. Overall, the fate of sucralose in humans was found to be similar to that found in rats, dogs and mice, suggesting that these species are good models for assessing the safety of sucralose in man. To further complement the safety evaluation process of sucralose, two supplementary human tolerance studies were conducted. This paper reviews the results of two ascending dose human tolerance studies on sucralose, the ®rst lasting 17 days and the second 13 wk.
0278-6915/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved. Printed in Great Britain PII S0278-6915(00)00035-1
S124
I. McLean Baird et al. MATERIAL AND METHODS
Methods Study 1: 17-day ascending dose study Clinical protocol. The ®rst ascending dose study was divided into two separate phases. Phase 1 involved single ascending doses on days 1, 3, 5, 7 and 9 at doses up to 10 mg/kg, while Phase 2 involved 3 days of continuous doses of 2 mg/kg followed by 4 days of continuous doses of 5 mg/kg up to day 17. Eight normal subjects (four male and four female) aged between 18 and 48 yr (mean 32 yr) and weighing between 57 and 80 kg (mean 70 kg), were recruited into the study at Princess Margaret Hospital, Windsor, UK. An Ethical Committee approved this study, and each subject gave written, informed consent before the study commenced. Volunteers, many of whom were scienti®cally oriented, fully understood the trial. The inclusion criteria for participation included good physical health, normal electrocardiogram (ECG, CardioMini, Fukuda Denshi Co., Japan), and normal values for the haematologic, biochemical and urinary parameters monitored. Study subjects were inpatients on metabolic wards at Princess Margaret Hospital during the course of this investigation. Sucralose was supplied by Tate & Lyle Group Research & Development as an aqueous solution (5 mg/ml). The dosage for each subject (mg/kg) was determined and independently veri®ed before administration on each occasion. The dose was supplied with 100 ml water. Throughout the study, subjects were authorized to fast from 21.00 hr the night prior to dosing. At 1 and 2 hr following each dosing, all subjects received 200 ml water. There were no other restrictions on food intake apart from the consumption of alcohol, which was limited to the equivalent of one glass of wine each day. Phase 1Ðsingle ascending doses Dosing was 0 mg/kg on day 1, 1.0 mg/kg on day 3, 2.5 mg/kg on day 5, 5 mg/kg on day 7 and 10 mg/kg on day 9. No doses were given on days 2, 4, 6 and 8. Each dose was given at approximately 09.00 hr followed by physical observations, including vital signs, at times 0 (dosing), 2, 4, 8, 12, 16, 20 and 24 hr post-dosing. A blood sample was taken for blood insulin assay 0.5 hr after each dose (Randle et al., 1963). 24 hr after dosing, ECGs were repeated and blood samples were taken for haematological (haemoglobin, RBC, PCV, MCV, MCH, MCHC, WBC, ESR and platelets) and biochemical (thyroxin, total protein, albumin, globulin, calcium, phosphorus, urea, uric acid, creatinine, total bilirubin, alkaline phosphatase, SGOT, SGPT, gamma GT, glucose, cholesterol and triglyceride) analyses.
Urine was collected daily in 24-hr aliquots during the 9-day period and was examined for pH, ketones, blood, glucose, bilirubin, protein, urobilinogen, speci®c gravity, white cells, red cells, squamous cells, crystals and organisms. After the 10 mg/kg dose, serial blood samples were taken with heparinized tubes at times 0, 0.5, 1, 2, 3, 4, 6, 8 and 12 hr post-dosing, stored at 48C and then analysed for sucralose. Phase 2Ðcontinuous dosage On day 11, when all the biochemical screening test results were available (within 48 hr), the second phase commenced. The same subjects as in phase 1 received a single dose of sucralose each morning for 7 days. On days 11, 12 and 13, a dose of 2 mg/kg was administered, and on days 14, 15, 16 and 17 a dose of 5 mg/kg was given. Again, after an overnight fast, each dose was given with 100 ml water. Physical observations were made at times 0, 2, 4, 8, 12, 16 and 24 hr post-dosing, and ECGs were done daily. Blood samples for insulin assays were taken 0.5 hr after each sucralose dose, and blood samples were taken on days 14, 18 and 25 for haematological and biochemical testing. Urine was collected in 24-hr aliquots and examined daily. On day 25 of the study, blood insulin levels were measured 0.5 hr after administration of 50 g standard sucrose, as a 10% solution. Haematology, urinalysis and biochemical analyses were performed throughout by J. S. Pathology Laboratory using standard clinical practice methods. The normal values were taken from established values. A trend analysis was performed on the obtained variables. ECG tracings were scanned by two physicians independently for any abnormality. Blood sucralose analysis Blood sucralose assays were performed by Tate & Lyle Research & Development, Whiteknights, Reading, UK. Blood (5 ml) was centrifuged at 5000 rpm for 10 min and the plasma retained. The cells were resuspended in isotonic saline (5 ml) and centrifuged, and the supernatant was added to the plasma. The solution was injected, using a 10 ml Luer syringe, into a Sep-Pak cartridge (Waters Associates, UK) previously activated with methanol (2 ml) and water (1 ml). The column was washed with water (5 ml) and the washings discarded. The retained material was then eluted with methanol (2 ml) into a silylation tube. The solvent was removed in a stream of air, and a solution (250 ml) of the internal standard, 6,1',6'-trichloro-6,1',6'-trideoxy-sucrose, in pyridine (Fisons, UK), was added followed by 250 ml trimethylsilyllimidazole (Sigma, UK). The concentration of the internal standard used depended on the expected level of sucralose in the sample. The vial was sealed and heated at 708C for 30 min. A range of external sucralose standards
Dose study of sucralose in humans
to cover the predicted level of sucralose in the silylated samples was prepared. The samples and standards were analysed by gas chromatography±mass spectrometry (Kratos MS 80 Rf mass spectrometer linked to a Carlo Erba capillary GC) using the following conditions: column, 2 m glass 3% OV-1; injection temperature, 2808C, helium ¯ow rate, 40 ml/min; temperature program, 2758C for 1 min increasing to 3008C at 88C/min and holding for 10 min; injection volume, 1 ml. Jet separator temperature, 2708C; source temperature, 2508C; single ion monitoring, m/z 308. Duplicate injections of each sample and standard were made. The average recovery of sucralose added to control blood has been shown to be 68 with a variation of 10%. The response factor of the external sucralose standard was obtained using the expression: Response factor Rf
S125
aqueous solution (pH 7.67) was stored at 308C for approximately 1 yr (McNeil, 1987). The control group received 5 g fructose (Finn Sugar Company) daily throughout the study. Each sweetener was administered twice daily in equal doses at approximately 10.00 hr and 16.00 hr. Each subject was provided weekly with a supply of sweetener and a diary card to complete. Each dose was recorded and signed by an independent witness to con®rm compliance. Blood and urine analyses were carried out at the end of wk 3, 7 and 13. A repeat physical examination including an ECG was conducted at the end of the study. Blood and urine samples were analysed at J.S. Pathology Services, London, using standard automated analytical procedures utilizing the Technician SMA II analyser for chemistries and
Peak area sucralose ppm internal standard in silylated soln Peak area internal standard ppm sucralose in silylated soln
The concentration of sucralose in blood is then given by: Peak area sucralose ppm internal standard in silylated soln 1 100 Peak area internal standard Rf concn factor 68
Study 2: 13-wk repeat dose study Clinical protocol. The study was a single-blind randomized controlled study conducted over a period of 13 wk at two sites in the Greater London area. The objective of the study was to compare the eects of sucralose in human volunteers with that of fructose administered daily. The study was approved by a local Ethical Committee, and written informed consent was obtained from each subject. 118 volunteers ful®lling the inclusion and exclusion criteria as follows were recruited into the study. Eligibility criteria included a normal physical examination, a 12-lead electrocardiogram, a complete blood count including prothrombin time and partial thromboplastin time, normal serum biochemistry and urinalysis. Subjects with a history of serious illness, metabolic disease, including diabetes, chronic active gastrointestinal disease, a history of drug or alcohol abuse or reported sensitivity to sugars or sweeteners were excluded. The recruited subjects were randomly assigned to sucralose or fructose (control sweetener) in a 2:1 ratio. The sucralose, obtained from Tate & Lyle, Whiteknights, Reading, as an aqueous solution (35 g/litre), was administered daily for 90 days at doses of 125, 250 and 500 mg/day during wk 1±3, 4±7 and 8±12, respectively. Fresh aqueous solutions were prepared on six separate occasions and forwarded to the lab. Aqueous solutions of sucralose have been found to be stable almost inde®nitely. No detectable loss of sucralose occurred when an
the Technician H6000 for haematology. Urine samples were analysed using standard clinical laboratory procedures. Prothrombin time was performed by the method of Quick (1966) and partial thromboplastin time by the method of Procter and Rapaport (1961). ECG recordings were analysed for heart rate, PR interval, QRS duration and QTc interval using Bazett's formula. In addition, sucralose blood level determinations were performed on 10 of the subjects (®ve males and ®ve females) receiving 250 mg sucralose twice daily during wk 12. Blood samples (5 ml) were obtained on 5 consecutive days, both immediately before and 2 hr after the morning sucralose dose. Blood was analysed for sucralose as described above in the 17-day ascending dose study with modi®cation for multiple ion monitoring at M/Z 308 and 310. Also a detailed ophthalmological examination, including slit-lamp examination using a Hamblin SL 202, was carried out before and after the study in a randomly selected group of 24 subjects of whom 18 were taking sucralose (15 male and three female) and six fructose (®ve male and one female). RESULTS
17-Day ascending dose study All eight subjects who were entered completed the 17-day ascending dose study. During the study there were no abnormal variations in body tempera-
S126
I. McLean Baird et al.
Fig. 1. Sucralose blood levels after 10 mg/kg dose. On day 9 of study one, a single dose of 10 mg/kg sucralose was administered orally in the fasting state to the eight subjects. Blood was obtained for sucralose assay at times 0.5, 1, 2, 3, 4, 6, 8 and 24 hr post-dosing. Peak blood concentrations were detected at 1 hr; sucralose was barely detectable at 24 hr. (N = 8).
ture, pulse rate, blood pressure or respiratory rate following either single ascending or continuous doses. In addition, no adverse comments attributed to sucralose were made. The general physical examinations at the end of each phase of the study were normal. During the study, mean haematological values remained stable. There were occasional results outside the normal range for the laboratory, but no trends were apparent and these variations were, therefore, not clinically meaningful. The 17 biochemical values monitored also remained stable throughout the study. No treatment eect or trends were evident, and values remained in the normal range. Two females taking oral contraceptives had marginal elevations in the pre-study thyroxin values, which were unaected by sucralose administration. All urinalysis results taken in Phase 1 and 2 were found to be normal. No eect on urine volume, urine chemistries or sediment was observed. Similarly, 12- lead ECG readings were unaected by sucralose ingestion over the 17-day period. The absorption study of sucralose indicated a general pattern of peak blood levels 1 hr post-dosing which then decreased thereafter. Levels were
barely detectable 24 hr post-dosing (Fig. 1). Sucralose had no eect on fasting serum insulin levels as no variation from the fasting range was detected for each individual (McNeil, 1987). 13-Week repeat dose study A total of 118 subjects were recruited, of which 108 completed the study. Of these, 77 received sucralose (47 males, 30 females) and 31 received fructose (17 males, 14 females) (Table 1). 10 subjects withdrew from the study, three of these taking sucralose and seven taking fructose. None of the withdrawals in the sucralose group was due to adverse experiences. Withdrawals in the sucralose group were due to a loss of appetite during wk 2, a concern about consuming low-calorie sweeteners and non-compliance with the dosing schedule. One subject, taking fructose, exhibited an apparent sensitivity to the compound, while most other dropouts in this group were related to the high sweetness of the test material (Table 2). Overall, a common complaint throughout the study was the intense sweetness of the test beverages. The vast majority of subjects were able to adjust to this, but most indicated it was far in excess of their normal sweetener intake.
Table 1. Composition of sucralose and control groups
Sexes Mean age in yr (SD) Mean weight in kg (SD)
Male Female Total Male Female Overall Male Female Overall
Sucralose group
Control group
47 30 77 33.7 (9.73) 35.9 (11.71) 34.6 (10.52) 78.7 (10.70) 60.2 (6.69) 71.5 (12.97)
17 14 31 34.0 (10.76) 33.7 (11.72) 33.9 (11.02) 73.1 (9.33) 64.6 (7.00) 69.3 (9.28)
Dose study of sucralose in humans
S127
Table 2. Withdrawals from study Sucralose (three subjects) Male (No. 58) Male (No. 110)
Wk 2 Wk 5
Female (No. 32)
Wk 8
Fructose (seven subjects) Male (No. 113) Male (No. 24) Male (No. 98)
Day 2 Day 3 Wk 2
Male (No. 107) Female (No. 12) Female (No. 38) Female (No. 16)
Wk Wk Wk Wk
2 2 8 5
``Loss of appetite'' ``Did not like the idea of taking an arti®cial sweetener for any length of time. OtherwiseÐno complaints or side-eects Not taking sucient ``doses'', therefore, considered an unreliable subject and she was asked to withdraw. No side-eects ``Too sweet'' ``Too sweet'' Skin reactionÐcleared on cessation of fructose intake and reappeared when subject challenged again with two ``doses'' of fructose Persistent sore throat which resolved when fructose intake stopped ``Too sweet'' ``Too sweet'' Decided (reluctantly) to stop as she had to have a minor operation unrelated to the study
There were no changes between initial and ®nal physical examinations in those completing the study, and there was no evidence of weight loss. Mean BP recordings before and after 13 wk showed no changes. There were no signi®cant changes observed in any of the haematologic parameters measured. There was an upward trend in the prothrombin time measurements in both the sucralose and control groups, but there was no statistical dierence between the two groups. There were no signi®cant changes in any of the biochemical parameters or urinalysis. Tracings from the 12-lead ECGs were compared before and after the 13-wk period. Heart rates, PR, QRS and QTc intervals showed no signi®cant changes after sucralose administration. Of the 24 individuals examined by an ophthalmologic consultant, 18 were taking sucralose (15 male and three female) and six fructose (®ve male and one female). Ophthalmoscopic and slit-lamp examination revealed no abnormalities attributable to either sweetener. The daily blood samples taken from 10 subjects each morning during wk 12 immediately before and 2 hr following the administration of the morning dose of sucralose (250 mg) revealed no trend towards increasing fasting or post-dose concentrations of sucralose over time (Table 3). DISCUSSION
The levels of sucralose used in these two studies were in excess of expected human intakes. To understand how much sucralose can be safely con-
sumed by an individual, it is useful to consider three related concepts: highest-no-adverse-eect level (HNEL), acceptable daily intake (ADI) and estimated daily intake (EDI). In order to assess the safety of a new food additive, long-term animal studies are conducted to determine the highest daily dose that can be given to test animals without adverse eects. In the sucralose safety studies, no adverse eects were seen in rodents given up to approximately 1500 mg/kg/day for 104 wk (Goldsmith et al., 2000; Mann et al., 2000a,b). This amount, therefore, was considered the HNEL of sucralose. An ADI represents the amount of a food additive that, if consumed daily throughout a person's life, would be considered safe by a wide margin. This amount is established by dividing the product's HNEL by a large safety factor, usually 100. Dividing the HNEL of sucralose (1500 mg/kg/day) by this 100-fold safety factor yields an ADI of 15 mg/kg/day. The Joint Expert Committee on Food Additives (JECFA) has allocated sucralose an ADI of 15 mg/kg/day (WHO, 1991). Also, the Australian National Food Authority has similarly approved an ADI of 15 mg/kg/day for sucralose. The EDI of a food additive is the amount that is expected to be consumed daily by an individual. To further ensure a product's safety, this projected consumption level should average less than the acceptable daily intake level established for the product. To estimate the EDI of sucralose, an independent research organization, Market Research Corporation of America (MRCA), asked members of more than 2000 households to keep diaries listing all the foods and beverages consumed over a 2-wk
Table 3. Mean and standard deviation (SD) of daily concentration of sucralose in whole-blood during wk 12 of study (ng/ml) while subjects received 5 mg/kg/day Day Time after dosing 0 hr 2 hr
1
2
3
4
5
30 (20) 150 (60)
40 (30) 190 (120)
30 (20) 100 (60)
20 (3) 120 (70)
50 (50) 180 (90)
S128
I. McLean Baird et al. Table 4. EDI, ADI and HNEL of sucralose for individuals of dierent age groups EDI* Mean
90th percentile
Age
mg
mg/kg/day
mg
mg/kg/day
0±23 months 2±5 yr 6±12 yr 13±17 yr 18±44 yr 45±64 yr 65+ yr All ages
22 46 58 57 51 55 51 52
2.0 2.8 2.0 1.0 0.8 0.8 0.8 1.1
49 77 100 110 100 100 93 98
4.3 5.1 3.9 1.9 1.5 1.5 1.5 2.3
ADI** (mg/kg/day)
HNEL*** (mg/kg/day)
15
1500
*EDI = estimated daily intake. **ADI = acceptable daily intake as allocated by JECFA, June 1990. ***HNEL = highest-no-adverseeect level.
period. In their analysis, the researchers assumed that sucralose would replace virtually all sweeteners including sugar in all the food and beverage categories in which it would be approved for use. As it is extremely unlikely that this would ever be the case, the EDI provides an excessive estimate of how much sucralose might be consumed. Despite this very exaggerated set of assumptions, the mean EDI of sucralose for people of all ages is 1.1 mg/kg/day (McNeil, 1987). The EDI was also estimated for individuals of all ages who might consume sucralose-sweetened foods and beverages in great amounts on a continuous basis throughout their lifetime (consumption at the 90th percentile). For these individuals, the EDI is 2.3 mg/kg/day (Table 4). The EDI with its overestimation of consumption has a built-in margin of safety and, consequently, does not remotely approximate a toxic threshold. The overestimation of intake provided by the EDI can be quite large. For example, for another widely used low-calorie sweetener, the EDI is 5 mg/kg, but actual Canadian intake levels are below 1.3 mg/kg (Heybach and Ross, 1989). In terms of human consumption, a 160-lb (73 kg) adult would have to drink 1500 12-oz (360 ml) soft drinks sweetened with sucralose every day to consume an amount comparable to the HNEL. Similarly, a 50-lb (23 kg) child would have to drink 480 12-oz (360 ml) sucralose-sweetened soft drinks daily to reach the HNEL. To reach the ADI of 15 mg/kg/day, a 160-lb (73 kg) adult would have to ingest an amount of sucralose equivalent in sweetness to approximately 1 lb of sugar a day. Even if an individual occasionally consumed sucralose in amounts greater than the ADI, this would pose absolutely no health risk. For instance, if an individual on occasion consumed 10 mg/kg, that level would still be 1/150th the dose which caused no adverse eects in animals. In the 17-day ascending dose study, sucralose doses were administered which were as high as nine times the projected EDI for users of low-calorie
sweeteners. Even given this much higher than expected dietary intake, none of the study subjects reported adverse eects, and their physical, chemical and haematological data were unaected by sucralose ingestion. Sucralose was found to be well tolerated at a single dose of up to 10 mg/kg and then when administered daily in doses of 2 mg/kg for 3 days, then 5 mg/kg for 4 days. Fasting insulin values were also unaected by repeated sucralose ingestion, and the insulin response to sucrose 1 wk after the last dose of sucralose followed the expected pattern. These results indicate that humans, like other mammalian species tested, do not biochemically recognize sucralose as a disaccharide. To assess the eects of chronic sucralose exposure, a single blind sweetener-controlled 13-wk study was undertaken with fructose as the control. During the dosing period, the daily intakes of sucralose varied between 4.8 and 8.0 mg/kg for males and 6.4 and 10.1 mg/kg for females. Many subjects disliked the intense sweet taste of both beverages used to achieve the desired intake level of sucralose by the end of the study, approximately ®ve times the EDI. Several subjects elected to drop out in both groups. However, of those three subjects who dropped out of the sucralose group, none did so due to an adverse experience. During the study, no group dierences in physical, haematological or biochemical data were detected between the sucralose and control groups. Neither were there changes in these parameters over the 13wk period. Similarly, the ECG tracings were unaffected by chronic sucralose ingestion, and ophthalmologic examination remained normal after 13 wk of ingestion in the subjects examined. Consistent with the results from the 17-day study, sucralose was found to be well tolerated in normal volunteers, even when dosed subchronically at high intakes. Analysis of the blood samples taken following the 10 mg/kg single oral dose showed that the peak sucralose concentrations occurred after 1 hr (0.62 0.30 ppm) and decreased over the time period
Dose study of sucralose in humans
with virtually complete elimination after 24 hr (0.02 2 0.02 ppm). These data support the human pharmacokinetic observations which showed that peak plasma sucralose concentrations occurred 1.5 to 3 hr after an oral dose and that negligible accumulation would occur during consumption consistent with the normal pattern of use of a lowcalorie sweetener. (Roberts et al., 2000). The lack of sucralose blood bioaccumulation was shown following the analysis of blood samples after chronic ingestion of 5 mg/kg in wk 12 of the 13-wk study. In this study, there was no indication of rising blood sucralose levels with prolonged high-dose exposure. Both of the human studies show that sucralose is well tolerated in healthy volunteers, even when dosed subchronically at levels greater than would be consumed during normal patterns of use. REFERENCES
Borzelleca J. F. (1992) Macronutrient substitutes: safety evaluation. Regulatory Toxicology and Pharmacology 16, 253±264. Goldsmith L. A. (2000) Acute and subchronic toxicity of sucralose. Food and Chemical Toxicology 38 (Suppl. 2), S53±S69. Goldsmith L. A. and Grice H. C. (2000) SucraloseÐan overview of the toxicity data. Food and Chemical Toxicology 38 (Suppl. 2), S1±S6. Heybach J. P. and Ross C. (1989) Aspartame consumption in a representative sample of canadians. Journal of the Canadian Dietetic Association 50, 166±170. International Food Information Council (1990) Aspartame consumption found safe in new studies. Food Insight, Current Topics in Food Safety and Nutrition. Vol. 6. John B. A., Wood S. G. and Hawkins D. R. (2000) The pharmacokinetics and metabolism of sucralose in the rabbit. Food and Chemical Toxicology 38 (Suppl. 2), S111±S113.
S129
Kille J. W., Ford W. C. L., McAnulty P., Tesh J. M., Ross F. W. and Willoughby C. R. (2000) Sucralose: lack of eects on sperm glycolysis and reproduction in the rat. Food and Chemical Toxicology 38 (Suppl. 2), S19±S29. McNeil Specialty Products Food Additive Petition 7A3987 (1987) (Sucralose). Mann S. W., Yuschak M. M., Amyes S. J. G., Aughton P. and Finn J. P. (2000a) A combined chronic toxicity/ carcinogenicity study of sucralose in Sprague±Dawley rats. Food and Chemical Toxicology 38 (Suppl. 2), S71± S89. Mann S. W., Yuschak M. M., Amyes S. J. G., Aughton P. and Finn J. P. (2000b) A carcinogenicity study of sucralose in the CD-1 mouse. Food and Chemical Toxicology 38 (Suppl. 2), S91±S98. Market Research Corporation of America. (1989) Unpublished report on sucralose intake. Vols I±V. Procter R. R. and Rapaport S. I. (1961) The partial thromboplastic time with kaolin. American Journal of Clinical Pathology 36, 212±219. Quick A. J. (1966) The prothrombin consumption time test. American Journal of Clinical Pathology 45, 105± 109. Randle P. J., Hales C. N., Garland P. B. and Newsholme E. A. (1963) The glucose fatty acid cycle, its role in insulin sensitivity and the metabolic disturbances of diabetes mellitus. Lancet 785±789. Roberts A., Renwick A. G., Sims J. and Snodin D. J. (2000) Sucralose metabolism and pharmacokinetics in man. Food and Chemical Toxicology 38 (Suppl. 2), S31±S41. Sims J., Roberts A., Daniel J. W. and Renwick A. G. (2000) The metabolic fate of sucralose in rats. Food and Chemical Toxicology 38 (Suppl. 2), S115± S121. WHO (1991) Evaluation of Certain Food Additives and Contaminants. Food Additives Series, No. 37, World Health Organization, Geneva. Wood S. G., John B. A. and Hawkins D. R. (2000) The pharmacokinetics and metabolism of sucralose in the dog. Food and Chemical Toxicology 38 (Suppl. 2), S99± S106.