RESEARCH ARTICLE

Snack-Eating Patients Experience Lesser Weight Lossafter Roux-En-Y Gastric Bypass Surgery

Silvia Leite Faria & Emily de Oliveira Kelly &

Orlando Pereira Faria & Marina Kiyomi Ito

Received: 13 June 2008 /Accepted: 10 September 2008 /Published online: 2 October 2008# Springer Science + Business Media, LLC 2008

AbstractBackground In bariatric surgery patients, weight loss andlong-term weight maintenance are related to food intake andeating patterns. To improve the diet orientation in the bariatricsurgery postoperative period, we assessed the postoperativeeating patterns and related them to weight loss.Methods This was a transversal, analytical, and descriptivestudy that assessed body mass index (BMI) values andpercentage of excess weight loss (%EWL) in patients whohad undergone Roux-en-Y gastric bypass (RYGBP) surgery.The eating pattern and energy intake were investigated basedon data collected through a 4-day food intake record. Fromthese records, we assessed the number of daily meals, thequantity of food per meal, and calorie value of snacks betweenmain meals. Based on these records, patients were classifiedunder sweet-eating, snack-eating, or normal-eating patterns.Results Seventy-five patients met our inclusion criteria. Thenormal-eating pattern group was the one with the greatestweight loss with an average %EWL of 71.4±21%, followedby the sweet-eating pattern with 69.9±16.8%, and the snack-eating pattern with 56.4±16.7%. This difference was signif-icant only between the first and the third group (p=0.04). Thesnack-eating patients had the highest caloric intake andhighest number of daily meals (p<0.01).

Conclusion Postoperative eating pattern influenced post-bariatric surgery weight loss. In the present study, thesnack-eating pattern was associated with the worst weightloss outcome, followed by the sweet-eating and normal-eating patterns. A screening and a differential approach topatients according to their eating patterns may lead to betterresults of weight loss.

Keywords Eating pattern . Snack-eating . Sweet-eating .

Weight loss . Roux-en-Y gastric bypass

Introduction

In bariatric surgery patients, weight loss and long-termweight maintenance are related to food intake and eatingpattern, particularly after the sixth postoperative month [1].An individual’s eating pattern is an important diet indicatorthat reflects the profile of calorie, protein, vitamin, andmineral intake and frequency of consumption of differentfood groups [2].

Bariatric surgery forces patients to restrict their intakevolume, but it does not necessarily help improve intakequality. A frequent consumption of snacks—with highcontent of fat and sugar—may lead to excessive calorieintake, which can be related to insufficient weight loss orweight regain in the late postoperative period (after the 24thpostoperative month) [3, 4].

Brolin et al. [3] assessed different eating patterns andweight loss after bariatric surgery, underscoring sweetseaters (defined as individuals who consume sweets orsugar-added liquids that add up to more than 150 kcal perportion three times or more per week) and snack eaters(defined as those who consume in between meals foods thathave calorie values greater than 150 kcal per portion). The

OBES SURG (2009) 19:1293–1296DOI 10.1007/s11695-008-9704-7

S. Leite Faria (*) :M. Kiyomi ItoUniversidade de Brasília,Brasilia, Brazile-mail: silvialeitefaria@brturbo.com.br

S. Leite Faria :O. Pereira FariaGastrocirurgia de Brasília,Brasília, Brazil

S. Leite Faria : E. de Oliveira KellyGastronutrição Nutrição Bariátrica,Brasília, Brazil

authors suggests that snack-eating patients will have lesserweight loss and that Roux-en-Y gastric bypass (RYGBP)patients who are not in the habit of consuming sweets, icecreams, and snacks have a bigger chance of maintainingtheir weight loss in the long-term [3, 5].

Other authors [6, 7] have assessed food consumptionafter bariatric surgery and verified that, after the sixthpostoperative month, there is an adaptation in the intakevolume and patients start returning to their usual eatinghabits. Andersen and Pedersen [8] observed the importanceof food intake in gastroplasty patients. His work reports thatthere was no improvement in food intake quality aftersurgery, which reinforces the importance of postoperativenutritional follow-up.

Aiming to provide information that enriches dietorientation in the bariatric surgery postoperative periodand which favors weight loss and long-term weightmaintenance, this research sought to assess postoperativeeating patterns, relating them to weight loss.

Methodology

This was a transversal, analytical, and descriptive study.The research encompassed patients who underwentRYGBP technique by the same surgical team at theGastrocirurgia de Brasília clinic and by the same surgeon(Dr. Orlando Pereira Faria), thus forming a conveniencesample. These patients were followed-up by the nutritionalservice of our clinic between the years of 2003 and 2006.

The research subjects were selected based on follow-ing inclusion criteria: each had been operated on by thesame surgeon, using the RYGBP technique; were aged16–65 years; had undergone surgery at least 1 yearearlier; and had had periodical medical and nutritionalfollow-ups (minimum of twice a year). The exclusioncriteria were: patients who did not meet the inclusioncriteria; patients who worked in the nutrition area or whohad studied nutrition; patients with psychiatric disease(depression, phobias, etc.) who required psychotropictreatment; and bulimic and/or anorexic patients. Theresearch was approved by the Research Ethics Commit-tee of the University of Brasília’s Health SciencesFaculty, filed under no. 052/2002.

Patients had their weights and heights measured by thesame anthropometrist. Body weight was measured afterovernight fasting with a digital scale with maximum capacityof 300 kg and sensitivity of 100 g (Filizola®). Height wasmeasured with a wall stadiometer with 5 mm sensitivity.

The study assessed body mass index (BMI) values andthe percentage of excess weight loss (%EWL) with theideal weight calculated based on values taken from theMetropolitan Life Foundation table [9].

Eating patterns and energy intake were investigatedbased on data collected with a 4-day food intake record.From these records, we assessed the number of daily meals,the quantity of food per meal, and the caloric value ofsnacks taken between main meals.

Using a nutrition software application called Nutrisurvey[10], we assessed the mean daily caloric intake, mean dailyprotein intake in grams and as a percentage of total energyintake (TEI), mean daily lipid intake in grams and as apercentage of TEI, and mean daily carbohydrate intake ingrams and as a percentage of TEI. We verified thedistribution (frequency and quantity) of foods in theindividual’s usual eating habits, focusing on the quantityand frequency of the consumption of sweets and snacks.

Patients were classified as displaying sweet-eating,snack-eating, or normal-eating patterns according to thefollowing parameters:

& Sweet-eating pattern: Patients who consumed 150 kcal ormore per portion in the form of sweets between mainmeals (breakfast, lunch, and dinner). This occurred at leasttwice during the 4 days of the food intake record [3].

& Snack-eating pattern: Patients who consumed 150 kcalor more per portion in the form of snacks in betweenmeals. This occurred at least twice during the 4 days ofthe food intake record. The frequency of the consump-tion of snacks was defined arbitrarily [3].

& Normal-eating pattern: Patients were classified asdisplaying normal-eating pattern when they did notconform to any of the behaviors described above.

Statistical Analysis

The statistical program used was SAS version 8.2. Avariance test was used to assess whether caloric intake,weight loss, protein intake, carbohydrate intake, and thenumber of meals were equal among the three eating patterngroups. When equality among groups was rejected, Tukey’smultiple comparisons test was applied. In all analysis, alevel of significance of 0.05 was used.

Table 1 Patient characteristics before and after surgery

Mean

Male/female 15/60Age (years) 36.8 (±10.7)Average time of surgery (months) 23.0 (±10.3)Preoperative weight (kg) 120.4 (±19.5)Postsurgical weight (kg) 81.9 (±15.6)Preoperative BMI (kg/m2) 43.0 (±5.5)Postsurgical BMI (kg/m2) 29.8 (±4.6)Excess weight (kg) 58.5 (±16.3)%EWL 67.5 (±18.8)

1294 OBES SURG (2009) 19:1293–1296

Results

Of the 100 patients assessed, only 75 met our study’sinclusion criteria. Most of the patients were females (80%).The average age was 36.8±10.7 years, ranging from 19 to64 years. Patients had undergone surgery 23±10.3 monthsearlier, on average. Anthropometric and demographicvariables for the whole sample are summarized in Table 1.

The studied population had an average preoperative BMIvalue of 43±5.5 kg/m2 and a postsurgical BMI value of 29.8±4.6 kg/m2. Average %EWL after surgery was of 67.5±18.8%.The population’s average energy intake was 1,475±546 kcaland the mean protein intake was 73.4±30.9 g.

After classifying the patients according to their eatingpatterns, we verified that the normal-eating pattern was themost frequent, comprising 45.3% of the sample, followedby the sweet-eating pattern at 36%, and the snack-eatingpattern at 18.6%.

When the different eating patterns were analyzed individ-ually, we noted several differences among the three groups.The normally eating group had the greatest weight loss withan average %EWL of 71.4±21%, followed by the sweet-eating group at 69.9±16.8%, and the snack-eating group at56.4±16.7% (Fig. 1). These differences were significant only

between the first and third groups (p=0.04). Differences inthe constituent macronutrients and in energy intake were alsoobserved among the diets, as shown in Table 2.

Discussion

Our results show a 67.5% loss of excess weight in the studiedpatients, which represents a satisfactory loss after bariatricsurgery. Our patient sample consumed an average TEI of1,475 kcal/day 23 months after surgery, in a range from 530 to3,055 kcal. Values in the literature vary: 1,100±426 kcal/dayafter 1 year [3, 11], 1,500 kcal/day at 18 months [12], and1,800 kcal/day 2 years after surgery [13]. All these studiesinvolved patients who had undergone surgery with theRYGB technique.

The snack-eating pattern observed in this work wasdetermined based on the criteria established by Brolin et al.,emphasizing the nutritional characteristics of the diet. Inother studies, this pattern has been defined as a subclinicaleating disorder, characterized by the consumption of smallquantities of foods throughout the day with a sensation ofloss of control [1, 14]. These studies relate this pattern toworse weight loss results and underscore the importance ofcognitive–behavioral therapy in treating this disorder.

In our study, the snack-eating-pattern group showedlower %EWL than that of the groups with other eatingpatterns. This may be attributable to the greater caloricintake and to the higher number of daily meals observed inthis group. The work of Brolin et al. [3] suggested that thesnack-eating pattern results in less weight loss than theother patterns or even weight regain because of the higherassociated caloric intake. We also observed that the snack-eating group had a higher calorie intake per kilogram ofbody weight, a parameter that was significantly different inall the studied groups (p<0.01).

The higher caloric intake of the snack-eating pattern isrelated to the type of food consumed by this group. Themost commonly consumed foods (snacks such as crackers,potato chips, high-fat microwave popcorn, etc.) crumbleeasily, demand less chewing time, and empty quickly fromthe gastric pouch. This eating pattern is also related to alower sense of satiety [4].

Table 2 Diet composition and number of meals of the three eating patterns

Energy intake(kcal)

Kilocalorie perkilogram of weight

Protein (g) Carbohydrate (g) Fat (g) Numberof meals

Snackers 2,086* (±480) 24.3* (±8.4) 94.8* (±27.5) 247.1*(±50.7) 30.7 (±6.3) 7.4* (±1.81)Sweet eaters 1,530* (±481) 19.5* (±5.6) 73.7 (±31.9) 178.8* (±53.4) 32.1 (±4.4) 5.2 (±1.2)Normal 1,179* (±379) 15.3* (±5.1) 64.4* (±27.6) 127.6* (±35.9) 32.4 (±6.1) 4.9 (±0.8)

*p<0.01

Fig. 1 %EWL in relation to eating pattern

OBES SURG (2009) 19:1293–1296 12951295

Another important factor that may be related to the lowerweight loss among snackers is the constant insulin peaksproduced by the continuous consumption of carbohydrate-rich foods. Insulin is a lipogenic hormone, the consequen-ces of which are weight gain and appetite increase [15].

In their study, Kenler et al. [5] emphasized theimportance of replacing snacks with foods that are proteinsources (beef, chicken, fish, egg, and cheese) because thistype of food stays longer in the gastric pouch, thusincreasing satiety. Sugerman et al. [4] suggested replacingsnacks with fiber-rich foods such as raw, unpeeledvegetables and fruits, for the same reason.

The average protein intake of the total population, of thesweet-eating group, and of the normal group remainedbelow the recommended daily value of 80 g of proteins,taking into consideration that the population was formedpredominantly of women (recommended daily value formen is 100 g) [11]. Only the snack-eating group had anadequate intake, probably a reflection of the greater foodconsumption shown by this group. According to Moize etal. [11], bariatric surgery patients tend to consumeinadequate quantities of protein. A low-protein diet is notdesirable because it causes lean mass loss and a consequentreduction in the basal metabolic rate, making weight lossand weight maintenance more difficult.

Sugerman et al. [4] and others have postulated thatRYGBP patients end up avoiding sweet foods for fear ofdeveloping “dumping syndrome.” However, the presentstudy identified many patients displaying the sweet-eatingpattern, as have other studies of the RYGBP postoperativeperiod [3, 5]. A probable explanation for this is that there isa gradual increase in tolerance for sweets throughout thepostoperative period. Whereas the %EWL of the sweet-eating group was not significantly different from that of thenormally eating group, the former group showed a caloricintake significantly higher than that of the latter group, andthis may negatively influence weight loss [5, 6].

There was no statistical difference in the ingestion of fatamong the three groups (p=0.63), and there was norelationship between fat consumption and weight loss. Animportant bias of the present study is that the researcherwas the nutritionist who followed-up the group of patients.This may have influenced the patients’ answers.

Conclusion

The postoperative eating pattern influenced postbariatricsurgery weight loss. In this resent study, the snack-eatingpattern was associated with the worst weight loss outcome,

followed by the sweet-eating and normal-eating patterns.Screening and a differential approach to patients according totheir eating patterns may lead to better weight loss results. Thesnack-eating pattern has not often been studied in the bariatricsurgery population. More studies are required to determine theinfluence of this factor on surgery results.

The simple performance of the RYGBP technique is notsufficient to achieve good outcomes. The nutritionist is aprofessional who has several tools available for assessing andorienting patients’ eating patterns. The use of these tools isnecessary because the goal of surgery is excess weight lossand a proper eating pattern is essential in this process.

References

1. Colles SL, Dixon JB, O’Brien PE. Grazing and loss of controlrelated to eating: two high-risk factors following bariatric surgery.Obesity 2008;16(3):615–22.

2. Togo P, Osler M, Sorensen TI. Food intake patterns and bodymass index in observational studies. Int J Obes Relat MetabDisord 2001;25:1741–51.

3. Brolin RE, Robertson LB, Kenler HA, et al. Weight loss anddietary intake after vertical banded gastroplasty and Y-en-Rouxgastric bypass. Ann Surg 1994;220(6):782–90.

4. Sugerman HJ, Kellum JM, Engle KM, et al. Gastric bypass fortreating severe obesity. Am J Clin Nutr 1992;55:560S–6S.

5. Kenler HA, Brolin RE, Cody RP. Changes in eating behavior afterhorizontal gastroplasty and Roux-en-Y gastric bypass. Am J ClinNutr 1990;52:87–92.

6. Shah M, Simha V, Garg A. Review: long-term impact of bariatricsurgery on body weight, comorbidities and nutritional status. JClin Endocrinol Metab 2006;91:4223–31.

7. Kolanowski J. Surgical treatment for morbid obesity. Br Med Bull1997;53(2):433–44.

8. Andersen T, Pedersen BH. Pouch volume, stoma diameter, andclinical outcome after gastroplasty for morbid obesity. Scand JGastroenterol 1984;19(5):643–9.

9. Metropolitan Life Foundation. Metropolitan height and weighttables. Metropolitan Life Foundation, Statistical Bulletin 1983;64(1):2–9.

10. Nutrisurvey. Available at http://www.nutrisurvey.de/. AccessedDecember 2002.

11. Moize V, Geliebter A, Gluck ME, et al. Obese patients haveinadequate protein intake related to protein intolerance up to1 year following Roux-en-Y gastric bypass. Obes Surg2003;13:23–8.

12. Blake M, Fazio V, O’Brien P. Assessment of nutrient inassociation with weight loss after gastric restrictive proceduresfor morbid obesity. Aust N Z J Surg 1991;61:195–9.

13. Wardé-Kamar J, Rogers M, Flancbaum L, et al. Calorie intake andmeal patterns up to 4 years after gastric bypass surgery. Obes Surg2004;14:1070–9.

14. Saunders R. Grazing: a high risk behavior. Obes Surg2004;14:98–102.

15. Kersten S. Mechanisms of nutritional and hormonal regulation oflipogenesis. EMBO Rep 2001;2(4):282–6.

1296 OBES SURG (2009) 19:1293–1296