An overview of a systematic review and meta-analysis
by Yeongkeun Kwon, MD, and Sungsoo Park, MD, PhD
Dr. Yeongkeun Kwon is from the Department of Family Medicine, Korea University College of Medicine, Seoul, Korea. Dr. Sungsoo Park is from the Department of Surgery, Korea University College of Medicine.
Bariatric Times. 2016;13(2):8–10.
The gastric bypass operation, which was developed by Drs. Edward Mason and Chikashi Ito in the 1960s, was based on two earlier procedures—Billroth I and Billroth II—named after Dr. Theodor Billroth, who successfully carried out the first gastroduodenostomy in 1881. In the Billroth I (BI) procedure, the surgeon removes the pylorus and connects the remaining portion of the stomach with the duodenum. In 1885, Billroth I was modified to Billroth II (BII) when Billroth and von Hacker joined the remaining stomach to the jejunum in a side-to-side manner.(gastrojejunostomy). Although both procedures share some technical characteristics, they differ in that the BII procedure diverts food away from the duodenum and the BI procedure restores the normal configuration of the gastrointestinal tract. Although these procedures were originally performed to treat gastric ulcers, researchers later found that type 2 diabetes mellitus (T2DM) was improved in non obese patients who underwent BII.
Since then, studies have shown that BII reconstruction after subtotal gastrectomy for cancer or intractable ulcers can more effectively improve type 2 diabetes mellitus (T2DM) than Billroth I reconstruction by allowing patients to achieve normoglycemia without or with lower doses of diabetes medications. Given the strong association between diabetes and mortality in cancer patients, we sought to study this unique patient population to further examine the effect of metabolic surgery on T2DM.
In the November–December 2015 issue of Surgery for Obesity and Related Diseases,3 we reported the results of a systematic review and meta-analysis conducted with Dr. Hyun Jung Kim from the Institute for Evidence-based Medicine, The Korean Branch of Australasian Cochrane Center, Department of Preventive Medicine, Korea University College of Medicine, in Seoul, Korea; and Drs. Emanuele Lo Menzo, Samuel Szomstein, and Raul J. Rosenthal from Bariatric and Metabolic Institute, Section of Minimally Invasive Surgery, Cleveland Clinic Florida, Weston, Florida.
This study sought to 1) compare the diabetes improvement rate between patients who underwent BI reconstruction and those who underwent BII reconstruction, and 2) determine any potential baseline characteristics that might differentiate patients who will achieve diabetes amelioration following subtotal gastrectomy using Billroth techniques from those who will not.
First, we searched published scientific literature that were considered relevant to our topic and published between 1950 and June 25, 2014. We used MEDLINE, EMBASE, and the Cochrane Central Register of Controlled Trials.
We considered papers eligible if they 1) were done in gastric cancer or intractable ulcer patients with previously diagnosed T2DM; 2) studied the effect of Billroth techniques on diabetes status by reporting the means and standard deviations of fasting plasma glucose or glycated hemoglobin levels at baseline and after subtotal gastrectomy; and 3) assessed patients for diabetes amelioration after any duration of follow up.
Two pooled relative risks (RR) were calculated: one for diabetes remission and one for diabetes amelioration. Remission status was defined as “cure” or “resolution” in each study. Amelioration status excluded the words “unchanged,” “worsened,” “same,” “stationary,” or “invalid” in each study).
We found eight studies that met criteria—six studies in Korea and one each in China and Taiwan.3–10 These studies provided data for 972 participants. The patients included in the our systematic review had a baseline glycated hemoglobin level of 7.2 to 8.7 percent and a baseline BMI of 23.8 to 27.3 kg/m2.
Among the 972 participants, BII reconstruction was associated with a higher RR of postoperative diabetes amelioration and remission than BI reconstruction (Figures 1A AND 1B).
To identify other sources of the slight residual difference between trials, we performed meta-regression analyses of the baseline age, baseline BMI, baseline fasting plasma glucose level, baseline glycated hemoglobin level, diabetes duration, and BMI reduction rate. We found that the BMI reduction rate was borderline significant (P=0.069) on the univariate meta-regression models and became significant (P=0.047) when age was included. The BMI reduction rate was inversely related to the RR (effect size [Figure 2]. The coefficient of BMI reduction rate in the meta-regression model was -0.18 (95% CI, -0.36 to -0.004).
Table 1 shows the meta-analysis comparing the baseline clinical characteristics (mean difference) between patients who experienced postoperative diabetes amelioration and those who did not. Patients who achieved amelioration had a higher baseline BMI (weighted mean difference, 0.88; 95% CI, 0.38 to 1.38) and shorter diabetes duration at baseline (weighted mean difference, -0.40; 95% CI, -0.23 to -0.70).
Our analysis demonstrated that BII reconstruction is more effective than BI reconstruction for achieving postoperative diabetes control.
Limitations. This study was limited in that it was an analysis of only non-randomized controlled trials and, therefore, should be interpreted with caution as selection bias may exist. Only four of the eight studies we analyzed included a covariate-adjusted odds ratio; therefore, pooling the adjusted odds ratios was impossible. It’s important to note that the selection of reconstruction method (BI or BII) is not related to patient diabetes status in clinical practice. Additionally, the inclusion of patients with comorbidities, other organ resections, and a BMI of 30kg/m2 or more had a limited impact on the overall effect sizes.
Since all of the included studies were performed in Asian countries, we concluded that the results might not be generalizable to other populations. For instance, studies have shown that Asian populations typically have a higher prevalence of diabetes than the Western population when matched for BMI and waist circumference.
We were not able to draw conclusions on long-term postoperative diabetes status as only one study provided relevant data (>3 yr) follow up. We believe this should be an area of focus in additional studies.
Finally, the quality of the individual studies may have been limited in that the presence of gastric neoplasms might have some unknown metabolic or anorectic effects in this patient population.
In summary, we concluded that based on our analysis of the literature, BII reconstruction is more effective than BI reconstruction for achieving postoperative diabetes control. These findings could help guide the individualized choice of post-gastrectomy intestinal reconstruction and the design of new operations intended specifically to treat T2DM. Additional research should be conducted to assess postoperative diabetes status long-term (more than 3 years) and among patients who are not of Asian descent.
Editor’s Note: To read the article that was published in the journal Surgery for Obesity and Related Diseases, visit http://www.soard.org/article/S1550-7289%2815%2900004-0/abstract. Access to full text varies.
1. Friedman MN, Sancetta AJ, Magovern GJ. The amelioration of diabetes mellitus following subtotal gastrectomy. Surg Gynecol Obstet. 1955;100:201-4.
2. Kwon Y, Jung Kim H, Lo Menzo E, Park S, Szomstein S, Rosenthal RJ. A systematic review and meta-analysis of the effect of Billroth reconstruction on type 2 diabetes: A new perspective on old surgical methods. Surg Obes Relat Dis. 2015;11(6):1386–1395. Epub 2015 Jan 8.
3. An JY, Kim YM, Yun MA, Jeon BH, Noh SH. Improvement of type 2 diabetes mellitus after gastric cancer surgery: short-term outcome analysis after gastrectomy. World J Gastroenterol. 2013;19:9410–9417.
4. Hu DP, Cheng XZ, Jing WT, Lai RM, Guo TK. Impact of 3 kinds of digestive tract reconstruction on carbohydrate and lipid metabolism in non-obese type 2 diabetes mellitus patients. World Chin J Digestology. 2013;21:93–97.
5. Kang KC, Shin SH, Lee YJ, Heo YS. Influence of gastrectomy for stomach cancer on type 2 diabetes mellitus for patients with a body mass index less than 30 kg/m2. J Korean Surg Soc. 2012;82:347–355.
6. Kim JW, Cheong JH, Hyung WJ, Choi SH, Noh SH. Outcome after gastrectomy in gastric cancer patients with type 2 diabetes. World J Gastroenterol. 2012;18:49–54.
7. Kwon Y, Abdemur A, Lo Menzo E, Park S, Szomstein S, Rosenthal RJ. The foregut theory as a possible mechanism of action for the remission of type 2 diabetes in low body mass index patients undergoing subtotal gastrectomy for gastric cancer. Surg Obes Relat Dis. 2014;10:235–242.
8. Lee W, Ahn SH, Lee JH et al. Comparative study of diabetes mellitus resolution according to reconstruction type after gastrectomy in gastric cancer patients with diabetes mellitus. Obes Surg. 2012;22:1238–1243.
9. Wang KC, Huang KH, Lan YT et al. Outcome after curative surgery for gastric cancer patients with type 2 diabetes. World J Surg. 2014;38:431–438.
10. Jo JH. Influence of gastrectomy for stomach cancer on type 2 diabetes mellitus [Dissertation]. Kosin University Graduate School of Medicine, Busan, Korea; 2010.
11. Hsu WC, Boyko EJ, Fujimoto WY et al. Pathophysiologic Differences Among Asians, Native Hawaiians, and Other Pacific Islanders and Treatment Implications. Diabetes Care. 2012;35:1189–1198.
FUNDING: No funding was provided.
DISCLOSURES: The authors report no conflicts relevant to the content of this article.