Bariatric surgery is the most effective treatment for severe obesity often resulting in excess weight loss of 50–80% [
Type 2 diabetes and lack of physical activity in bariatric patients have previously been associated with decreased weight loss after surgery [
In addition, lack of physical activity has been associated with poor mental health and reduced HRQOL in the bariatric population [
To address these shortcomings in the literature, the aim of this study was to prospectively investigate the effect of objectively measured baseline type 2 diabetes and physical fitness level on mental health, HRQOL, and weight-related body image after RYGB.
The current study is part of the larger GASMITO study (the National Committee on Health Research Ethics, study protocol number HC-2009-050) that prospectively investigates psychological and physiological factors in severely obese RYGB patients with and without type 2 diabetes from the time before weight loss is started until weight stability is achieved 1.5 years after RYGB. All patients were tested at four follow-ups: (A) at baseline, prior to weight loss, (B) about two months later after a diet induced weight loss, (C) 4 months after RYGB, and, finally, (D) 18 months after RYGB. Each test session consisted of three test days which included a number of physiological tests including oral and intravenous glucose tolerance tests, 2-hour hyperinsulinemic euglycemic clamp, dual X-ray absorptiometry scan, and incremental bicycle VO2max test. Furthermore, the psychological profile of the patients was assessed using a battery of standardized questionnaires including measures of mental symptoms, HRQOL, body image, and lifestyle, which were administered at each of the four follow-ups.
In total, 40 GASMITO participants completed the psychological and physiological tests at baseline, 32 participants at follow-up (B), 31 participants at follow-up (C), and finally 23 participants at follow-up (D). The main reason for dropout was the time associated with the multiple tests days and some tests were furthermore associated with some pain or discomfort. Participants were sampled consecutively at the bariatric clinic at Hamlet Hospital or at Køge Hospital, two hospitals located in Copenhagen and suburban Copenhagen area, respectively. Participants were contacted at their first presurgical consultation at the hospital or contacted over the phone after their initial consultation and invited to an information meeting before inclusion. All GASMITO participants were informed verbally and in writing about the procedures and were asked to sign a consent form before being included in the study. Inclusion criteria were in agreement with the Danish guidelines for gastric bypass surgery prior to January 2011, which allowed individuals between 18 and 60 years of age with a BMI > 40 kg/m2 or BMI > 35 kg/m2 and obesity-related comorbidities (e.g., type 2 diabetes) to be considered for surgery. Exclusion criteria for the current study comprised heart disease, renal disease, or hyper- or hypothyreosis.
To assess mental symptoms, a short version of the Danish
To assess quality of life and general health status of the patients, we used the
Information about the patients’ background, lifestyle, and medical status was obtained using a short version of the questionnaire that was developed as part of the
The
Diabetes status was determined by oral and intravenous glucose tolerance tests. The plasma samples were cooled and centrifuged at 2000 ×g for 10 minutes before the plasma was stored at −80°C until analysis was conducted. Remission of type 2 diabetes was defined as HbA1c < 42.1 mmol mol−1, fasting glucose < 5.6 mmol L−1, and no medication.
The VO2max (mL/min/kg) is an objective measure of physical fitness level and a direct indicator of the extent of physical activity. Participants performed an incremental bicycle VO2max test on a stationary bike until exhaustion. A test was considered valid when leveling off was achieved, which was defined as unchanged oxygen uptake despite increasing work load, and when a respiratory exchange ratio greater than 1.15 was observed. Maximal oxygen uptake was calculated as the highest average uptake during 20 seconds of exercise.
Finally, a dual X-ray absorptiometry scan (DXA scan) was performed to examine the body composition and body fat distribution of the patients.
Descriptive statistics were used to investigate patient characteristics at baseline. To avoid losing information due to missing follow-up data points, we analysed the repeated data using a random intercept multilevel model, allowing intercepts to vary between subjects, with within-subjects level 1 (follow-up time) and between-subjects level 2 with the subjects GASMITO ID as grouping variable for the measures. The first-order autoregressive covariance structure was used. All analyses of the repeated data were performed using the GENLINMIXED procedure in SPSS 20.0 with the robust standard error option to accommodate nonnormality in some of the outcome variables. To investigate the effect of baseline diabetes and physical fitness level on the psychological outcome variables, the multilevel model included follow-up time, sex, age, baseline BMI, baseline VO2max, baseline diabetes, and their interaction with follow-up time as fixed effects covariates. Age, BMI, and VO2max at baseline were included as continuous variables in the interaction analyses with follow-up time.
In total, 40 bariatric patients were included in the GASMITO study, 30% (
Table
Improvement of weight, BMI, physical fitness (VO2max), mental health (SCL-90), HRQOL (SF-36), and body image (BIQ) from baseline to time point (B), from time point (B) to time point (C), and from time point (C) to time point (D). SE: standard error; GSI: General Severity Index.
Baseline |
Follow-up (B) |
Follow-up (C) |
Follow-up (D) | |
---|---|---|---|---|
|
40 | 32 | 31 | 23 |
Physical measures | ||||
Weight (kg) | 127. 8 (3.0) |
|
|
|
BMI (kg/m2) | 42.7 (0.7) |
|
|
|
VO2max (mL/min/kg) | 21.1 (0.8) | 0.8 (0.5) |
|
|
|
||||
SCL-90 | ||||
Somatization | 1.73 (0.1) | −0.11 (0.1) | −0.005 (0.1) |
|
Depression | 1.79 (0.1) | −0.05 (1.1) |
|
|
Anxiety | 1.46 (0.1) | −0.02 (0.1) |
|
−0.07 (0.1) |
GSI | 1.68 (0.1) | −0.06 (0.1) |
|
|
|
||||
SF-36 | ||||
Physical function | 68.94 (3.2) |
|
|
|
Physical role | 74.51 (4.9) | −5.27 (7.0) |
|
3.83 (2.7) |
Bodily pain | 56.25 (3.5) | 4.15 (3.7) |
|
3.75 (3.3) |
General health | 60.89 (2.9) | 5.43 (3.6) |
|
4.14 (3.0) |
Vitality | 55.10 (3.4) |
|
|
|
Social function | 88.21 (2.9) | −2.53 (3.5) | 1.42 (4.5) |
|
Mental role | 81.00 (4.7) | −1.88 (5.9) |
|
3.95 (2.9) |
Mental health | 79.74 (1.9) | −0.19 (1.8) |
|
1.23 (1.9) |
|
||||
BIQ | ||||
Weight-related body image | 2.35 (0.1) |
|
|
|
Weight and BMI improved significantly in the bariatric patients. These weight-related improvements were highly significant from baseline to follow-up (B) immediately before surgery and continued to be highly significant throughout the surgical course. At follow-up (D), 18 months after RYGB, the mean percent excess weight loss (% EWL) from baseline reached 65% (±12) and percent weight loss (% WL) was 30% (±6), indicating a successful surgical weight loss outcome [
At 18-month follow-up, the bariatric patients reported a significant reduction of mental symptoms and a significant increase in HRQOL and weight-related body image compared with baseline values, indicated by lower scores on the somatization, depression, anxiety, and GSI subscale of the SCL-90 and higher scores on the eight SF-36 subscales and the weight-related BIQ subscale (
In the applied multilevel model, the interaction between follow-up time and type 2 diabetes at baseline was found to be significant for six of the thirteen investigated subscales after adjusting for main effects of sex, age, baseline BMI, baseline VO2max, and their interaction with follow-up time. Figure
Estimated means and standard errors (bar) across follow-ups for patients with and without type 2 diabetes at baseline. Only subscales with a significant interaction effect of type 2 diabetes and follow-up are illustrated. The somatization, depression, anxiety, and General Severity Index (GSI) subscales of the “Symptoms Checklist” (SCL-90) are scored from 1 to 4 with higher scores indicating more mental distress. The physical function and physical role subscales of the “Short Form Health Survey 36” (SF-36) are scored from 1 to 100 with higher scores reflecting better health-related quality of life.
The only significant main effect of baseline diabetes was observed for the emotional role subscale of the SF-36 (
Physical fitness (VO2max) made modest contributions to variations in mental symptoms and HRQOL but not weight-related body image. Main effects of VO2max were observed for the physical function (
Estimated means and standard errors (bars) across follow-ups for patients with a baseline physical fitness level lower and higher than the sample median (VO2max = 21.7 mL/min/kg). Only significant interaction effects of physical fitness level and follow-up are illustrated. The somatization subscale of the “Symptoms Checklist” (SCL-90) is scored from 1 to 4 with higher scores indicating more mental distress. The physical role and general health subscales of the “Short Form Health Survey 36” (SF-36) are scored from 1 to 100 with higher scores reflecting better health-related quality of life.
Several significant predictors were observed in addition to baseline diabetes and baseline physical fitness. More specifically, the main effect of sex was significant for physical function (
To our knowledge, this study was the first study to prospectively investigate the importance of type 2 diabetes and physical fitness at baseline for mental health, HRQOL, and weight-related body image among bariatric patients undergoing RYGB. In line with consistent findings from prior research [
The interaction between follow-up time and baseline diabetes was significant for mental health and physical HRQOL outcomes indicating that the difference in mental symptoms and physical HRQOL between diabetic and nondiabetic patients declines across follow-ups. Diabetic patients endorsed fewer mental symptoms and had a higher HRQOL and better weight-related body image preoperatively compared with nondiabetic patients. This indicates that diabetic patients seeking RYGB may be motivated by their diabetes in contrast to nondiabetic patients that may seek out the surgical option due to substantial obesity-related distress. However, interestingly, the differences in mental symptoms and HRQOL between patients with and without diabetes at baseline resolved around the time of surgery. This also suggests that the psychological impairments in the nondiabetic patients may be related to the obese state and therefore immediately influenced by weight loss. It could be speculated that positive expectations to the operation outcome generate an experience of optimism that positively affects the level of mental symptoms and HRQOL in the nondiabetic patients around the time of surgery. In fact, improvements in mental health and HRQOL were more pronounced initially after surgery in patients that did not have preoperative diabetes compared with patients with diabetes at baseline in spite of the fact that diabetes resolved in 50% of these patients after surgery. One possible explanation is that the differences in psychological wellbeing between patients with and without diabetes reflect preoperative differences in weight-related mental health and HRQOL. Prior research has suggested that type 2 diabetes is not associated with major impairments of mental and physical functioning among bariatric candidates [
Significant increases in physical activity have consistently been reported in prior research [
Finally, numerous attempts have been made to identify demographic predictors of weight loss after bariatric surgery with conflicting results [
The current study is an important contribution to the very limited literature investigating type 2 diabetes and psychological outcome after bariatric surgery. The prospective study design and the use of objective methods to assess weight, BMI, type 2 diabetes, and physical fitness are important strengths of the study. In addition, the assessment of four psychological factors contributes to a fairly detailed psychological profile of the bariatric patients. However, some limitations should be mentioned. The relatively small sample size, questioning the power of the study, and the lack of comparison with a nonsurgical obese control group are both limitations of the study. Thus, significant findings may have been missed and, furthermore, it remains unclear whether the results are specific to the bariatric population or whether they also apply to the severely obese individuals in traditional weight loss programs. Also, though prospective, by design, the study was correlational. Thus, it is not possible to infer causality and therefore not possible to examine whether baseline type 2 diabetes or baseline fitness level has causal effects on psychological outcome after RYGB surgery. Finally, patients were followed up to 18 months to two years after their operation. However, the American Society for Bariatric Surgeons has recommended research to conduct at least 5-year follow-ups and the results of the current study should therefore be considered preliminary [
This study found significant improvements in BMI, mental health, HRQOL, weight-related body image, and physical fitness among the GASMITO patients undergoing RYGB. Psychological differences in patients with and without baseline diabetes depended on time of assessment. The diabetic group had better mental health and HRQOL on all psychological measures before surgery. However, these differences resolved around the time of surgery and no significant differences between the two groups were observed postoperatively. Thus, this study found no evidence that baseline diabetes should be considered either a positive or a negative predictor of long-term mental health and HRQOL after RYGB surgery. Furthermore, level of physical fitness (VO2max) as an indicator of physical activity was not a main predictor of mental health or HRQOL outcome. However, the literature is limited and further research is therefore highly needed. Future studies should prospectively investigate the long-term (>5-year follow-up) effects of type 2 diabetes and physical activity on mental health and HRQOL in a larger bariatric sample using objective methods to obtain information about weight, BMI, diabetes status, and physical activity. Specifically, physical activity is of importance as this factor is clinically meaningful and can potentially be modified. Research should therefore seek to clarify the most optimal type and intensity of physical activity that result in successful outcome with regard to both weight loss and psychological wellbeing after bariatric surgery.
The authors declare that there are no competing interests regarding the publication of this paper.