Gastrointestinal Symptoms in Relation to Quality of Life After Metabolic Surgery in Adolescents

Abstract

Background

Adolescent obesity is associated with significant comorbidities including decreased quality of life (QOL). QOL improves after metabolic and bariatric surgery (MBS), but recent studies have demonstrated that certain gastrointestinal symptoms (GIS) worsen after surgery, including reflux symptoms, nausea, bloating and diarrhea.

Objective

The aim of this study was to evaluate QOL and the effect of these symptoms on QOL after bariatric surgery.

Setting

Five academic centers performing adolescent MBS in the United States.

Methods

We prospectively studied 228 adolescents undergoing MBS using the Teen-Longitudinal Assessment of Bariatric Surgery (Teen-LABS) cohort. Gastrointestinal symptoms and QOL scores were assessed before surgery, at 6-months and yearly to 5 years after surgery. Analysis involved linear models examining QOL and the association between GIS and QOL adjusting for a priori determined covariates.

Results

Adjusting for body mass index (BMI) change over time, the physical component score (PCS) of the Short Form-36 (SF-36) increased after surgery from 44.2 at baseline to 54.4 at 5 years (p<0.0001). The mental component score (MCS) did not significantly change over time. The SF-36 domains that showed the biggest increase after surgery were physical functioning, physical role functioning, and general health. The SF-36 PCS decreased significantly over time post-surgery in those with GIS of reflux, nausea, and diarrhea but remained higher than baseline SF-36 PCS. There was no statistically significant change in MCS or IWQOL-KIDS scores in those with or without GIS.

Conclusion

QOL, specifically the SF-36 PCS, increases after MBS. Reflux symptoms, nausea, and diarrhea reduce the degree of improvement in QOL in adolescents following MBS. Patients should be monitored and treated for these symptoms to address this decreased QOL.

INTRODUCTION

Childhood and adolescent obesity is increasing in prevalence and has far reaching psychological and physiological effects including lower quality of life, depression, obstructive sleep apnea, dyslipidemia, and diabetes^(1-3). Metabolic and bariatric surgery (MBS) has proved to be an effective therapy for adolescent obesity, specifically vertical sleeve gastrectomy (VSG) and Roux-en-Y Gastric bypass (RYGB)⁽²⁾. However, the long-term consequences of surgery are still under investigation. Previous work has demonstrated that overall quality of life (QOL) as well as weight related quality of life (WRQOL) is improved after surgery ^(2,4-8).

Although it is known that weight loss after MBS in adolescents is associated with improved health related QOL, it is unknown how other post-surgical factors affect health related QOL, specifically gastrointestinal symptoms after surgery. This is important, as gastrointestinal symptoms such as nausea, bloating, diarrhea, and reflux symptoms have been shown to increase after MBS in adolescents^{(9) (10,11)}. Comparing gastrointestinal symptoms between sleeve gastrectomy and RYGB, reflux symptoms occur more frequently after sleeve gastrectomy, which mirrors adult studies ⁽¹²⁾. During the development of an adult bariatric QOL tool, one study found that as weight loss plateaus, gastrointestinal symptoms more strongly affect QOL⁽¹³⁾. There are no data regarding how gastrointestinal symptoms affect QOL in adolescents. The aim of this study was to evaluate how health and weight related QOL changes and how gastrointestinal symptoms affect quality of life after bariatric surgery in adolescents. We hypothesized that health and weight related QOL would improve after bariatric surgery but that gastrointestinal symptoms after metabolic surgery would mitigate those improvements relative to mild or no concomitant symptoms.

METHODS

Study Design and Subjects

Methodology and design of the Teen-Longitudinal Assessment of Bariatric Surgery (Teen-LABS; NCT00465829) study have been described previously.⁽¹⁴⁾ Consecutive adolescents (<19 years of age and Tanner Stage 4 or greater) with a body mass index (BMI) ≥ 35kg/m² who met eligibility guidelines for MBS were offered enrollment into Teen-LABS at one of five U.S. centers (February 28, 2007- December 30, 2011). The study protocol, data and safety monitoring plans, and assent and/or consent forms were approved by all institutional review boards and also an independent data and safety monitoring board convened by the National Institutes of Health (NIH). Participants who underwent adjustable gastric band (AGB) insertion (n=14) were excluded from the current analysis, resulting in 228 patients who underwent vertical sleeve gastrectomy (VSG; n=67) or Roux-en-Y gastric bypass (RYGB; n=161).

Data Collection

Data were collected at baseline (≤30 days prior to surgery), 6 months, 1 year, 2 years, 3 years, 4 years, and 5 years post-operatively. The majority of follow-up research visits and associated data collection occurred at Teen-LABS centers (89%), while the remaining visits were conducted by trained field examiners at participants’ homes or via telephone and/or electronic communication. The completion percentage for the GI symptoms and QoL was 82% overall at 5 years (137/161 (85%) for RYGB patients and 49/67 (73%) for VSG patients) and 84% over the whole study period overall; 86% for RYGB and 78% for VSG.

Data Points and Definitions

Demographic data:

Study participants reported their age (years) at the time of surgery, race (white or non-white), sex (male or female), and household income.

Weight Loss:

Weight and height were measured, and BMI calculated at each pre- and post-operative visit. Percent change in BMI from baseline was calculated for each visit.

Early post-operative complications:

Complications within 30-days post-surgery were classified as major, minor or none. Major complications were re-operation due to bowel obstruction, gastrointestinal leak or sepsis, post-operative transfusion, anticoagulant use for DVT or PE, unplanned splenectomy or suicidal ideation. Minor complications were gastrointestinal leak (minimal), postoperative bleeding, atelectasis/pneumonia, urinary tract events, bowel injury, solid organ injury, mesenteric bleeding, over-sedation, hypertension, TPN after discharge, gastrojejunal anastomotic stricture, wound infection, ileus obstruction, abdominal pain/dehydration, or acute pancreatitis.

Gastrointestinal Symptoms:

Participants completed a 15-item gastrointestinal symptom (GIS) rating scale⁽¹⁵⁾ before surgery and at all post-operative study visits for 5 years. Potential responses for each symptom ranged from none through very severe. For clinical relevance and analyses, we considered symptoms reported as “moderate”, “moderate/severe”, “severe” and “very severe” as symptom present, and symptoms reported as “none”, “minor” and “mild” as symptom absent. Symptoms which significantly worsened over time in previous work⁽¹¹⁾ were selected for further analysis: gastroesophageal reflux symptoms (GERS), nausea, bloating, and diarrhea.

Quality of Life Metrics:

Two QOL measures were used, including one that captured weight related QOL (IWQOL-Kids) and one that captured general health related QOL (SF-36). IWQOL-Kids is a 27-item instrument ⁽¹⁶⁾ designed to assess aspects of weight-related quality of life in adolescents, aged 11–19 years. Factor analyses identified four factors, including Physical Comfort, Body Esteem, Social Life, and Family Relations, as well as a Total score. This measure has strong psychometric properties, discriminates among weight status groups, and is responsive to weight change⁽¹⁶⁾. Scaled scores were standardized and range from 0 to 100, with higher scores representing better weight-related quality of life.

The short-form (SF)-36⁽¹⁷⁾ is a 36 item scale that assesses eight health concepts (none of which are specific to weight status or obesity): 1) physical functioning; 2) physical role functioning 3) mental health; 4) bodily pain; 5) social functioning; 6) emotional role functioning; 7) vitality; and 8) general health perceptions. The survey was constructed for self-administration by persons 14 years of age and older, and for administration by a trained interviewer in person or by telephone. Scores range from 0 to 100, with higher scores representing better QOL.

Statistical Methods

Descriptive statistics were reported as mean ± standard deviation (SD) or standard error of the mean (SEM) for continuous variables or frequency and percentage for categorical variables. Distributional properties of continuous variables were assessed, and potential outliers examined. The minimal clinically important difference (MCID) for the SF-36 PCS and MCS was established at 5 points as per previous studies⁽¹⁸⁾.

Analysis involved mixed linear models to examine the two measures of QOL and their components at 6-months and yearly at 1-year through 5-years post-surgery. QOL measures were adjusted for percent BMI change from baseline. Presence of gastrointestinal symptoms was the independent variable of interest. Specific gastrointestinal symptoms examined included GERS (defined as presence of either heartburn or acid reflux), nausea, bloating and diarrhea as previous literature has demonstrated that these symptoms significantly increase over time after bariatric surgery⁽¹¹⁾. Covariates included in the initial model were baseline values for: 30-day post-operative complications, income, type of surgery, sleep apnea, diabetes mellitus, sex, race, and pre-surgery QOL scores. Visit number (representing time since surgery), percent change in BMI from baseline and Beck Depression Inventory (BDI) were included as time-dependent covariates. These covariates were selected based on previous literature reporting effects of these factors on QOL^(19-21). The initial model included all covariates, backward elimination was used and only those covariates significant at p<0.05 were retained in the model. To examine if the association between QOL and severity of symptoms over time, the interaction between visit number and symptom severity was examined. If the interaction was significant, it was included in the model and also analyses were examined by visit. SAS®, version 9.4 (SAS Institute, Cary, NC) was used for analysis. A p-value of <0.05 was considered statistically significant unless stated otherwise.

RESULTS

Cohort Characteristics

The mean age at surgery was 16.5 ± 1.6 years. The majority of participants were female (75.0%) and white (72.0%). The baseline BMI was 52.6 ± 9.4 kg/m²(Table 1).

Table 1:

Baseline Demographics of Cohort

	Overall
N	228
Age at Surgery (years)	16.5 (1.6)
Sex (Female)	171 (75.0)
Race:
White	164 (72.0)
Black	50 (22.0)
Asian	2 (1.0)
More than one race	12 (5.0)
Household income:
<$25,000	87 (38.0)
$25,000-$49,999	46 (20.0)
$50,000-$74,999	40 (17.5)
≥$75,000	55 (24.1)
BMI (kg/m2)	52.6 (9.4)
Baseline QOL
IWQOL	61.1 (18)
SF-36
MCS	48.7 (10.9)
PCS	44.8 (9.6)

BMI: body mass index; SF-36: short form 36; IWQOL: impact of weight on quality of life; MCS: mental component score; PCS: physical component score

Data are presented as mean (standard deviation) or n (%)

Baseline mean QOL total scores were 61.1 ± 18 for IWQOL-KIDS, 48.7 ± 10.9 for the mental component score (MCS) of the SF-36, and 44.8 ± 9.6 for the physical component score (PCS) of the SF-36. In terms of the specific components of the SF-36, the mean scores were as follows: 61.1 for physical functioning, 69.4 for physical role functioning, 80.1 for emotional role functioning, 52.4 for vitality, 72.3 for mental health, 74.5 for social function, and 49.2 for general health (Table 1). Examining individual SF-36 questions, the lowest mean baseline scores were for the following questions/statements: “My health is excellent” (31.4), health limiting vigorous activity (34.2), “I am as healthy as anybody I know” (36.7), and status of general health (35.7).

Quality of Life After MBS

At the 6 month visit after surgery, the mean SF-36 MCS increased to 51.1 from 49.4, and the mean PCS increased to 53.3 from 44.2. Adjusting for percent BMI change, baseline MCS, sex, race, post-operative complications, income, and BDI score, the SF-36 MCS did not significantly change over time in the overall cohort or when comparing surgery type (RYGB and VSG). Adjusting for percent BMI change, baseline PCS, sex, post-operative complications, and BDI score, the SF-36 PCS did significantly increase over time in the overall cohort (p<0.0001) and in the RYGB subgroup (p<0.0001) (Figure 2). The % weight loss between the RYGB (28%) and VSG (26%) subgroups were similar.

Figure 2 a and b:

SF-36 PCS and MCS over time after MBS by type of surgery (RYGB versus VSG)

The largest increases in mean SF-36 component scores from baseline to the 6-month visit were seen in the general health score (49.1 to 68.7), vitality score (52.4 to 67.9), physical functioning (61.1 to 85.8), and physical role functioning (69.4 to 91.8). Scores were stable over time with the exception of vitality which decreased by the 60-month post-operative visit (Figure 3, Supplement Table A). The MCID was exceeded for all component scores of the SF-36 after MBS.

Figure 3:

Mean SF-36 component scores at baseline and over time after MBS

For the majority of individual SF-36 questions, there was an increase in score after baseline that persisted through 5 years after surgery with the exception of the question asking, “compared to one year ago, how would you rate your health now?” This question’s score initially increased from 56.1 at baseline to 95 at 6 months and then had a downward trend to 62.0 at 5 years post operatively (Supplement Table B).

Quality of Life in Relation to Gastrointestinal Symptoms

There was no overall significant difference in IWQOL-KIDS between those with and without symptoms after accounting for sex, race, 30-day post-operative symptoms, income BDI and percent change in BMI for GERS, nausea, bloating, or diarrhea (Table 2). The subscales of the IWQOL-KIDS measure were also examined for GERS, nausea, and diarrhea, and there were no significant differences between those with and without symptoms (Supplement, Table C). Those with bloating symptoms did have significantly lower IWQOL-KIDS score for the subscale of social life (−8.3 (−15.8, −0.8)) at 48 months after MBS (p=0.03), but no other subscales demonstrated a significant difference for those with bloating symptoms (Supplement, Table B).

Table 2:

Linear Regression Estimates for the Effect of Moderate GI symptoms of GERS, Nausea, Bloating, or Diarrhea and on QOL (IWQOL, SF-36 PCS and MCS).

	Coefficient (95% CI) for symptoms	P-value
IWQOL1
GERS	−0.9 (−3.3, 1.5)	0.5
Nausea	1.7 (−0.5, 3.9)	0.1
Bloating	0.2 (−2.2, 2.6)	0.9
Diarrhea	0.9 (−2.4, 4.2)	0.6
SF-36 MCS2
GERS	−1.1 (−2.8, 0.5)	0.2
Nausea	−0.1 (−1.7, 1.4)	0.9
Bloating	−1.0 (−2.6, 0.7)	0.3
Diarrhea	−1.4 (−10.1, 7.3)	0.8
SF-36 PCS3
GERS	−3.1 (−4.5, −1.7)	<0.0001
Nausea*
12 months	−2.7 (−7.2, 1.8)	0.2
24 months	−2.5 (−7.0, 2.1)	0.3
36 months	−6.7 (−10.8, −2.7)	0.004
48 months	−7.2 (−10.6, −5.5)	0.0004
60 months	−1.9 (−6.1, 2.2)	0.4
Bloating	−1.1 (−2.5, 0.4)	0.2
Diarrhea*
12 months	−2.9 (−11.9, 6.2)	0.5
24 months	−8.3 (−12.2, −4.5)	0.007
36 months	−6.7 (−11.8, −1.7)	<0.0001
48 months	−4.4 (−12.0, 4.1)	0.3
60 months	5.0 (−3.5, 13.4)	0.3

GI: gastrointestinal; QOL: quality of life; PCS: physical component score; MCS: mental component score; SF-36: short form 36; MBS: metabolic and bariatric surgery; IWQOL: impact of weight on quality of life; GERS: gastroesophageal reflux symptoms

^*There was a significant interaction between visit and GI symptoms, so estimates by visit are also reported adjusting for sex, race, 30-day postoperative symptoms, income, surgery type, BDI score and percent weight change

The SF-36 MCS demonstrated no overall statistically significant difference in change of QOL scores between those who had GI symptoms and those who did not for GERS, nausea, or bloating after adjusting for sex, race, 30-day postoperative symptoms and BDI (Table 2). However, there were significant differences at 60 months post-surgery for diarrhea (Table 2).

The physical component score of the SF-36 measure demonstrated statistically significant changes in QOL scores between those who had GI symptoms and those who did not with the exception of the symptom of bloating. These differences were detected by a significant interaction of symptom by visit, after adjusting for sex, 30-day post-operative symptoms, sleep apnea BDI and % BMI change from baseline. For those who had GERS, their overall PCS was 3.1 (95% CI 1.6, 4.5) points lower than those with mild or no symptoms (p<0.0001). Individual post-operative visit differences were then examined for nausea and diarrhea. The only significant differences were for nausea at 36 and 48 months. For nausea at 36 months, the PCS for those with nausea was 6.7 (95% CI 2.7, 10.8) points lower those with mild or no symptoms (p=0.004), and at 48 months, the PCS was 7.2 (95% CI 5.5, 10.6) points lower (p=0.0004) (Table 2). The MCID was reached at certain times points for the PCS in patients with GERS, nausea, and diarrhea (Figure 4).

Figure 4a-d:

Difference in Adjusted Mean PCS after Surgery by Symptoms of GERS, Nausea, Bloating and Diarrhea

An analysis was performed to examine the interaction between surgical procedure and symptom severity. There were no significant differences for the gastrointestinal symptoms under study and effect on QoL between the two surgical procedures.

DISCUSSION

The presence of symptoms of GERS, nausea, and diarrhea are associated with diminished improvement in health related QOL in adolescents following bariatric surgery compared to those without symptoms. This diminished health related QOL is demonstrated in the physical component of the SF-36 score. The current study is an important contribution to the literature as data regarding the effect of GI symptoms on health related QOL after bariatric surgery in adolescents is lacking. The results of this study suggest that adolescents should be monitored and treated for these symptoms after surgery to address their diminished health related QOL.

The baseline component scores in the current study were similar to the published SF-36 norms for severely obese adults⁽²²⁾ with the exception of pain, physical role functioning and emotional role functioning which were 15, 15, and 9 points higher in the current adolescent cohort, respectively. This suggests that severely obese adolescents experience a similar decrease in health related QOL to severely obese adults despite a higher level of family support. This may be due to the effect of perceived classmates’ lack of support⁽²³⁾. Demonstrating the effect of obesity on health related QOL, a comparison to adolescent SF-36 norms showed that most baseline component scores were lower in the current study, with the exception of MCS, emotional role functioning, and mental health⁽²⁴⁾. Interestingly, at 5 years after surgery, these component scores reached or exceeded the levels of these published norms with the exception of general health (10 points lower) and physical function (6 points lower), indicating the significant effect of MBS on health related QOL. This improvement in health related QOL after MBS has been seen in an adult cohort undergoing gastric bypass surgery. The adult cohort also demonstrated a greater increase in the SF-36 PCS after surgery compared to the SF-36 MCS⁽²⁵⁾ suggesting surgery has a larger impact on the physical aspects of health related QOL.

In regard to gastrointestinal QOL after surgery, the adult literature is heterogeneous and has demonstrated mixed outcomes. The most commonly used assessment tool is the gastrointestinal quality of life index (GIQLI). This tool has several domains including core symptoms, physical, psychological, social, and disease specific⁽²⁶⁾. Comparing the GIQLI to the SF-36, there is overlap in the themes of vitality, mental health, physical functioning, physical role functioning, and emotional role functioning. There is the addition of asking how often a variety of GI symptoms have occurred over the 2 previous weeks, similar to the GIS rating scale used in the current study. Most studies find an overall increase in GIQLI after surgery with follow up ranging from 1-2 years^(27-29). However, one study found improvement in the social, physical, and psychologic domains but not in the domains regarding GI symptoms⁽³⁰⁾. Another study specifically evaluated GIQLI and food tolerance after surgery and found that food tolerance correlated with the GI symptom domains of the GIQLI and that the food tolerance and overall GIQLI improved after surgery, more for sleeve when compared to RYGB⁽²⁸⁾ emphasizing the important relationship between gastrointestinal symptoms and health related QOL. Although we did not use the GIQLI in the current study, we also found a relationship between gastrointestinal symptoms and a diminished health related QOL.

It was not surprising that there was no significance difference noted between groups in the IWQOL-KIDS measure as this more specific tool measures weight related QOL and may not capture QOL changes that are not directly related to weight changes. Although, there may be a small effect because when comparing to the published overall IWQOL-KIDS scores of the same cohort at 3 years⁽²⁾, the mean IWQOL-KIDS scores for symptoms of nausea and bloating were lower at that same time point⁽²⁾. Reviewing the SF-36 components, it also seems reasonable that the effect gastrointestinal symptoms would be seen the physical component score rather than the mental component score as it is a more physical effect. Reviewing the overall trends of the physical component score of the SF-36, none of the GIS had scores that decreased to preoperative nadir with the exception of nausea. Despite not reaching the baseline low QOL scores, the decreases in the PCS were clinically significant as the previously established MCID was reached.

There are some limitations to the study. The current analysis was limited by the QOL measures that were included in the original study as well as a lack of a non-operative control group. As there was decreased health related QOL for those with symptoms, as captured by the physical component of the SF-36, this study suggests that a more GI specific measure such as the Gastrointestinal Quality of Life Index (GIQLI) tool would be helpful to further delineate the effect of GI symptoms on health related QOL and represents an area for future research. As multiple outcomes were measured, there is a risk of type 1 error. However, a tukey adjustment was used to account for the multiple comparisons between visits. Despite these limitations, the prospective design, uniformity of data collection, length of follow up, and high subject retention rate add to the overall strength and value of this study.

CONCLUSION

In conclusion, health related QOL increases after MBS, specifically in the physical health components of the SF-36. However, the presence of gastrointestinal symptoms, specifically reflux symptoms, nausea, and diarrhea, mitigate the increase in health related QOL of adolescents following bariatric surgery. Patients should be monitored and treated for these symptoms to address this decreased health related QOL.

Supplementary Material

2

Table A: SF-36 component scores at baseline and over time after MBS

1

Table B: SF-36 individual question scores at baseline and over time after MBS

3

Table C: Linear Regression Estimates for the Effect of GI symptoms of GERS, Nausea, Bloating, or Diarrhea and on IWQOL-KIDS subscales (physical, body esteem, social life and family relationships).

Figure 1:

Variable Diagram

Highlights:

• The SF-36 PCS increased significantly over time after MBS in adolescents

• This increase was diminished by the presence of moderative to severe reflux symptoms, nausea, and diarrhea

• The domains that showed the biggest increase after surgery were physical functioning, physical role functioning, and general health