Polydextrose and Soluble Corn Fiber Increase Five-Day Fecal Wet Weight in Healthy Men and Women 1 3

Transcription

1 The Journal of Nutrition Nutrient Requirements and Optimal Nutrition Polydextrose and Soluble Corn Fiber Increase Five-Day Fecal Wet Weight in Healthy Men and Women 1 3 Derek A. Timm, 4 William Thomas, 5 Thomas W. Boileau, 6 Patricia S. Williamson-Hughes, 7 and Joanne L. Slavin 4 * 4 Department of Food Science and Nutrition, University of Minnesota, St. Paul, MN; 5 Division of Biostatistics, School of Public Health, University of Minnesota, Minneapolis, MN; 6 General Mills Inc., Bell Institute of Health and Nutrition, Minneapolis, MN; and 7 Tate and Lyle Health and Nutrition Sciences, Hoffman Estates, IL Abstract Dietary fiber has well-established beneficial effects on laxation. Many fibers have been developed with positive sensory properties and 2 such fibers are polydextrose (PDX) and soluble corn fiber (SCF), which can be added to many commercially produced products. We conducted a randomized, double-blind, placebo-controlled, crossover study comparing the laxative effects of PDX and SCF at a dose of 20 g/d with a low fiber control (LFC) eaten daily as a muffin and cereal in 36 healthy men and women. Each treatment period was 10 d with a 2-wk washout period between. Participants collected fecal samples during the last 5 d of each treatment and completed food diaries and gastrointestinal tolerance questionnaires on d 1, 2, and 10 of each treatment period. Five-day fecal wet weight was higher after the PDX and SCF treatments than the LFC treatment (P # ). The number of stools per day and daily fecal output also were significantly greater during the PDX treatment compared with the LFC treatment. The whole gut transit time did not differ among treatments. The PDX treatment resulted in a softer stool (P = 0.002) than the SCF and LFC treatments. Fecal ph was lowered by the PDX treatment (P = 0.02), whereas SCF tended to lower it compared with the LFC treatment (P = 0.07). When the participants consumed PDX and SCF, they reported significantly more flatulence and borborygmi compared with when they consumed the LFC. Consumption of PDX and SCF at a dose of 20 g/d results in a mild laxative effect with nominal gastrointestinal tolerance issues. J. Nutr. 143: , Introduction Various types of dietary fiber have been shown to influence one or more of the following well-noted physiological effects, including improved laxation, plasma glucose attenuation, and blood lipid reduction (1). Despite these benefits, the mean fiber consumption remains low at ~12 18 g/d, substantially lower than the recommended amounts of 25 g/d for women and 38 g/d for men (1). A possible way to reduce this gap is by incorporating fibers into commercially produced products. With many new fibers being developed with broad food application potential, it is imperative to examine their physiological effects. Two such fibers are polydextrose (PDX) 8 and soluble corn fiber (SCF), which are not broken down in the small intestine but are 1 Supported by General Mills, Inc., Minneapolis, MN and Tate and Lyle, Decatur, IL. 2 Author disclosures: T. W. Boileau works for General Mills, Inc., which provided funding for the study. P. S. Williamson-Hughes works for Tate and Lyle, which provided funding for the study. D. A. Timm, W. Thomas, and J. L. Slavin, no conflicts of interest. 3 This study was registered at clinicaltrials.gov as NCT Abbreviations used: LFC, low fiber control; PDX, polydextrose; ROM, radio-opaque marker; SCF, soluble corn fiber; WGTT, whole gut transit time. * To whom correspondence should be addressed. jslavin@umn.edu. partially fermented in the colon (2 4). PDX is formed from random polymerization of glucose with sorbitol, with the predominant linkage being a-1,6 (2). SCF is produced from the hydrolysis of corn starch and consists of a mixture of a-1,4, a-1,6, a-1,2, and a-1,3 glycosidic linkages (5). The literature on the laxation effects of functional fibers such as PDX and SCF is limited. Laxation typically includes changes in stool weight, stool frequency, stool consistency, or whole gut transit time (WGTT). Stool weight is generally measured during 4 7 d to account for variations in daily stool output. The passage of 80% of inert radio-opaque pellets through the gastrointestinal tract is an accepted way to approximate WGTT (6). Stool weight outcomes for PDX are mixed, with some studies showing significant increases and others showing no difference (4,7 9). Some studies have found a significantly softer stool with the consumption of PDX, whereas others showed no difference in stool consistency (5,7,9). To our knowledge, no difference in WGTT has been reported with PDX compared with a control using radio-opaque markers (ROMs) (4,8). Only 2 crossover studies have been conducted on SCF to date and one reported no differences after a 4-d fecal collection in the number of stools, total stool weight, stool ph, or total SCFA concentration (5). The other study observed a significant increase in 5-d fecal dry ã 2013 American Society for Nutrition. Manuscript received September 25, Initial review completed November 2, Revision accepted January 24, First published online February 20, 2013; doi: /jn

2 matter and a significant decrease in stool ph but no difference in total SCFA concentration compared with control (7). Both studies reported significantly higher flatulence compared with control (5,7). Because the literature is mixed regarding the gastrointestinal effects of these fibers, we designed a study to investigate the gastrointestinal effects of PDX and SCF compared with a low fiber control (LFC). We hypothesized that PDX and SCF would exert significantly higher laxative effects than the LFC. Laxation was examined by measuring 5-d fecal wet weights, stool frequency, gastrointestinal transit time, and stool consistency. Fecal chemistry was investigated by measuring the fecal SCFA concentration and ph. Additionally, we used a selfreported gastrointestinal tolerance questionnaire to measure flatulence, bloating, cramping, borborygmi, nausea, diarrhea, and constipation. Participants and Methods Subjects. Thirty-six participants (n = 18 males, n = 18 females) were recruited from the University of Minnesota by fliers posted on campus and screened by telephone. The participants were required to be 18 y or older, have a BMI between 18.5 and 30, be free of medication use, and have no chronic diseases. All participants had to be free of antibiotic use for at least 3 mo prior to study enrollment. Exclusion criteria included diagnosis of heart, liver, or kidney disease, diabetes, any gastrointestinal conditions, taking antidiarrheal or constipation medications, vegetarianism, consumption of prebiotic or probiotic supplements, and nonbreakfast eaters. All enrolled participants completed the study. The participant characteristics can be found in Table 1. The study protocol was approved by the University of Minnesota Institutional Review Board and all participants gave informed consent prior to beginning the study. Study design. The study used a randomized crossover design with 2-wk washout periods between the 10-d treatment periods. Each participant completed 3 treatment periods where they consumed either 20 g/d of PDX (STA-LITE Polydextrose, Tate and Lyle Ingredients), SCF (PROMITOR Soluble Corn Fiber, Tate and Lyle Ingredients), or an LFC in the form of a breakfast cereal and a muffin. The participants were instructed to consume one packet of cereal and one muffin each day for the 10-d treatment period. Threemenandthreewomenwererandomlyassignedtoeachofthe6 sequences for the 3 treatments, balancing any carryover effects. The treatment cereal and muffins were formulated and produced by General Mills and matched for energy and macronutrient content. The cereal and muffins were analyzed for total fat, protein, dietary fiber, and weight by Medallion Labs. AOAC methods were used for the analysis of total fat (AOAC ), protein (AOAC ), soluble and insoluble fiber (AOAC and ), and total dietary fiber (AOAC and ) (10). Resistant oligosaccharides were determined according to the AOAC method (AOAC (10). Total carbohydrate was calculated by subtraction. The nutritional information of the test products can be found in Table 2. Participants were screened to be on low fiber background diets and our diet records support that the participants consumed a background diet that contained <15 g fiber during each of the 3 diet periods. Participants were instructed to consume their usual diet during the study and include the provided cereal and muffin daily. Food records are collected for 3 d during the study and were averaged to obtain an estimate of usual nutrient intake. We picked d 1, 2, and 10 as the 3 d we would collect the diet records from the participants. We also collected the gastrointestinal tolerance measures on the same days, mostly for participant convenience rather than an expectation that tolerance would be different on any day during the study. Our study was designed to examine whether intake of isolated fiber would change stool weight and we included collection of 5 d of stool weight to get our best measure of stool weight. The collection of fecal samples is a large participant burden but the most important variable in our study. Our design allowed us to compare the effects of 2 fibers, compared with the same treatment without fiber, on stool weight. 474 Timm et al. TABLE 1 Anthropometric data of participants at the start of the trial 1 n Age Height Weight BMI y cm kg Total Men Women Values are mean 6 SD. Stool collection and analysis. On the morning of d 6 of each treatment period, the participants swallowed a single dose of 20 ROM and recorded the time at which this was done. The time at which the pellets were swallowed served as the baseline for transit time calculations. Participants then collected all their fecal samples using a Commode Specimen Collection system (Sage Products) for the next 5 d and maintained a bowel movement diary, where they recorded the time and date of each bowel movement. Each fecal sample was then placed on ice until it could be returned to our laboratory and frozen at 220 C. The fecal samples were individually weighed in a bag, the mean weight of an empty bag was subtracted, and total 5-d stool weight calculated. WGTT was investigated by x-raying each sample to observe the passage of the ROM pellets. The 80% transit time method was used to calculate WGTT (6). The final fecal sample of each treatment period was used for determination of stool consistency, ph, and SCFA. The investigators assessed stool consistency using the 7-point Bristol stool consistency chart, which quantitatively assigns each stool a number based on comparison with the chart (11). Stool ph was assessed by directly inserting an electrode ph probe (PerpHecT_ LogR_ meter model 350, Orion Research) into the homogenized stool sample. SCFA extraction was completed using diethyl ether, with ethyl butyrate serving as an internal standard as described earlier (5). SCFA concentrations were determined using a Hewlett-Packard 5890 Series GC with Stabilwax-DA fused silica column (30 m long; i.d., 0.52 mm; film thickness, 1 mm; Restek). Helium served as the carrier gas with a flow rate of 30 ml/min. The oven, injector, and detector temperatures were 110, 220, and 240 C, respectively. Gastrointestinal tolerance. Participants completed a self-reported gastrointestinal tolerance survey on d 1, 2, and 10 of each treatment period, the same days that the diet records were completed. Each survey consisted of 7 questions regarding levels of flatulence, bloating, cramping, stomach noises, nausea, constipation, and diarrhea they experienced each day. The participants ranked each symptom based on a 0 10 scale, where 0 corresponds to no symptom and 10 correspond to the worst imaginable symptom. This scale was adapted from an earlier published scale (5). The mean of these 3 d was used to determine the tolerance of each treatment. In addition, participants were asked a quality-of-life question regarding how changes in bowel function may have influenced their daily life, ranging from not at all to always. This quality-of-life question was adapted from a longer validated bowel function instrument created for rectal cancer patients (12). Dietary intake. Instructions on how to complete diet records, including a serving size guide, were given to all participants prior to the start of the study. Food records were completed on d 1, 2, and 10 of each treatment period to determine nutrient intakes on habitual diets and document that the dietary treatment did not alter background nutrient intake. Days 1, 2, and 10 were chosen to coincide with the gastrointestinal tolerance questionnaire to ensure participant compliance. Three diet records are considered more reliable than only a 1-d record and can provide estimates of nutrient intake while consuming habitual diets. All caloric foods and beverages except test products were recorded and entered into Nutrition Data System for Research. Total energy, protein, fat, carbohydrates, and dietary fiber of the background diet are displayed as a mean of the 3-d diet records and compared among treatments. Statistical analysis. Participant demographic data are presented as means 6 SDs. Treatments were compared using mixed-effects linear

3 TABLE 2 Nutritional composition of study cereals and muffins 1 Product Serving size Moisture Fat Protein Sugar Soluble TDF 2 Insoluble TDF RO 2 TDF + RO 2 Energy g % g kj LFC Cereal Muffin PDX Cereal Muffin SCF Cereal Muffin Values are mean of analytics after product preparation. LFC, low fiber control; PDX, polydextrose; RO, resistant oligosaccharide; SCF, soluble corn fiber; TDF, total dietary fiber. 2 TDF and RO were analyzed by AOAC method models with treatment and visit as fixed effects and with a random intercept for each participant to model correlation between repeated measurements from the same participant. For each outcome, equal carryover and treatment-visit interaction were checked by these mixedeffects models. Differences of least square means were used to determine differences among treatments. Significance was achieved at P < 0.05 and 0.05 < P # 0.10 was considered a trend. Data analysis was performed by SAS version 9.2 (SAS Institute). Results All 36 participants finished the study and completed study measurements. The participantsõ self-reported ethnicity was as follows: 27 Caucasian, 5 African American, and 4 Asian. Excluding test products, there were no differences between treatment periods for dietary fat, protein, carbohydrate, dietary fiber, or energy (Table 3). Nutrient intakes did not vary among d 1, 2, and 10 on any treatment. The mean dietary fiber, excluding test products, in the participantsõ background diet was ;14 g/d with no differences among the treatment periods (Table 3). The 5-d stool wet weights of the participants during the PDX and SCF periods were greater than during the LFC period (P # ) (Table 4). Also, the number of stools passed by participants during the PDX and SCF periods was significantly higher than during the LFC period (P # ) (Table 4). However, a treatment-period interaction was observed for the number of stools (P = 0.05). This indicates that participants passed more stools during the fiber treatments, but this varied by treatment period. In contrast, no differences in the mean weight per stool were observed among treatment periods (Table 4). Daily fecal output during the PDX period was greater than during the LFC period (P = 0.02), whereas that during the SCF period tended to be greater (P = 0.08) than during the LFC period and did not differ from the PDX period. Fecal wet weight increased by 2.07 g/g PDX and 1.62 g/g SCF. All ROM markers swallowed by participants passed during the 5-d collection period of each treatment, indicating excellent compliance with the study protocol. WGTT did not differ among treatment periods (Table 4). The consistency of the stools passed by the participants was determined by the investigators based on the 7-point Bristol Stool Consistency Chart and showed that PDX consumption resulted in softer stools (P =0.002)comparedwithboththeSCFandLFC treatments (Table 4). Additionally, fecal ph was lowered (P = 0.02) by consumption of the PDX treatment compared with the LFC treatment, whereas ingestion of the SCF treatment tended (P = 0.07) to lower fecal ph compared with the LFC treatment. The total fecal SCFA concentration was higher when participants consumed the LFC treatment compared with the PDX treatment, with the SCF treatment being similar to both the LFC and PDX treatments (Table 5). Participants consuming the SCF and LFC treatments had higher fecal acetate concentrations than during the PDX period. Fecal propionate concentrations significantly differed among all treatments periods with LFC > SCF > PDX. Fecal butyrate concentrations were higher (P = 0.02) during the LFC treatment period than the PDX treatment period was but similar to the SCF period. When examined as ratios of total SCFAs, the concentrations of acetate, propionate, and butyrate were 33.6:31.5:34.9 during the LFC treatment period, 33.8:32.0:34.2 during the PDX period, and 35.3:30.9:33.8 during the SCF treatment. The SCF treatment had a higher (P = 0.03) acetate ratio than the LFC treatment and tended (P = 0.07) to be higher than PDX. The propionate and butyrate ratios did not differ among the treatments periods. The subjective tolerance scores were all minimal, indicating that the fiber treatments were well tolerated by the study participants (Table 6). No differences were found among results on d 1, 2, and 10 for the subjective tolerance scores. Significantly higher flatulence and borborygmi were reported when participants consumed the PDX and SCF compared with the LFC treatment. For all other questions, no differences among treatments were noted. Mean daily total tolerance scores were higher (P = 0.03) during the SCF period compared with the LFC period, TABLE 3 Daily energy and nutrient intakes of men and women who consumed for 10 d each muffins and cereals containing LFC, PDX, or SCF 1,2 LFC PDX SCF Fat, g/d Protein, g/d Carbohydrate, g/d Soluble fiber, g/d Insoluble fiber, g/d Total fiber, g/d Energy, kj/d Values are mean 6 SD, n = 36. LFC, low fiber control; PDX, polydextrose; SCF, soluble corn fiber. 2 Values reflect dietary intake of the participants not including study products. Laxative effects of functional fibers 475

4 TABLE 4 WGTT, 5-d stool output, fecal ph, and stool consistency of men and women who consumed for 10 d each muffins and cereals containing LFC, PDX, or SCF 1 LFC PDX SCF WGTT, h d fecal wet weight, g b a a Stools in 5-d collection, n b a a Fecal wet weight, g/stool Stool wet weight, g/d b a ab Fecal weight increase above control period, 2 g/g fiber supplemented Fecal ph a b ab Stool consistency b a b 1 Values are mean 6 SD, n = 36. Labeled means in each row without a common letter differ, P, LFC, low fiber control; PDX, polydextrose; SCF, soluble corn fiber; WGTT, whole gut transit time. 2 Mean fecal weight during the LFC period was subtracted from weight during the PDX and SCF treatment periods divided by the amount of supplemental fiber consumed during the 5-d collection period. 3 Values are rated on the Bristol stool consistency chart, where 1 = separate hard lumps and 7 = entirely liquid. whereas the PDX period was similar to both the SCF and LFC periods. Lastly, we did not observe differences among the treatment periods for the bowel habit quality-of-life question. Discussion Although the laxative effects of fiber are generally accepted, few studies exist on the ability of functional fibers to affect laxation. Laxation is relatively loosely defined, but objective measures, including stool weight and transit times, are good markers to determine if fiber improves laxation. Our study shows modest laxative effects of both PDX and SCF as evident by increases in 5-d stool weight and frequency of defecation compared with LFC. PDX and SCF were well tolerated at a dose of 20 g/d in addition to the usual background fiber intake of our participants. Participants consumed ~13 14 g/d of dietary fiber based on diet records, suggesting that both SCF and PDX can be successfully added to usual American diets, thus improving fiber intake and increasing stool weight without causing excessive gastrointestinal tolerance issues. Some earlier studies have observed inconsistent results of the effect of PDX and SCF on stool weight; however, these studies fed these fibers in doses of 8 12 g/d (5,9,13). A recent study investigated thelaxativeeffectsof21g/dofpdxandscfinadultmenand reported PDX and SCF increased fecal wet weight compared with control, which corroborates our findings (7). Interestingly, this study observed a greater laxative effect of SCF than PDX, whereas we observed no differences in stool weight between the fibers. It is worth noting that various fibers affect stool weight differently. The laxative effects of fiber can be investigated by expressing the increase in stool weight, above baseline or control period, per gram of fiber fed. The laxative effects of some other common fibers such as wheat bran (; g wet weight/g), cellulose and cellulose derivatives (;3 4 g wet weight/g), psyllium (;3 4 g wet weight/g), and pectin (;1 2 g wet weight/g) suggest that fibers not digested during gut transit, i.e., wheat bran and cellulose, are more effective at increasing stool weight than fibers that are extensively broken down by microbes in the gut, i.e., pectin (14). We observed increases of 2.07 g and 1.62 g fecal weight/g fiber fed as PDX and SCF, respectively, compared to the LFC period. PDX and SCF are partially fermented in the large intestine by microbes and their laxative effects are similar to pectin, which is also fermentable in the large intestine (2 4,13). A previous study of PDX and SCF noted 476 Timm et al. a similar trend but with lesser increases of 1.4 and 0.9 g fecal weight/g of PDX and SCF, respectively (7). Taken together, our study and the previous study show mild laxative effects of PDX and SCF at a dose of 20 g/d (7). PDX and SCF significantly increased participantsõ daily fecal output to 166 and 157 g/d compared with 125 g/d during the LFC treatment. This is a noteworthy finding, because it has been suggested that fecal weights >150 g/d may reduce the incidence of colon cancer (15,16). We did not observe significant differences in WGTT, measured by ROM, in this study. Previous results regarding PDX and transit time are mixed. One study reported no change in transit time when consuming 30 g/d PDX (8). In contrast, another study reported 8 g/d PDX decreased WGTT; however, this study used colored dyes and not ROM to examine transit time (9). Changes in WGTT are most important when examined in the context of diarrhea (fast transit) or constipation (slow transit); however, we screened for these conditions, so it may not be a surprise that no differences in transit time were observed. It is worth noting that faster WGTT is associated with increased stool weight, which may reduce the chance of developing colon cancer due to decreased exposure of cytotoxic compounds to the colonic epithelial cells (15). The fecal ph of participants was significantly lower when they ingested PDX compared with the LFC treatment. When participants consumed SCF, the fecal ph was similar to the LFC treatment in this study and was also observed in a previous study (5). In contrast, a previous study reported that when participants ingested SCF, fecal ph was significantly lower compared with TABLE 5 Fecal SCFA concentrations of men and women who consumed for 10 d each muffins and cereals containing LFC, PDX, or SCF 1 LFC PDX SCF mmol/g stool wet weight Acetate a b a Propionate a c b Butyrate a b ab Total SCFA a b ab 1 Values are mean 6 SD, n = 36. Labeled means in each row without a common letter differ, P, 0.05.LFC,lowfibercontrol;PDX,polydextrose;SCF,solublecorn fiber.

5 TABLE 6 Mean daily gastrointestinal tolerance scores of men and women who consumed for 10 d each muffins and cereals containing LFC, PDX, or SCF 1 Symptom 2 LFC PDX SCF Flatulence b a a Bloating Cramping Borborygmi b a a Nausea Constipation Diarrhea Mean daily total b ab a 1 Values are mean 6 SD, n = 36. Labeled means in each row without a common letter differ, P, LFC, low fiber control; PDX, polydextrose; SCF, soluble corn fiber. 2 Each symptom rated on a 0 10 scale. 3 Mean sum of all questions per day. the control period, whereas consumption of PDX resulted in a similar fecal ph to both the control and SCF periods (7). A randomized, parallel study of 120 Chinese men and women showed a dose-response decrease in fecal ph with doses of PDX of 0, 4, 8, and 12 g (13). Overall, the current study and literature show significant decreases in fecal ph as a result of PDX and SCF consumption. Decreases in fecal ph may inhibit the growth of some pathogenic bacteria, including Clostridium perfringens, decrease conversion of bile acids to carcinogenic secondary bile acids, and make minerals more soluble, thereby increasing absorption (17,18). A significantly softer stool consistency was observed with the PDX treatment, with the SCF and LFC treatments being similar. Other recent studies on PDX and SCF did not observe any differences in stool consistency compared with controls (5,7,19). One potential explanation for the mixed results is that we used the 7-point Bristol stool consistency scale, whereas other studies used 4- or 5-point scales, which may not be as sensitive to changes as the 7-point scale. The benefits of SCFAs on gut health are well noted, especially the role of butyrate as an energy source for the colonocytes and cell proliferation and differentiation (20). Contrary to what we would expect, the total fecal SCFA concentration was significantly lower for PDX compared with LFC, whereas SCF was similar to both treatments. With the consumption of 20 g of fermentable fiber, one would expect higher fecal SCFA concentrations compared with a LFC; however, it is possible that the increased fecal weight may have diluted the SCFA concentrations. Interestingly, a dilution effect has been suggested in other recent studies examining PDX and SCF (7,19). Also, it should be noted that the majority of SCFAs are produced in the proximal colon and ~95% are absorbed soon after production; therefore, fecal SCFAs are more representative of distal colon concentrations rather than the proximal colon (17). Furthermore, in vitro studies of PDX show it produces low SCFA concentrations compared with other fibers, thereby supporting our current findings (21,22). Apart from looking at concentrations, proportions of SCFAs are a useful way to compare SCFAs across studies. Comparing our SCFA proportions with previous studies shows a marked difference in the SCFA proportions. One previous clinical trial reported higher acetate, lower propionate, and lower butyrate ratios for PDX (68:18:15) and SCF (67:19:15), whereas our results were 33.8:32.0:34.2 for PDX and 35.3:30.9:33.8 for SCF. In contrast, our SCFA proportions for SCF compare more favorably with an earlier study that reported SCF had a ratio of 43:26:31 (5). These differences in ratios may be due to differences in gut microflora, SCFA extraction method, study participants, diet consumed, or other factors. The previous study extracted SCFAs from acidified fecal matter and expressed as mmol/g fecal dry matter, whereas our study and the earlier study from our laboratory group extracted SCFAs using diethyl either and expressed as mmol/g fecal wet weight (5,7). Our SCFA results disagree with previous data and highlight the need for a standardized SCFA extraction method so that results can be compared among studies. Progress in the SCFA field will be limited unless standardized methods are used across all human studies. Measurement of fecal SCFAs is definitely affected by fecal collection procedures, fecal handling, solvent extraction methods, and GC protocols, and without accepted standards in the field, results must be considered preliminary and not overinterpreted. Interest in fecal SCFAs continues, as there are limited measures of fiber fermentation in human trials. The increase in flatulence from the fiber treatments is expected due to an increase in fermentable substrate in the colon. Although this increase in flatulence was significant, the absolute level on the scale was still relatively mild, with a mean of 3 of 10. Recent studies indicate that a significant increase in flatulence is common for both fibers (5,7,19). Moreover, the similarity in the bowel health quality-of-life question further supports that these fibers are well tolerated. Overall, the PDX and SCF treatments did cause significant increases in tolerance symptoms; however, no serious incidences were reported, indicating that a split 20-g dose of PDX and SCF is well tolerated. Thus, both PDX and SCF increase stool weight in healthy individuals when the fibers are added to cereal and muffins. Because an increase in stool weight is accepted as a biomarker of improved laxation, these results demonstrate these fibers improve laxation when added to processed foods and beverages. Acknowledgments The authors thank Michelle Manderfield and Fern Panda of General Mills, Inc. for formulating and creating the test products used in this study. Nort Holschuh of General Mills, Inc. completed the sample size calculation. Abby Klosterbuer assisted with laboratory work. D.A.T., T.W.B., and J.L.S. designed the study; D.A.T. and J.L.S. conducted research; W.T. provided guidance on statistical analysis; D.A.T. analyzed data; and D.A.T., P.S.W.-H., and J.L.S. wrote the paper. All authors read and approved the final manuscript. Literature Cited 1. Institute of Medicine. Dietary, functional, and total fiber. Dietary reference intakes for energy, carbohydrate, fiber, fat, fatty acids, cholesterol, protein, and amino acids ed. Washington, DC: The National Academies Press; p King NA, Craig SA, Pepper T, Blundell JE. Evaluation of the independent and combined effects of xylitol and polydextrose consumed as a snack on hunger and energy intake over 10 d. Br J Nutr. 2005;93: Maathuis A, Hoffman A, Evans A, Sanders L, Venema K. The effect of the undigested fraction of maize products on the activity and composition of the microbiota determined in a dynamic in vitro model of the human proximal large intestine. J Am Coll Nutr. 2009;28: Achour L, Flourie B, Briet F, Pellier P, Marteau P, Rambaud JC. Gastrointestinal effects and energy value of polydextrose in healthy nonobese men. Am J Clin Nutr. 1994;59: Laxative effects of functional fibers 477

6 5. Stewart ML, Nikhanj SD, Timm DA, Thomas W, Slavin JL. Evaluation of the effect of four fibers on laxation, gastrointestinal tolerance and serum markers in healthy humans. Ann Nutr Metab. 2010;56: Hinton JM, Lennard-Jones JE, Young AC. A new method for studying gut transit times using radioopaque markers. Gut. 1969;10: Boler BM, Rossoni Serao MC, Bauer LL, Staeger MA, Boileau TW, Swanson KS, Fahey GC. Digestive physiological outcomes related to polydextrose and soluble maize fibre consumption by healthy adult men. Br J Nutr. 2011;106: Tomlin J, Read NW. A comparative study of the effects on colon function caused by feeding ispaghula husk and polydextrose. Aliment Pharmacol Ther. 1988;2: Hengst C, Ptok S, Roessler A, Fechner A, Jahreis G. Effects of polydextrose supplementation on different faecal parameters in healthy volunteers. Int J Food Sci Nutr. 2009;60 Suppl 5: AOAC. Official methods of analysis. 18th ed. Washington, DC: AOAC; Lewis SJ, Heaton KW. Stool form scale as a useful guide to intestinal transit time. Scand J Gastroenterol. 1997;32: Temple LK, Bacik J, Savatta SG, Gottesman L, Paty PB, Weiser MR, Guillem JG, Minsky BD, Kalman M, Thaler HT, et al. The development of a validated instrument to evaluate bowel function after sphincterpreserving surgery for rectal cancer. Dis Colon Rectum. 2005;48: Jie Z, Bang-Yao L, Ming-Jie X, Hai-Wei L, Zu-Kang Z, Ting-Song W, Craig SA. Studies on the effects of polydextrose intake on physiologic functions in Chinese people. Am J Clin Nutr. 2000;72: Cummings JH. The effect of dietary fiber on fecal weight and composition. CRC handbook of dietary fiber in human nutrition. 2nd ed. Boca Raton (FL): CRC Press; p Cummings JH, Bingham SA, Heaton KW, Eastwood MA. Fecal weight, colon cancer risk, and dietary intake of nonstarch polysaccharides (dietary fiber). Gastroenterology. 1992;103: Birkett AM, Jones GP, de Silva AM, Young GP, Muir JG. Dietary intake and faecal excretion of carbohydrate by Australians: importance of achieving stool weights greater than 150 g to improve faecal markers relevant to colon cancer risk. Eur J Clin Nutr. 1997;51: Topping DL, Clifton PM. Short-chain fatty acids and human colonic function: roles of resistant starch and nonstarch polysaccharides. Physiol Rev. 2001;81: Thornton JR. High colonic ph promotes colorectal cancer. Lancet. 1981;1: Fastinger ND, Karr-Lilienthal LK, Spears JK, Swanson KS, Zinn KE, Nava GM, Ohkuma K, Kanohoris S, Gorden DT, Fahey GC. A novel resistant maltodextrin alters gastrointestinal tolerance factors, fecal characteristics, and fecal microbiota in healthy adult humans. J Am Coll Nutr. 2008;27: Wong JM, de Souza R, Kendall CW, Emam A, Jenkins DJ. Colonic health: fermentation and short chain fatty acids. J Clin Gastroenterol. 2006;40: Vester Boler BM, Hernot DC, Boileau TW, Bauer LL, Middelbos IS, Murphy MR, Swanson KS, Fahey GC. Carbohydrates blended with polydextrose lower gas production and short-chain fatty acid production in an in vitro system. Nutr Res. 2009;29: Hernot DC, Boileau TW, Bauer LL, Middelbos IS, Murphy MR, Swanson KS, Fahey GC. In vitro fermentation profiles, gas production rates, and microbiota modulation as affected by certain fructans, galactooligosaccharides, and polydextrose. J Agric Food Chem. 2009; 57: Timm et al.