Back to 2017 Posters

---

SPHINCTER OF ODDI DISEASE IN GASTRIC BYPASS PATIENTS: IS SURGERY COMPARABLE TO ENDOSCOPIC TREATMENT?
Patrick B. Schwartz^*1, Jeffrey J. Easler², Shadi Aboudi¹, William P. Lancaster¹, Ihab I. El Hajj², James L. Watkins², Evan L. Fogel², Stuart Sherman², Glen A. Lehman², Michael G. House¹, Nicholas J. Zyromski¹, C. Max Schmidt¹, Attila Nakeeb¹, Eugene P. Ceppa^1
1Surgery, Indiana University School of Medicine, Indianapolis, IN; ²Gastroenterology, Indiana University School of Medicine, Indianapolis, IN

Introduction: With a 38% incidence of obesity, treatment via Roux-en-Y gastric bypass (RYGB) for morbid obesity is more common than ever. However, this creates a predicament for the management of Sphincter of Oddi Dysfunction (SOD), a disease affecting up to 1.5% of the population. Traditional diagnosis and treatment with endoscopic retrograde cholangiopancreatography (ERCP) requires advanced techniques (e.g. gastrostomy access) after RYGB. Transduodenal sphincteroplasty is also performed, yet carries the risks of surgery. We hypothesized that surgical sphincteroplasty would have increased morbidity and mortality, but provide a more durable remission of symptoms with increased time between interventions relative to endoscopic treatment.
Methods: All consecutive patients between 2004-2016 with prior RYGB for morbid obesity chosen by treating physicians to undergo endoscopic or surgical management for Type I (16 patients) or Type II (22 patients) SOD by the modified Milwaukee classification system were included in the study. Patients with Type III SOD, or who underwent RYGB for non-obesity indications, were excluded.
Results: Thirty-eight patients (87% Female) with a mean age of 48 years were included. ERCP after surgical gastrostomy was initially performed in 50% (n=19) of patients while open transduodenal sphincteroplasty was performed in another 50% (n=19). Mean BMI (29), weight loss (54 kg) and time since RYGB (7 years) were similar. The 30-day mortality was 0% in our cohort, and 30-day morbidity was similar between endoscopic and surgical cohorts (37% vs. 37%; p=1); post-procedural pancreatitis or pain requiring hospitalization were most common. Successful treatment, or resolution of SOD symptoms, after initial therapy was seen in 42% (8/19) and 74% (14/19) in the ERCP and surgery groups (p=0.05). Re-intervention was performed in 8 ERCP patients and 1 surgical patient; the median interval until re-intervention was 10.2 and 2.4 months, respectively (p=0.9). Of these 9 patients, 5 underwent ERCP and 4 surgery. 30-day mortality was 0%, but 30-day morbidity was higher in both relative to the index intervention (40% vs. 75%) yet not statistically different (p=0.45). Re-intervention was less successful than initial treatment for both groups (20% vs. 25%; p=0.72); one ERCP patient underwent a second re-intervention 5.6 months later. Median remission of symptoms was longer for those treated with surgery (70 vs. 43 months), but was not statistically significant (p=0.097).
Conclusions: Resolution of symptoms was similar between surgical and ERCP groups with similar rates of morbidity for index intervention. However, ERCP patients required more interventions and surgical sphincteroplasty had a higher rate of successful symptom relief after initial therapy. Further studies are needed to elucidate under which conditions each treatment is most beneficial.

		Gastrostomy Access ERCP	Surgical Sphincteroplasty	p Value
Patient #		19	19
Average BMI		30.2	28.6	0.385
SOD
	Type I	7	9	0.743
	Biliary	5	8	0.41
	Pancreatic	2	1
	Type II	12	10	0.743
	Biliary	8	9	0.41
	Pancreatic	4	1
Treatment Success
	Index Procedure	8/19 (42%)	14/19 (74%)	0.05
	Repeat	1/5 (20%)	1/4 (25%)	0.722
30-Day Morbidity
	Index Procedure	7/19 (37%)	7/19 (37%)	1
	Repeat	2/5 (40%)	3/4 (75%)	0.524
LOS after Index Procedure (days)		1.5	4	0.07
Median Interval Between Re-Intervention (mo.)		10.2	2.4	0.9
Median Duration of Symptom Remission (mo.)		43	70.3	0.097

---

Back to 2017 Posters