Weight regain after Bariatric Surgery increases the request for Traditional Chinese Medicine Results from the cross-sectional BARBQTCM survey


 Background: To investigate patient attitudes towards Traditional Chinese Medicine (TCM) after bariatric surgery with special emphasis on postoperative weight regain (WR).Methods: We retrospectively compared experiences with Complementary Additive Medicine (CAM) and the desire for additional TCM between patients with stable weight (SW) and WR by means of the disease-specific BARBQTCM survey during postoperative follow-up in a single bariatric center.Results: Out of 467 participating patients, 150 (32.1%) had WR after bariatric surgery. Median age was 49 years and 78.4% of the patients were female. 233 patients (49.9%) had already had experience with CAM and 125 (26.8%) with TCM. 381 patients (81.6%) were interested in using TCM at the time of the survey. Acceptance of TCM was lower in men (OR 0.35, P<0.01) and decreased with age (OR 0.95, P<0.001). Overall, the request for TCM was lower in the SW cohort compared to the WR group (77.3% versus 90.7%; OR 0.29, P<0.01). There was a positive correlation between preoperative experience with CAM and postoperative request for TCM (OR 5, P<0.001). The proportion of patients who were interested in outpatient TCM twice or more monthly was higher in patients with WR than in those with SW (94.0% versus 84.2%, P<0.01).Conclusions: The overall acceptance of CAM and TCM was high in our bariatric surgical patients. WR increased the rate of acceptance of TCM. We conclude that CAM and TCM should be offered during postoperative follow-up after bariatric surgery as part of an integrative medical approach.

David L. Sackett, the founder of the modern "Evidence-based Medicine (EbM)" underlined three essential cornerstones of EbM: (1) relevant external scienti c evidence in combination with (2) patients' values and preferences leading to (3)

solid internal clinical experience and vice versa. Meta-Analyses and Systematic
Reviews (MASR) achieve the highest level of scienti c evidence according to the requirements of the Scottish Intercollegiate Guidelines Network (SIGN).
According to the EbM concept of Sackett, TCM has been proven to have a broad-based cornerstone of relevant external evidence to support its use in the eld of weight control and obesity. In view of a generally growing acceptance of CAM and TCM (24), in this study we wanted to gain insight into patient attitudes toward CAM and TCM after bariatric interventions, especially in those with WR.
We hypothezised that patients with a secondary increase in weight may be more interested in TCM therapy than patients who retain a stable weight. We therefore compared experiences with CAM and the desire for a complementary TCM treatment between patients with stable weight (SW) and patients who experienced weight regain (WR) during postoperative follow-up period.
To the best of our knowledge, this is the rst study to investigate CAM and TCM preferences of obese and overweight patients after bariatric surgery using a disease-speci c questionnairy.

Methods
This is a retrospective single-center sub-analysis of the prospectively-collected multi-center BARBQTCM survey (BARiatric Basic Questionnairy for Traditional Chinese Medicine). The anonymous BARBQTCM questionnairy is a self-rated 7-item disease-speci c inventory focussing on overweight, obesity and PSALMS. It includes demographic parameters, former treatments for obesity, prior experience with CAM, request for TCM therapies, disease-speci c complaints that patients would treat with TCM, frequency with which patients would visit a TCM outpatient clinic per month, and the amount of money patients would be willing to spend per month for TCM therapies (Supplementary Materials). We analyzed the data from the Center for Bariatric Surgery, Bern, Switzerland. We received permission from the ethics committee (EK-Nr E17-N01-02) to perform the study. It was conducted according to the criteria of Good Clinical Practice, as edited by the Medical Faculty of the Sigmund Freud University in Vienna and the Declaration of Helsinki of 1964, updated in Fortaleza, Brasil 2013. Inclusion criteria were age over 18 years and treatment at an outpatient obesity center. Exclusion criteria were patients who did not have bariatric surgery, missing information about surgical intervention, and missing information about post-operative weight course.
We de ned WR as a secondary weight gain of more than 10 kg after bariatric surgery. Our primary outcome was the correlation between WR and the request for support by TCM. In addition, we analyzed differences between patients who had WR and those without WR. Language Reference, version 2.7. http://www.python.org) and R (R Core Team (2017), R Foundation for Statistical Computing, Vienna, Austria, https://www.R-project.org/). The descriptive analysis was followed by assessing the statistical differences between both groups. This was performed using the Kruskal-Wallis test for continuous variables and Chi-square test for discrete variables. We considered results as statistically signi cant if P was less than 0.05. Numerical variables are given as median and 95% Con dence Interval (95% CI) and categorical variables as numbers and percentages. In the percentages, we deducted missing information from the denominator throughout. Variables which yielded a statistically signi cant difference were then analyzed using multi-and univariate logistic regression (including stepwise regression) for their correlation with WR.
Results 500 patients completed the BARBQTCM questionnairy. We excluded 33 patients (6.6%) who did not undergo surgery (n=10), indicated no details about the surgical intervention (n=3), and/or did not give information about post-operative weight course (n=20), leaving 467 patients for nal analysis. 317 patients (67.9%) were satis ed with a SW course after the initial postoperative weight loss, whereas 150 patients (32.1%) suffered from a secondary WR. Median age was 49 years old, and 78% of the participants were female; this was not signi cant between the two groups (Table 1). Median BMI at the time of survey was higher in the WR cohort compared to the SW group (32kg/m 2 versus 29kg/m 2 , respectively, P<0.001) as presented in Table 1.
Prior to the survey, patients with WR had taken more weight loss drugs than those with a stable weight (68.7% versus 46.7%, respectively, P<0.001; Table 1). Half of the WR patients had prior experience with CAM and one third of patients tried out TCM in the past. There was no signi cant difference in both study cohorts concerning CAM experience ( Table 1).   Out of all BARBQTCM participants, 49 patients (10.5%) refused outpatient TCM treatment. The vast majority reported interest in visiting a TCM outpatient clinic one to two times per month (61.2%). 61 patients (13.1%) reported interest in visiting a TCM ward less than once per month and 37 patients (7.9%) would go three or more times a month. 281 patients with SW (88.6%) and 137 patients with WR (91.3%) stated that they wanted to receive outpatient TCM treatment (Figure 2). When comparing both study cohorts, more SW patients were interested in outpatient TCM treatment once per month on average (≤1 monthly: 53.9% versus 45.3%, NS) whereas more WR patients were interested in visiting the TCM outpatient clinic two or more times monthly (≥2 monthly: 26.5% versus 40.7%, P<0.01) as presented in were not willing to spend money on TCM. Comparing the study groups, more SW patients were interested in spending 1 to 50 Swiss Francs for TCM treatment (1-50 CHF monthly: 18.6% versus 16.0%, NS) whereas WR patients were interested in spending a larger amount of money (51-200 CHF monthly: 53.3% versus 62.7%, NS) as shown in Figure 3.

Discussion
In the present study, we aimed to analyse the importance of CAM and TCM for overweight and obese patients after bariatric surgery. The main nding of our study is that patients who had undergone bariatric surgery were highly interested in TCM (82%), and those who experienced WR were even more so (91%). To our knowledge, the desire for CAM and TCM in this clinical scenario of patients after bariatric surgery has not yet been described in literature.
In the BARBQTCM survey, TCM and acupuncture use was reported in 27% of patients in the SW group and in 33% in the WR patients. Interest in TCM exceeded 80% which is much higher than 2% of surveyed patients using acupuncture in the National Health Interview Survey (NHIS) 2002 and 2012 (24).
Comparable rates of TCM herbal use (obese women: 26.2%) were reported from a survey in Taiwan prior to professional obesity treatment (25). In contrast to our data, two NHIS analyses of more than 31 000 adults in the USA found that obesity correlated with reduced usage of CAM therapies (OR 0.83 and 0.82, P<0.0001). Yoga (OR 0.35, P<0.0001), tai qi (OR 0.59, P<0.05) and even acupuncture (OR 0.56, P<0.05) were utilized less frequently by obese patients (26,27). One could argue that TCM is more common in eastern countries compared to western countries. On the other hand, another NHIS analysis investigated CAM use among 15 400 people with chronic diseases in the USA (28), including hyperecholesterolemia, hypertension, diabetes and obesity (i.e. PSALMS). Nearly two thirds of the survey participants had more than one comorbidity. Of that cohort, up to 30% of participants used CAM, which is nearly equal to the ndings from the BARBQTCM survey. Summing up ndings from that analysis and the BARBQTCM survey, CAM is a viable and accepted treatment option for at least one third of obese patients with a concomitant PSALMS including multiple chronic complaints.
The BARBQTCM single-center analysis gives insight about patients with WR after bariatric surgery, which comprises at least one third of our survey participants. Gastric bypass (RYGB and OAGB) had acceptable WR rates and sleeve gastrectomy (SG) achieved best weight loss results in the BARBQTCM cohort. Patients with a history of multiple revisional bariatric procedures reached WR rates of 67%. In addition, we found that gastric banding (GB) was the index operation most frequently followed by WR. Our observation does not coincide with results from a study by Courcoulas et al. who did not observe higher WR after GB compared to RYGB or SG during a seven to 10 year follow up (29,30). In contrast to those ndings Baig et al. reported the highest WR proportion after SG (35%), followed by RYGB (15%) and OAGB (3%) in a large cohort of more than 9600 indians (31). Clapp et al. con rmed a WR rate of 28% after SG at 7-year follow up (32). The lack of knowledge of the time between surgery and our survey might have in uenced the percentages of patients who have regained weight in the different intervention groups. Since SG is a novel intervention that has been introduced more recently in our center and the other interventions are "older," it is possible that SG patients have not yet had the time to regain weight. This could also explain why patients with several bariatric procedures had more weight regain, since 88% of these patients initially had a GB, the oldest intervention, that was removed followed by a conversion to another procedure.
In general, revisional bariatric surgery is associated with high complication and readmission rates and a greater overall morbidity as revealed in a MBSAQIP (Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program) database analysis of more than 2700 conversion procedures (33). Another multicenter study experience revealed WR in one quarter of RYGB patients, who were treated with endoscopic transoral outlet reduction. Those revisional procedures resulted in chronic pain in 18% of patients, nausea in 14%, and repeat endoscopic procedures in 8% of patients (34). In this context, one could imagine the di cult situation chronic overweight patients are in, after already having undergone multiple procedures, and being open to alternative medical treatments. The willingness to accept CAM and TCM might have been in uenced by previous positive experiences with the respective treatments, since we found a correlation between postoperative willingness for CAM or TCM and preoperative usage.
In the scienti c literature, acupuncture has been shown to be effective in treating different addictions. There exist 11 MASR on the treatment of food addiction and obesity by acupuncture (8-12, 14-16, 23, 35), which underline that body acupuncture should be combined with lifestyle modi cations such as diet and exercise and with other therapy variations including ear acupuncture, electroacupuncture and acupressure to improve outcome. A recent network meta-analysis of 2283 patients from 34 trials emphasized the superiority of the combination of TCM therapies over placebo, pharmacotherapy and life style modi cation alone for both body weight and BMI reduction (15). Furthermore, phytotherapy has been shown to interact with pathophysiological hunger pathways (36). These ndings are veri ed by another 6 MASR (17)(18)(19)(20)(21)(22). All 17 MASR report signi cant body weight and BMI reduction with a high procedure safety and low adverse events. Nevertheless, they conclude that ndings have to be treated with caution due to data heterogeneity and unclear and high risk of bias.
A weak point of our work is that there were many confounding factors in the survey. We found that previous experiences with CAM and TCM increased the request for TCM. However, since there was also an interest in CAM in patients without WR, other factors such as satisfaction with weight, satisfaction with ones' body image and other factors could have in uenced their response.
We also consider the de nition of the weight regain of 10 kg to be a weak point of the study. There are currently various measures of weight regain that might be assessed and that might be more accurate (3). A more precise documentation of the weight course could have examined whether there is a correlation between the amount of WR and the request for CAM or TCM.
As already mentioned, the questionnairy did not include information about the time between surgery and our survey. It would be interesting to investigate if there was a correlation between desire to bene t from CAM or TCM and the time from surgery.
Due to the unicentric setting, it is not clear whether our results can be transferred to other populations. However, the results of the entire BARBQTCM trial including 5000 patients from Austria, Germany and Switzerland will reveal if patients in the respective countries have similar attitudes towards CAM and TCM.
Despite these limitations, we believe our data to be informative and convincing given the large number of included patients treated by one medical team, thereby reducing in uencing factors from medical treatments, thus allowing for robust statistical results.

Conclusions
We were able to show that TCM might play a crucial role in the weight management of obese patients who have undergone bariatric surgery. We also found that overall willingness to attend outpatient TCM sessions was high, and even higher in patients with WR. The extent of the bene t of such multimodal therapy regimens on weight loss outcomes is yet to be determined.
In the context of EbM, external scienti c evidence and patients' preferences de nitively support an integrative medical approach, combining conventional and complementary medicine in the struggle against the world's most heavyweighted pandemic. We want to stress our conclusion that multicenter network research with standardized programs and the acquisition of longterm results will establish and consolidate the third cornerstone of Evidence-based Medicine, which contributes to a solid internal clinical experience.

Declarations
Ethics approval: We received permission from the ethics committee (EK-Nr E17-N01-02) to perform the study.
Consent for publication: not applicable.
Availability of data and materials: not applicable.
Competing interests: The authors have no competing interests to declare.
Funding: There was no external funding.
Authors' contributions: All authors substatially added to the present manuscript. All authors read and approved the nal manuscript. BW, UK, and MK equally contributed to the present work; they analyzed data, wrtote the initial draft, and approved the nal manuscript. UK, SR, CT, and JZ were implicated in data collection. CT, JZ and PP were major contributors in developing and revising the manuscript. SR, JZ and PP developed the project and monitored the course of the study. PP is the garantor. Frequency of visits to a TCM outpatient clinic (*more WR patients were interested in visiting the TCM outpatient clinic two or more times monthly, P<0.01, respectively; the other differences between the two cohorts were not signi cant)