Pregnancy May Increase the Risk of Proximal Gastric Pouch Dilatation after LAGB Surgery

Abstract

Interpretation of data: Finding no difference is not the same as reassuring us that there isn't one! It is all a matter of confidence. Pregnancy after bariatric surgery is an important topic and one in which we have had a great interest. We have published widely on the outcomes of laparoscopic adjustable gastric banding (LAGB) surgery including two manuscripts specifically on pregnancy and birth outcomes [3, 4]. These provided significant new insights into the management of the band during pregnancy. We have an issue with the interpretation of the data and subsequent conclusion of the manuscript by Rebecca Haward and colleagues, titled “Does pregnancy increase the need for revisional surgery after laparoscopic adjustable gastric banding?” [6]. The authors conclude that it is unlikely that pregnancy has an effect on the need for LAGB revisions. We believe that the data presented does not support this conclusion. The data signals concern that pregnancy may increase the risk of proximal pouch dilatation (PPD) and need for revisional surgery. We have also shared the authors' concern that pregnancy with an LAGB in situ could increase the risk of PPD having observed pregnant women who sometimes reported the onset of obstructive symptoms during pregnancy which continued thereafter, and subsequently required revisional surgery for PPD. Our hypothesis was that the profound changes to smooth muscle, including that of the upper gastrointestinal tract, which are largely hormonally driven and occur during pregnancy [1, 5, 7], may lead to an increased risk of stretching or dilatation of the proximal stomach and /or lower esophagus in the gravid state. Other commonly used bariatric surgical procedures, including Roux-en Y gastric bypass and sleeve gastrectomy, also rely on the stability of the anatomical changes just below the gastroesophageal junction; hence, they may also be at risk of abnormal stretching or dilatation that may occur during pregnancy. Haward et al. set an arbitrary level of concern at a surgical revision rate that is 60% greater in the pregnancy group than in a matched control group. That is if the reoperation rate was greater than a 60% increase or a hazard ratio >1.6, this would signal an increased risk. Showing no significant difference between groups cannot reassure us that there is not an increased risk. To demonstrate that there is no meaningful increased risk, at their elected level, the upper 95% confidence interval (CI) of the hazard ratio should be at or below 1.6. The authors report a hazard ratio for primary revision of the band in the pregnancy group at 3 years of 1.7 with a CI of 0.9 to 3.3. For PPD, we calculate from their data an odds ratio of 1.9 with a CI of 0.9 to 4.1 (Table 3, p=0.08). While the study did not find a definite statistically significant increase in revisional surgery in those that had a pregnancy, it did not reassure us that there was no increase in risk, with both ratios above 1.6 and upper CIs well above the 1.6 level set by the authors for a signal of increased risk. Clearly, the power of the study was J. B. Dixon :M. E. Dixon Obesity Research Unit, Department of General Practice, Monash University, Melbourne, Australia