The International Bariatric Surgery Registry

by Edward E. Mason, MD, PhD, FACS; and Kathleen E. Renquist, BS

Dr. Mason and Ms. Renquist are from The Roy J. and Lucille A. Carver College of Medicine, Department of Surgery; The University of Iowa, Iowa City, Iowa.

Abstract

Objective: To report the 20-year experience of a large data repository for surgical treatment of obesity with emphasis on development, problems, and outcomes. Design: Standardized data collection and analysis of voluntarily collected preoperative, operative, and postoperative information following surgical treatment for obesity. Setting: This represents the first outcome research facility organized for bariatric surgery in the United States involving multiple data collection sites using standardized software (1986 to the present). Participants: The 85 data collection sites include 148 surgeons. Information represents 45,294 patients from seven international sites and 28 sites in the US. Measurements: NBSR and IBSR data collection software, SAS analyses, and National Death Index searches to study longevity and cause of death. Results: Early publications and COE requests have concentrated on early outcomes. From 1986 to 2005, changes occurred in operative selection from predominately open, simple procedures (63% to 6%) to complex, bypass operations (36.9% to 94.0%) with an increase in the laparoscopic technique (0% to 22%). In addition, mean days of postoperative hospital stay declined from 4.9 to 3.3, while major 30-day complications increased (2.6% to 4.1%). Conclusion: In 2007, reporting results for surgical treatment of obesity must focus more on long-term mortality and cause of death. The IBSR can obtain this information if sufficient financial support is attained.

INTRODUCTION

This is a review of a 20-year effort to find the best operation(s) for severely obese patients. Using standardized data collection, analysis of aggregated data from multiple data collection sites were studied for 1) initial visit, 2) operation, and 3) lifelong outcome. The first two objectives were attainable.[1–6] The last was the most difficult, but holds the most promise.[7–9] Lifelong follow-up for two important variables can be obtained through the use of the National Death Index (NDI). The NDI has made it possible to compare the effects of simple and complex operations upon longevity and cause of death. Longevity provides a summary of the most important effects of changes in body weight plus the risks associated with a chosen operation. Causes of death are subjects of study for further improvement in outcome.

Getting Started

In 1977, Alden published a comparison of gastric and intestinal bypass. In the discussion of his presentation he mentioned that a Registry for obesity surgery had been started at the University of Iowa.[10] However, it was not until January 1, 1986, that the National Bariatric Surgery Registry (NBSR) began with full financial support for two years from a surgical stapling company with the understanding that continued support would come from surgeon participants. In 1996, the NBSR became the International Bariatric Surgery Registry (IBSR) due to participation from surgeons outside the United States.

Computerized analysis of variance for clinical data began in the department of surgery at the University of Iowa in 1960.[11,12] In 1966, when stomach operations for obesity began; data were saved and analyzed on the University of Iowa mainframe computer. We had 20 years of experience with the computerized analysis of obesity surgery data by the time we started the IBSR in 1986. At that time, standardized NBSR software was offered free of charge to all ASBS members. A minority accepted the software and collected data. An additional expense and special effort was required to establish a satellite registry for each member. The data entry sites required on-the-job learning of obesity-surgery chart-review and data-entry. Data entry people varied from surgeons, nurses, and office personnel to non-medical people with computer skills. All data entry people obtained software support from the IBSR Manager to get started and for ongoing problems. Quarterly newsletters with separate medical and data management sections have been provided for members since the beginning in 1986. Recent newsletters may be viewed on http://www.healthcare.uiowa.edu/Surgery/ibsr/.
The IBSR software has evolved to force completeness and accuracy of entered data. However, examination of the aggregate data as it is assembled in the central registry takes additional time and effort in quality control. Data for each surgeon member remains available on the local computer, and the software provides assembly of reports and graphs for use with patients and staff.

It may seem that all of the data needed for all of the studies that any member wants to perform could be collected. However, we found that requesting too much information causes frustration during data entry, errors, missing data, and software performance issues. The solution to this dilemma in 1988 was to designate a limited number of ‘primary’ data fields that required data entry for every patient. Completion rate for primary field data became an important measure of quality of data collection. Many such refinements marked the learning curve of the IBSR.

Changing Priorities, Demands and Emphasis

There have been changes in the prevalence of obesity and the practice of surgical treatment.[13,14] The emphasis of the IBSR has also shifted from early to life-long outcome. Between 1980 and 2000, the prevalence of people with a BMI greater than or equal to 30 doubled from 15 to 30 percent of the US population, and the greatest increase in weight was in heavier people who were candidates for surgical treatment.[15] As a result, there are now millions of candidates. The laparoscopic approach to abdominal surgery also increased the demand for obesity operations (Figure 1). The number of surgeons in each center performing obesity surgery has increased.[16] A national move has developed to designate centers of excellence (COE) in response to this increase in surgical treatment of obesity, early complications (Figure 2), and to control cost. This involves a review of early outcome and requires collection of data, which the IBSR provides for its members.

In 1986 it was hoped all bariatric surgeons would join the IBSR and participate in the collection of data voluntarily. Data collection is no longer voluntary today. The focus for designating COEs is on early outcome because early complications put patients at risk and increase the cost of patient care. Longevity and late in life complications related to residual obesity and to operations used for treatment will surely have more impact upon the total cost of medical care than the first years of outcome. In 2005, the IBSR published the first paper from an ongoing survivorship study comparing 7,185 patients with simple restriction operations and 11,787 patients with complex bypass operations.[17] For this population of 18,972 patients, operated on from January 1, 1986 to December 31, 1999, no difference in survival to December 31, 2001, was found between the two operation categories. The average time between operation and follow-up (death or no death to the endpoint 12-31-2001) was 8.3 years, with 3.5 percent deaths. If sufficient funding is obtained, this IBSR study can be extended four more years to December 31, 2005.

Preliminary examination of causes of death between the two operation categories did show differences (unpublished data). Further study of these differences has been deferred until a larger percent of deaths can be included following an extended NDI search.

The IBSR goal of determining late in life outcome can be achieved by the IBSR. Newer registries do not have patients who have been followed long enough. If we do not move ahead with the ongoing IBSR study, which today could provide 7 to 20 years of follow-up, millions of operations may be performed that may not provide the greatest possible longevity. Severe obesity is a lifelong disease and a cause of many other lethal diseases. Simple restriction operations appear to preserve life as effectively as operations that bypass the upper digestive tract, and yet they are infrequently offered.[18] The operations offered should be based upon early and lifelong outcome. It is quite possible that one operation is not the best operation for all patients.

The Future of IBSR

Lifelong operations for a lifelong disease need lifelong information about outcome. Lifelong information has proven to be difficult to obtain by the majority of data sites. Some long-term information can be obtained from the Swedish Obesity Subjects (SOS) study, but only 265 of the 2000 patients (13%) had a gastric bypass. There were 1,368 (68%) subjects with vertical banded gastroplasty. Sjöström, the principle investigator for the SOS study, recently reported a reduction in overall unadjusted mortality of 24.6 percent (p=0.0346) for the patients who were treated with surgery compared with the matched control group of patients treated without an operation.19 Adams, et al., used the NDI to obtain an average follow-up of 8.4 years for 8,172 post gastric bypass patients.[20] These patients came from a single surgical practice with patient entry into the study beginning 18 years ago.

Compared to an equal number of population-based controls, matched for age at operation, gender and BMI, they reported a 50-percent reduction in mortality occurring after the first year.
The years of data entry for the Adams study, use of the NDI, and the results are valuable as they stand, but also provide a comparison for the IBSR findings. Adams, et al, provided a matched control group without operation similar to SOS. The IBSR study provides a comparison between bypass and simple restriction operations, which the Adams study lacks. The IBSR patients came from many sites in the United States (Figure 3) while the Adams study patients came from one surgical site. The two categories of operations provided over the years for IBSR patients are shown in (Figure 4). The IBSR study should be extended before the null hypothesis is accepted that it makes no difference whether a simple or complex operation is used in prolonging the lives of the severely obese. Many surgeons assume their form of gastric bypass is the gold standard for all patients. This may be true, but this hypothesis needs further study of longevity and causes of death.

Figure 3

Figure 4
The demand for medical care in the next 10 to 20 years will be affected by how obesity is treated now. Most patients will live longer if they weigh less but it is increasingly important to determine the degree to which different categories of operations introduce other risks that may become manifest as patients live longer. Patients deserve to know as much as possible about the lifelong effects of the operation they choose.21 The IBSR appreciates the contributions over 20 years from the 148 surgeons who have supported and participated in this work. Continued involvement in this effort is necessary and we invite others to participate, especially those who are concerned about the lifelong outcome of the operation(s) they are using.

Disclosure

IBSR financial support is from the University of Iowa Department of Surgery, the American Society for Bariatric Surgery, IBSR surgeon members, and Newsletter subscribers.

References

1. Mason EE, Renquist K, Jiang D. Perioperative risks and safety of surgery in severe obesity. Am J Clin Nutr 1992; 55(2):573S–576S.
2. Renquist KE. Surgical treatment of obesity in America: Data according to the National Bariatric Surgery Registry. Probs in Gen Surg 1992;9:231–8.
3. Mason E, Renquist K, Jiang D. Predictors of two obesity complications: Diabetes and hypertension. Obes Surg 1992;2:231–7.
4. Renquist KE, Mason EE, Tang S et al. Pay status as a predictor of outcome in surgical treatment for obesity. Obes Surg 1996;6:224–32.
5. Mason EE, Renquist K, Zimmerman B. Gallbladder management in obesity surgery. Obes Surg 2002;12:222–9.
6. Mason EE, Renquist K, Huang Y-H et al. Cause of 30-day obesity surgery mortality: With emphasis on bypass obstruction. Obes Surg 2007;17:9–14.
7. Renquist KE, Cullen JJ, Barnes D et al. The effect of follow-up on reporting success in obesity surgery. Obes Surg 1995;5:285–92.
8. Jiang D, Renquist KE, Mason EE. Weight loss curve analysis. Obes Surg 1991;1:373–80.
9. Jeng G, Renquist K, Doherty C, Mason EE. A study on predicting weight loss following surgical treatment of obesity. Obes Surg 1994;4:29–36.
10. Alden JF. Gastric and jejunoileal bypass. A comparison in the treatment of morbid obesity. Arch Surg 1977;112:799–806.
11. Mason EE. Computer analysis in development of new diagnostic methods. JAMA 1961;178:1088–90.
12. Mason EE, Bulgren WG. Computer Applications in Medicine. Springfield: Charles C. Thomas, 1964.
13. Mason EE, Tang S, Renquist KE, et al. A decade of change in obesity surgery. Obes Surg 1997;7:189–97.
14. Zhao Y (Social and Scientific Systems, Inc.), Encinosa W (AHRQ). Bariatric Surgery Utilization and Outcomes in 1998 and 2004. Statistical Brief #23. January 2007. Agency for Healthcare Research and Quality, Rockville, MD. Available at: http://www.hcup-us.ahrq.gov/reports/statbriefs/sb23.pdf, Accessed 01-12-2007.
15. Flegal KM, Troiano RP. Changes in the distribution of body mass index of adults and children in the US population. Int J Obes 2000;24:807–18.
16. Samuel I, Mason EE, Renquist KE et al. Bariatric surgery trends: An 18-Year report from the International Bariatric Surgery Registry. Am J Surg 2006;192:657–62.
17. Zhang W, Mason EE, Renquist KE et al. Factors influencing survival following treatment of obesity. Obes Surg 2005;15:43–50.
18. Mason EE. Development and future of gastroplasties for morbid obesity. Arch Surg 2003;138:361–6.
19. Sjöström L. Soft and hard endpoints over 5 to 18 years in the intervention trial Swedish obese subjects (abstract). Obes Rev 2006;7(Suppl 2):27.
20. Adams T, Gress R, Smith S, et al. Long-term mortality following gastric bypass surgery (abstract). Obes Rev 2006; 7(Suppl 2):94.
21. Mason EE, Hesson WW. Informed consent for obesity surgery. Obes Surg 1998;8:419–28.

Category: Commentary, Past Articles