Sleeve gastrectomy is rapidly increasing in popularity among various bariatric surgical options according to recent worldwide statistics.¹ Sleeve gastrectomy has been proven effective for achieving weight loss and managing obesity-related comorbidities simultaneously.² However, some unmet needs have been noted with this gastric-only procedure, including weight regain and the reemergence of diabetes in the long-term.³ An increasing number of patients are expected to require intestinal bypass in addition to gastric sleeve to overcome these shortcomings, which is why interest in the sleeve gastrectomy with duodenojejunal bypass has been increasing recently.^4,5 The article “Factors predicting weight loss after ‘sleeve gastrectomy with loop duodenojejunal bypass’ surgery for obesity” by Vennapusa et al.⁶ was quite informative on this topic.

However, there are critical limitations to their results and conclusion. Although the percentage of excess weight loss (%EWL) or percentage of excess body mass index loss (%EBMIL) had been commonly used in bariatric patients, this outdated %EWL/%EBMIL metric is largely influenced by preoperative body mass index (BMI), distorting true weight loss efficacy after bariatric surgery.^7-10 The percentage of total weight loss (%TWL) metric also generates deviation, but to a much lesser extent than %EWL/%EBMIL, and thus is the recommended tool for reporting weight loss outcomes as the absolute change in BMI or %TWL which is less dependent on the baseline BMI.¹¹ Again, the weight loss reporting metric should be selected carefully when comparing the efficacy of bariatric/metabolic surgery in patients with different baseline BMI to avoid potential bias.

The authors compared several factors including age, sex, preoperative BMI, and the presence of diabetes to identify which predict better postoperative weight loss after sleeve gastrectomy with loop duodenojejunal bypass. As expected, they demonstrated that higher preoperative BMI negatively predicted %EWL, while contradictory results were shown for %TWL. This conflicting result led to confusion when interpreting their conclusion. %EWL/%EBMIL should be replaced with the use of %TWL, which is the currently accepted standard weight loss reporting method to avoid confusion and controversy.

Then, the only independent factor that predicted both 1- and 3-year %TWL was the presence of diabetes. This is quite an interesting finding in that patients with diabetes lost significantly less weight and were more likely to regain their weight. This is inconsistent with recent study results demonstrating that the association between the presence of diabetes and poor weight loss was most likely related to older age in patients with diabetes rather than diabetes itself.¹² It would be informative if the authors could explore the reasons for this discrepancy by comparing the results between patients with and without diabetes. Also, if the presence of diabetes predicts poor weight loss, further analysis on weight loss difference according to the severity of diabetes could provide valuable information.

The author declares no conflict of interest.