Obesity Surgery

, Volume 21, Issue 11, pp 1724–1730

Factors Related to Weight Loss up to 4 Years after Bariatric Surgery

Authors

    • Department of Surgery and Anatomy, University Hospital, Faculty of Medicine of Ribeirão PretoUniversity of São Paulo
  • Júlia Lopes do Amaral
    • Department of Surgery and Anatomy, University Hospital, Faculty of Medicine of Ribeirão PretoUniversity of São Paulo
  • Carla Barbosa Nonino-Borges
    • Division of Nutrition, Department of Internal Medicine, University Hospital, Faculty of Medicine of Ribeirão PretoUniversity of São Paulo
Clinical Research

DOI: 10.1007/s11695-011-0420-3

Cite this article as:
Júnior, W.S., do Amaral, J.L. & Nonino-Borges, C.B. OBES SURG (2011) 21: 1724. doi:10.1007/s11695-011-0420-3

Abstract

Background

Gastric bypass is the bariatric surgery most frequently performed in the world. It is responsible for sustainable weight loss, resolution of comorbidities, and improvement of quality of life. However, weight loss is not homogeneous, at times being insufficient in some patients. Our objective was to assess which factors were important in influencing this differentiated weight loss over a period of 4 years after surgery.

Methods

In this retrospective study, we assessed several physical, socioeconomic, behavioral, surgical, and demographic factors in morbidly obese patients submitted to Roux-en-Y gastric bypass that might influence excess weight loss over a period of 4 years after surgery. The same factors were assessed in order to characterize insufficient excess weight loss (<50% EWL).

Results

Review of the medical records of 149 patients showed that type-2 diabetes mellitus and dyslipidemia were the most important factors related to a lower EWL up to the third year. Preoperative weight loss, lower schooling, and lack of adherence to nutritional guidelines were important after 2 and 3 years. The presence of depression and lack of adherence to nutritional guidelines were the factors related to EWL of less than 50%.

Conclusions

Special attention and clarification should be provided to patients with diabetes mellitus type 2 and dyslipidemia and to patients with depression and lower schooling, since these patients tend to lose less excess weight after surgery. Multiprofessional care should also be provided so that the patients will follow nutritional guidelines more rigorously after surgery.

Keywords

Bariatric surgeryWeight lossPredictors

Introduction

In parallel to obesity, which has become the epidemic of the third millennium, the number of bariatric surgeries has also greatly increased over the last 20 years. This operation accounts for a greater and more sustainable weight loss than conventional treatments [1].

Roux-en-Y gastric bypass is the bariatric surgery most commonly performed all over the world [2, 3]. When performed at large centers, it involves low complication rates associated with a significant weight loss and a consequent improvement of comorbidities and of quality of life [1, 46].

Studies have demonstrated that most of the operated patients maintain an excess weight loss of more than 50% [7]. However, up to 15% of patients do not lose sufficient weight even when the same surgical technique is used and is performed by only one surgeon, with the same pattern of postoperative follow-up [5, 8].

Several factors may favor this failure of weight loss. The behavioral factor, with poor adherence to nutritional guidelines and little physical activity, is usually considered to be the main cause of this failure [9]. However, the literature shows that several other factors are associated with insufficient weight loss regardless of nutritional pattern. These are social, demographic, and constitutional factors, the presence of certain diseases, genetic polymorphism, and small variations in the surgical technique [921].

Although weight loss is not the single factor to be considered when assessing surgical success, the identification of variables that influence this weight loss is important in terms of patient expectations and mainly for the identification of patients who require special dedication during postoperative nutritional and psychological follow-up.

An important aspect of the present study is that it involved a more prolonged period of evaluation than most of the other studies in the literature, which report their results for 2 years, or even only 1 year after surgery.

The objective of the present study was to assess which factors were important in influencing a greater or lower excess weight loss over a follow-up period of 4 years. An attempt was also made to determine which factors were responsible for insufficient weight loss (percentage of excess weight loss (PEWL) < 50%) after 18 months. Finally, to assess the influence of the various weight-loss factors regardless of the eating pattern of the patient, the variables were compared among patients with the same eating behavior and the same adherence to the nutritional guidelines.

Materials and Methods

A retrospective study was conducted by examining the medical records of 149 patients submitted to open Roux-en-Y gastric bypass at our institution from 2000 to 2007, who maintained regular postoperative returns. In all, during this period 168 patients were operated, but 19 of them have not had regular returns. The records referring to 4 years of follow-up were reviewed.

Surgery was indicated according to the National Institute of Health consensus guidelines for treatment of the morbidly obese patient [22]. Two surgeons were responsible for all operations, which included a gastric pouch of 20 to 30 ml, almost always a sylastic ring banding this pouch (143 with band and 6 without) with 6.2 or 6.5 cm perimeter, a 50- or 100-cm biliopancreatic limb, and a 100-, 120-, or 150-cm alimentary limb. All patients were stimulated to lose weight during the preoperative period.

The observations were made on the 6th, 12th, 18th, 24th, 36th, and 48th postoperative month. To compare the variables, three types of patient division were used. The patients were first divided into those who lost weight below mean PEWL and those who lost weight above mean PEWL during the study period. For the second division, 50% PEWL was used as the cutoff point to characterize insufficient weight loss. Finally, the patients were divided in a manner similar to that used for the first division using mean PEWL, but considering only the patients with the same pattern of adherence to the nutritional guidelines in order to remove this analysis factor.

Regarding this last division, we opted to use the group of patients with poor or no adhesion to the nutritional guidelines because this group included most of the patients and therefore permitted a more reliable statistical analysis.

The following variables were analyzed: age; gender; presence of diabetes, dislipidemia, and hypertension; psychological or psychiatric diagnosis of depression; smoking; marital status (single or divorced versus married or cohabiting); schooling; preoperative weight, body mass index (BMI), excess weight, preoperative weight loss, waist circumference, and percentage of lean mass characterized by bioimpedanciometry using a Quantum BIA 101Q-RJL Systems® instrument; maximum weight; adherence to postoperative nutritional guidelines regarding the diet and sweet ingestion; pregnancy; frequency of physical activity (number of times per week spent in more than 1 h of activity); size of the biliopancreatic and alimentary limbs; presence or absence of a containment ring; perimeter of the ring used; presence of some more important postoperative complication (digestive fistula, problems with the ring, gastro-gastric fistula, vomiting during most meals, and important iron deficiency anemia).

Regarding adherence to the nutritional guidelines, the patients were divided into two groups: those who followed few or no guidelines during the postoperative period despite all the interventions of the specialists in terms of quality, quantity, or frequency of food intake, and patients who followed the nutritional guidelines properly or fully.

For statistical analysis, categorical data were analyzed by the Fisher exact test, and continuous data were analyzed by the t test for unpaired samples and the Mann–Whitney test (variables with nonparametric distribution). The level of significance was set at p < 0.05.

Results

The characteristics of the patients operated in our service during the study period were: age 41.40 ± 9.75 years (mean ± SD), preoperative weight 138.59 ± 22.48 kg, preoperative BMI 52.13 ± 7.69 kg/m2, maximum weight 153.55 ± 27.55 kg, abdominal circumference 144.35 ± 16.29 cm, and percent lean mass 51.06 ± 5.99%. One hundred twenty-one patients were women and 28 were men.

Comparison of Patients whose Excess Weight Loss Was Below or Above the Mean, Per Period

A progressive loss of excess weight was observed along the observation periods up to the second year (45%, 64%, 70%, and 73% excess weight loss at 6, 12, 18, and 24 months, respectively). Weight recovery occurred thereafter (71% and 64% at 36 and 48 months).

During the first 2 years, the time of greater weight loss, the patients who were heavier during the preoperative period (greater absolute weight up to 12 months, greater BMI, greater maximum weight, and greater abdominal circumference up to the first year) lost less excess weight (Table 1). This observation was expected, since patients weighing less have less excess weight; therefore, they experience a greater percent of excess weight loss even when they lose small amounts of weight. Actually, the opposite occurred when we evaluated percent of total weight loss, with more obese patients with a higher BMI losing a greater percentage of total weight 3 and 4 years after surgery (Table 2).
Table 1

Mean values of each variable for the patients who lost weight below (first value presented) or above (second value) the mean percent excess weight loss, per postoperative period

 

6 months n = 149 (82 vs 67) 45.9%a

12 months n = 149 (76 vs 73) 64%a

18 months n = 149 (73 vs 76) 70%a

24 months n = 146 (71 vs 65) 73%a

36 months n = 119 (58 vs 61) 71%a

48 months n = 86 (40 vs 46) 64%a

Age

41.7 vs 41.0 (ns)

42.5 vs 40.2 (ns)

42.7 vs 40.1 (ns)

42.4 vs 40.2 (ns)

43.2 vs 41.0 (ns)

42.8 vs 42.7 (ns)

Gender—female

80.4% vs 82.0% (ns)

80.2% vs 82.1% (ns)

79.4% vs 82.8% (ns)

77.4% vs 85.3% (ns)

84.4% vs 83.6% (ns)

82.5% vs 86.9% (ns)

Diabetes

34.1% vs 43.2% (ns)

39.4% vs 36.9% (ns)

46.5% vs 30.2%(0.04)

45.0% vs 29.3% (0.05)

50.0% vs 29.5% (0.02)

45.0% vs 34.7% (ns)

Dyslipidemia

21.9% vs 32.8% (ns)

35.5% vs 17.8%(0.01)

38.3% vs 15.7%(<0.01)

35.2% vs 17.3% (0.01)

36.2% vs 19.6% (0.05)

32.5% vs 21.7% (ns)

Depression

12.1% vs 13.4% (ns)

17.1% vs 8.2% (ns)

19.1% vs 6.5% (0.02)

15.4% vs 9.3% (ns)

18.9% vs 9.8% (ns)

27.5% vs 10.8% (0.05)

Hypertension

75.6% vs 65.6% (ns)

80.8% vs 64.3% (ns)

75.3% vs 67.1% (ns)

73.2% vs 68.0% (ns)

79.3% vs 72.1% (ns)

82.5% vs 76.0% (ns)

Smoking

6.0% vs 14.9% (ns)

9.2% vs 12.3% (ns)

10.9% vs 10.5% (ns)

8.4% vs 13.3% (ns)

8.6% vs 16.3% (ns)

12.5% vs 19.5% (ns)

Marital status (single/divorced)

42.6% vs 37.3% (ns)

41.3% vs 38.0% (ns)

40.2% vs 39.1% (ns)

32.8% vs 47.9% (ns)

32.1% vs 46.6% (ns)

35.0% vs 37.2% (ns)

Schooling (up to incomplete high school)

60.6% vs 43.4% (ns)

54.3% vs 52% (ns)

59.6% vs 46.0% (ns)

64.1% vs 42.3% (0.03)

68.8% vs 43.9% (0.02)

65.6% vs 48.2% (ns)

Pre-op weight (kg)

144.1 vs 131.7(<0.01)

142.9 vs 134.0 (0.01)

141.4 vs 135.8 (ns)

142.0 vs 135.9 (ns)

137.5 vs 140.5 (ns)

133.7 vs 139.6 (ns)

Pre-op BMI (kg/m2)

54.5 vs 49.1 (<0.01)

54.3 vs 49.8 (<0.01)

53.6 vs 50.6 (0.01)

53.5 vs 50.9 (0.03)

52.7 vs 52.2 (ns)

50.5 vs 52.6 (ns)

Pre-op excess weight (kg)

82.4 vs 69.4 (<0.01)

81.5 vs 71.4 (<0.01)

79.8 vs 73.4 (0.05)

80.1 vs 73.7 (0.05)

76.6 vs 77.9 (ns)

72.1 vs 78.0 (ns)

Maximum weight (kg)

160.9 vs 144.5 (<0.01)

159.1 vs 147.7 (0.01)

158.7 vs 148.5 (0.02)

160.6 vs 147.3 (<0.01)

155.2 vs 152.4 (ns)

150.9 vs 152.7 (ns)

Pre-op weight loss (kg)

16.3 vs 12.7 (0.02)

15.7 vs 13.7 (ns)

17.6 vs 13.5 (0.05)

18.2 vs 11.5 (<0.01)

17.1 vs 11.8 (0.03)

16.7 vs 13.0 (ns)

% Pre-op weight loss (%)

9.6% vs 8.3% (ns)

9.3% vs 8.8% (ns)

9.9% vs 8.2% (ns)

10.7%vs7.4%(<0.01)

10.3% vs 7.3%(0.01)

10.3% vs 7.7% (ns)

Waist circumf. (cm)

148.2 vs 139.5 (0.01)

148.2 vs 140.2 (0.02)

146.8 vs 142.0 (ns)

146.8 vs 141.9 (ns)

145.8 vs 143.9 (ns)

141.2 vs 144.7 (ns)

% Lean mass

50.1% vs 52.0% (ns)

50.2% vs 51.8% (ns)

50.8% vs 51.2% (ns)

50.9% vs 50.9% (ns)

50.3% vs 49.8% (ns)

49.6% vs 48.4% (ns)

Pregnancy after

4.5% vs 7.2% (ns)

4.9% vs 6.6% (ns)

5.1% vs 6.3% (ns)

5.4% vs 6.2% (ns)

4.0% vs 7.8% (ns)

6.0% vs 2.5% (ns)

Diet adequacy (none or poor)

67.0% vs 59.7% (ns)

67.5% vs 58.3% (ns)

70.4% vs 56.0% (ns)

72.4% vs 55.4% (0.03)

76.7% vs 59.0% (0.04)

81.0% vs 71.4% (ns)

Sweet eater

56.0% vs 67.1% (ns)

57.8% vs 57.5% (ns)

61.6% vs 60.5% (ns)

69.0% 52.7% (ns)

72.4% vs 59.0% (ns)

77.5% vs 69.5% (ns)

Physical activity (≤2 days/week)

46.3% vs 44.7% (ns)

51.3% vs 39.7% (ns)

47.9% vs 43.2% (ns)

43.6% vs 48.0% (ns)

50.0% vs 52.4% (ns)

60.0% vs 54.3% (ns)

Biliopancreatic limb (50 cm)

52.4% vs 64.1% (ns)

56.5% vs 58.9% (ns)

56.1% vs 59.2% (ns)

56.3% vs 58.6% (ns)

62.0% vs 67.2% (ns)

80.0% vs 76.0% (ns)

Alimentary limb (100 cm)

78.0% vs 91.0% (0.04)

80.2% vs 87.6% (ns)

83.5% vs 84.2% (ns)

81.6% vs 85.3% (ns)

81.0% vs 83.6% (ns)

82.5% vs 86.9% (ns)

Presence of band

77(93.9%)vs66(98.5%) (ns)

73(96.0%)vs70(95.8%) (ns)

70(95.8%)vs73(96.0%) (ns)

67(94.3%)vs63(97.3%) (ns)

56(96.5%)vs59(96.7%) (ns)

38(95.0%)vs45(97.8%) (ns)

Perimeter (6.2 cm)

86.5% vs 85.0% (ns)

86.8% vs 84.9% (ns)

84.9% vs 86.8% (ns)

83.0% vs 88.0% (ns)

82.7% vs 86.8% (ns)

78.3% vs 86.9% (ns)

Complications

17.0% vs 22.3% (ns)

18.4% vs 20.5% (ns)

17.8% vs 21.0% (ns)

16.9% vs 22.6% (ns)

20.6% vs 24.5% (ns)

22.5% vs 23.9% (ns)

ns no significance

aMean % excess weight loss

Table 2

Mean values of each variable for the patients who lost weight below (first value presented) or above (second value) the mean percent total weight loss, per postoperative period

 

6 months n = 149 84 vs 65) 24.7%a

12 months n = 149 (71 vs 78) 34.6%a

18 months n = 149 (71 vs 78) 37.9%a

24 months n = 146 (68 vs 78) 38.2%a

36 months n = 119 (53 vs 66) 37.3%a

48 months n = 86 (39 vs 47) 34.8%a

Pre-op weight (kg)

139.3 vs 137.6 (ns)

137.4 vs 139.6 (ns)

135.5 vs 141.3 (ns)

135.9 vs 141.0 (ns)

132.4vs143.7(<0.01)

128.6vs143.6(<0.01)

Pre-op BMI (kg/m2)

52.4 vs 51.7 (ns)

52.0 vs 52.2 (ns)

51.2 vs 52.9 (ns)

50.8 vs 53.2 (ns)

50.0 vs 54.2 (<0.01)

48.4 vs 54.2 (<0.01)

Maximum weight (kg)

155.1 vs 151.5 (ns)

154.3 vs 152.8 (ns)

151.3 vs 155.5 (ns)

151.1 vs 155.8 (ns)

147.3 vs 158.0 (0.03)

143.3 vs 158.6 (0.01)

ns no significance

aMean % total weight loss

The patients who lost more absolute weight during the preoperative period lost lower percentage of excess weight during the postoperative period up to the third year. However, percent of preoperative weight loss was significant in determining lower excess weight loss only by 24 and 36 months after surgery (Table 1).

Type 2 diabetes mellitus and dyslipidemia were important factors also causing a smaller weight loss starting at 18 and 12 months, respectively (Table 1).

As isolated factors, lower schooling and poor adherence to nutritional guidelines during the postoperative period were predictors of a smaller excess weight loss at 24 and 36 months. Although statistical significance was observed only at 18 months, the presence of depression, in absolute numbers, was more frequent among the patients who lost less weight (Table 1).

Comparison of Patients who Lost Less or More than 50% Excess Weight (Insufficient Loss) Per Period

The presence of depression and low adherence to the nutritional guidelines were factors responsible for insufficient weight loss for a longer period of time (18 to 36 months). In an isolated manner, more advanced age was a predictive factor of insufficient weight loss at 18 months (Table 3).
Table 3

Mean values of each variable for the patients who lost weight below (first value presented) or above (second value) 50% of excess weight, per postoperative period

 

18 months n = 149 (18 vs 131) 12.0%a

24 months n = 146 (17 vs 132) 12.8%a

36 months n = 119 (18 vs 101) 15.1%a

48 months n = 86 (16 vs 70) 18.6%a

Age

45.7 vs 40.8 (0.04)

44.3 vs 40.95 (ns)

44.7 vs 41.6 (ns)

43.8 vs 42.5 (ns)

Gender—female

77.7% vs 81.6% (ns)

86.6% vs 80.9% (ns)

83.3% vs 84.1% (ns)

87.5% vs 84.2% (ns)

Diabetes

44.4% vs 37.4% (ns)

33.3% vs 37.4% (ns)

44.4% vs 38.6% (ns)

37.5% vs 40.0% (ns)

Dyslipidemia

44.4% vs 24.4% (ns)

40.0% vs 24.4% (ns)

33.3% vs 26.7% (ns)

37.5% vs 24.8% (ns)

Depression

38.8% vs 9.1% (<0.01)

40.0% vs 9.1% (<0.01)

33.3% vs 10.8% (0.02)

37.5% vs 14.2% (ns)

Hypertension

88.8% vs 68.7% (ns)

86.6% vs 68.7% (ns)

83.3% vs 74.2% (ns)

93.7% vs 75.7% (ns)

Smoking

11.1% vs 10.6% (ns)

20.0% vs 9.92% (ns)

11.1% vs 13.8% (ns)

12.5% vs 17.1% (ns)

Marital status (single/divorced)

33.3% vs 39.6% (ns)

33.3% vs 41.4% (ns)

22.2% vs 42.8% (ns)

26.6% vs 37.3% (ns)

Schooling (up to incomplete high school)

58.8% vs 52.2% (ns)

71.4% vs 50.5% (ns)

64.7% vs 55.0% (ns)

57.1% vs 57.4% (ns)

Pre-op weight (kg)

144.0 vs 137.8 (ns)

143.6 vs 138.3 (ns)

136.1 vs 139.6 (ns)

128.2 vs 138.8 (ns)

Pre-op BMI (kg/m2)

54.2 vs 51.8 (ns)

54.3 vs 51.9 (ns)

51.9 vs 52.5 (ns)

48.7 vs 52.3 (ns)

Pre-op excess weight (kg)

82.3 vs 75.7 (ns)

82.3 vs 76.1 (ns)

75.1 vs 77.6 (ns)

67.2 vs 77.1 (0.05)

Maximum weight (kg)

163.2 vs 152.2 (ns)

163.2 vs 152.7 (ns)

152.3 vs 154.0 (ns)

143.9 vs 153.7 (ns)

Pre-op weight loss (kg)

18.7 vs 14.2 (0.02)

19.1 vs 14.2 (<0.01)

15.6 vs 14.2 (ns)

15.0 vs 14.6 (ns)

% Pre-op weight loss (%)

11.05% vs 8.82% (ns)

11.54% vs 8.80% (ns)

9.81% vs 8.67% (ns)

9.90% vs 8.75% (ns)

Waist circumf. (cm)

148.0 vs 143.7 (ns)

142.3 vs 144.6 (ns)

143.1 vs 145.0 (ns)

141.7 vs 143.3 (ns)

% Lean mass

48.6% vs 51.3% (ns)

46.7% vs 51.3% (ns)

48.0% vs 50.3% (ns)

49.3% vs 48.9% (ns)

Pregnancy after

7.1% vs 5.6% (ns)

7.6% vs 5.6% (ns)

6.6% vs 5.8% (ns)

0% vs 5.0% (ns)

Diet adequacy (none or poor)

88.2% vs 59.6%(<0.01)

85.7% vs 61.2%(0.05)

94.1% vs 63.0% (0.01)

86.6% vs 68.1% (ns)

Sweet eater

77.7% vs 58.7% (ns)

73.3% vs 58.7% (ns)

72.2% vs 64.3% (ns)

81.2% vs 71.4% (ns)

Physical activity (≤2 days/week)

61.1% vs 43.5% (ns)

66.6% vs 43.5% (ns)

72.2% vs 47.5% (ns)

68.7% vs 54.2% (ns)

Biliopancreatic limb (50 cm)

55.5% vs 58.0% (ns)

46.6% vs 58.7% (ns)

55.5% vs 66.3% (ns)

75% vs 78.5% (ns)

Alimentary limb (100 cm)

66.6% vs 86.2% (0.04)

73.3% vs 84.7% (ns)

72.2% vs 84.1% (ns)

81.2% vs 86.9% (ns)

Presence of band

94.4% vs 96.1% (ns)

93.3% vs 96.1% (ns)

94.4% vs 97.0% (ns)

100% vs 95.7% (ns)

Band perimeter (6.2 cm)

88.8% vs 85.4% (ns)

93.3% vs 84.7% (ns)

94.4% vs 83.1% (ns)

93.7% vs 80.5% (ns)

Complications

16.6% vs 19.8% (ns)

20.0% vs 19.8% (ns)

16.6% vs 23.7% (ns)

31.2% vs 21.4% (ns)

ns no significance

aPercentage of patients that lost <50% of excess weight

Comparison of Patients with Poor/No Adherence to Nutritional Guidelines during the Postoperative Period who Lost Excess Weight Below or Above the Mean, Per Period

After exclusion of the eating behavior variable, diabetes mellitus type 2 and dyslipidemia were the most important factors differentiating an excess weight loss below the mean in the second and third years (Table 4).
Table 4

Mean values of each variable for patients with poor/no adherence to postoperative nutritional guidelines who lost weight below (first value presented) or above (second value) the mean excess weight loss, per postoperative period

 

24 months n = 91 (50 vs 41) 73%a

36 months n = 79 (43 vs 36) 71%a

48 months n = 58 (30 vs 28) 64%a

Age

41.7 vs 39.3 (ns)

42.7 vs 39.7 (ns)

42.3 vs 41.4 (ns)

Gender—female

74.0% vs 87.8% (ns)

81.3% vs 83.3% (ns)

80.0% vs 85.7% (ns)

Diabetes

44.0% vs 21.9% (0.04)

46.5% vs 22.2% (0.03)

43.3% vs 21.4% (ns)

Dislipidemia

38.0% vs 9.75% (<0.01)

39.5% vs 13.8% (0.01)

33.3% vs 14.2% (ns)

Depression

16.0% vs 12.1% (ns)

20.9% vs 13.8% (ns)

30.0% vs 10.7% (ns)

Hypertension

76.0% vs 75.6% (ns)

79.0% vs 77.7% (ns)

83.3% vs 82.1% (ns)

Smoking

10.0% vs 17.0% (ns)

11.6% vs 19.4% (ns)

13.3% vs 25.0% (ns)

Marital status (single/divorced)

34.0% vs 50.0% (ns)

38.0% vs 47.2% (ns)

40.0% vs 48.1% (ns)

Schooling (up to incomplete high school)

63.1% vs 48.1% (ns)

62.8% vs 52.1% (ns)

60.0% vs 56.2% (ns)

Pre-op weight (kg)

143.1 vs 138.8 (ns)

138.0 vs 144.1 (ns)

133.5 vs 142.7 (ns)

Pre-op BMI (kg/m2)

53.4 vs 51.4 (ns)

52.0 vs 53.1 (ns)

49.6 vs 52.3 (ns)

Pre-op excess weight (kg)

80.5 vs 76.0 (ns)

76.2 vs 80.9 (ns)

71.0 vs 79.5 (ns)

Maximum weight (kg)

160.2 vs 150.0 (ns)

154.1 vs 156.3 (ns)

150.4 vs 153.6 (ns)

Pre-op weight loss (kg)

16.7 vs 12.3(0.02)

15.7 vs 13.5 (ns)

17.4 vs 11.6 (ns)

Waist circunf. (cm)

146.1 vs 140.9 (ns)

145.7 vs 142.8 (ns)

140.0 vs 142.6 (ns)

% Lean mass

50.9% vs 50.5% (ns)

50.6% vs 49.5% (ns)

49.5% vs 49.0% (ns)

Pregnancy after

8.1% vs 5.5% (ns)

5.7% vs 6.6% (ns)

4.1% vs 4.1% (ns)

Physical activity (≤2 days/week)

48.0% vs 56.0% (ns)

53.4% vs 58.3% (ns)

63.3% vs 60.7% (ns)

Biliopancreatic limb (50 cm)

60.0% ns 65.8% (ns)

57.1% vs 72.2% (ns)

83.3% vs 82.1% (ns)

Alimentary limb (100 cm)

86.0% vs 85.0% (ns)

88.3% vs 82.5% (ns)

86.6% vs 85.7% (ns)

Presence of band

94.0% vs 95.1% (ns)

95.3% vs 94.4% (ns)

93.3% vs 96.4% (ns)

Band perimeter (6.2 cm)

78.0% vs 82.9% (ns)

79.0% vs 80.5% (ns)

66.6% vs 82.1% (ns)

Complications

14.0% vs 24.3% (ns)

18.6% vs 19.4% (ns)

16.6% vs 25.0% (ns)

ns no significance

aMean excess weight loss

Discussion

The main objectives of bariatric surgery are to promote a significant and sustainable weight loss, to improve or resolve comorbidities, and to promote a better quality of life, with low rates of perioperative and long-term complications. However, weight loss is not homogeneous in this population, even with technical standardization of the surgery. Even worse, some patients lose weight in an insufficient manner. Divergent reports have been published in the literature regarding the determination of the cutoff point for defining insufficient weight loss. Thus, in the present study, we considered losses of less than 50% excess weight [23, 24].

Adherence to nutritional guidelines after surgery and the practice of physical activities are fundamental factors for guaranteeing sufficient weight loss maintenance over the years. Inadequate eating behaviors and compulsory eating disorders must be corrected in order to obtain a satisfactory weight loss. However, even patients who follow these recommendations may lose less weight due to other factors contributing to this occurrence.

We observed that the presence of diabetes mellitus type 2 and of dyslipidemia was related to a lower excess weight loss after surgery both in overall evaluation and in the evaluation in which the adherence to nutritional guidelines was removed. However, these diseases were not related to insufficient weight loss, as also reported by others [11, 18, 23].

In an attempt to explain the fact that diabetes is related to a lower weight loss, investigators mention the fact that these patients usually take medications for the control of hyperglycemia (insulin or others), which somehow increase the circulating levels of insulin or the sensitivity to this hormone. Insulin is an anabolic hormone that promotes lipogenesis, adipocyte differentiation, and muscle synthesis, a fact that may explain this differentiated weight loss [2527].

Regarding the variations of the surgical techniques, there is some controversy in the literature when the differential weight loss is evaluated. The size of the gastric pouch is an example of this divergence [11, 28]. Although this was a retrospective study, we did not observe a significant difference in weight loss regarding the size of the alimentary and biliopancreatic limbs, as also reported by Christou [8]. However, other studies have demonstrated that the length of the alimentary limb is important for weight loss. These last studies, however, assessed short-term results [29, 30]. The evaluation of the presence or absence of a containment ring is hampered by the small number of cases in which we did not use this device.

Patients who were heavier during the preoperative period and who had a higher BMI lost less excess weight after surgery, a finding that agrees with literature data [11, 14, 18]. However, it should be pointed out that, as done here, all other studies use the value of excess weight loss. Patients weighting less have less excess weight; therefore, even when they lose small amounts of weight, their percent excess weight loss is higher. By assessing percent of total weight loss and not excess weight loss, we obtained an exactly opposite result, i.e., heavier patients with higher BMI during the preoperative period lost more weight after surgery. Thus, it would be a mistake to use this variable (excess weight loss) to assess the results, as already proposed by van der Laar et al. [31].

Regarding preoperative weight loss, Livhits et al. reported the results of a meta-analysis which showed that in most of the studies reviewed, the patients who lose more weight preoperatively are those who will lose more excess weight by 12 months after surgery. After this period, the results are no longer homogeneous [21]. In the present study, we observed the opposite, i.e., the patients who lost more weight before surgery were those who lost less excess weight by 2 and 3 years after surgery, with no significance at any other period of evaluation. Two possible explanations for this divergence can be proposed. First, preoperative weight loss was determined in a nonhomogeneous manner in the obese population studied, based on the perception of the surgeon regarding how much weight each patient should lose according to biotype and estimated fat content. Thus, more obese patients were stimulated to lose more than 10% their initial weight before being operated. Second, in contrast to the cited studies in which the preoperative weight loss was obtained in an acute manner during the weeks preceding surgery, in our patients the loss was achieved by means of a multiprofessional effort lasting several months. Thus, some of our patients lost much weight, up to 60 kg, and some were operated with up to BMI below 37 kg/m2, suggesting that the organism tries to establish and equilibrium during weight loss, taking into consideration all that was lost during the preoperative period.

In an attempt to justify this hypothesis, we may cite the study by Scopinaro et al. and by other authors who demonstrated that patients with lower BMI, when submitted to bariatric surgery, lose less weight than more obese patients and, therefore, do not present malnutrition, stabilizing their weight close to a BMI of 25 kg/m2 [32].

We assessed some socioeconomic aspects that might influence weight loss. Starting 2 years after surgery, single patients lost a little more weight than married patients, although the difference was not statistically significant, as previously demonstrated by Lutfi et al. [14]. After 24 and 36 months, patients with lower schooling lost a smaller percentage of excess weight.

An important factor that influenced insufficient weight loss was the presence of depression, diagnosed by the psychologist or psychiatrist of the group. These patients also had a bad adherence to the nutritional guidelines despite all the special care dedicated to them. Lutfi et al. [14] and Thonney et al [20] showed the opposite.

Thus, we conclude that the most important factors predicting a lower excess weight loss starting 1 year after surgery were the presence of type 2 diabetes mellitus and dyslipidemia, even when the groups were standardized in terms of adherence to nutritional guidelines. Patients who were heavier during the preoperative period and with higher body mass indices lost more percentage of absolute weight. After a period of 2 years, poor adherence to nutritional guidelines, a greater percentage of preoperative weight loss, and lower schooling were additional factors that predicted a lower excess weight loss. The two most important factors for the characterization of insufficient weight loss were the presence of depression and non-adherence to nutritional guidelines after surgery. A secondary conclusion of this study was that the way of presenting postoperative weight loss results in the various studies should be reevaluated. Reporting excess weight loss may lead to inappropriate conclusions.

Conflict of Interest

The authors declare that they have no conflict of interest.

Copyright information

© Springer Science + Business Media, LLC 2011