Covered Stents in Cervical Anastomosis Following Esophagectomy

Emily A Speer, MD1, Christy M Dunst, MD2, Amber Shada, MD1, Kevin M Reavis, MD2, Sara Moncrief, MD1, Lee L Swanstrom, MD2. 1Providence Portland Medical Center, 2The Oregon Clinic


Anastomotic complications after esophagectomy are frequent. The use of self expanding removable covered metal stents has been shown to be an effective initial treatment for esophagogastric anastomotic leaks after esophagectomy, but there is a paucity of literature regarding their long term sequelae. The aim of this study is to review the long term outcomes of anastomotic stenting after esophagectomy with cervical esophagogastric reconstruction, focusing on the safety and efficacy of current stent practices.


All stents placed across cervical anastomoses following esophagectomy from 2004-2014 were retrospectively reviewed. Indications for surgery and stent placement were collected. Leaks were graded on a scale from 1 to 4 according to the classification set forth by Lerut and colleagues [1]. Success was defined as resolution of indicated anastomotic problem for at least 90 days to ensure durability of outcome. Stents were routinely removed or exchanged at 3-6 week intervals. For patients with serial stents, each stent event was evaluated first separately and then as part of its series. Complications were defined as the development of any new anastomotic problems.


Twenty-three patients had a total of 63 stents placed (14% prophylactic, 38% leak, 48% stricture) with an average 2.7 stent events per patient. Median single stent duration was 25.5 days (IQR=13-39.8), and median overall stent duration per patient was 62 days (IQR=40-74.5). 60% of patients had successful resolution of their primary anastomotic problem. 33% resolved with only one stent event. Strictures healed in 25% of patients at a median of 42 days (IQR=32-67). Leaks (40% Grade 1, 30% Grade 2, 30% Grade 3) healed in 70% of patients at a median of 59 days (IQR=50.5-66).

Overall, stent related complications occurred in 78% of patients. The average number of complications per stent event was 1.3. Complications (per stent event) included 62% migration, 9.5% clinically significant tissue overgrowth, 7.9% minor erosion (ulcers), and 9.6% major erosion (into surrounding structures). Stents placed for stricture were more likely to result in complications (80% vs 60%, p=0.09), especially migration which reached significance (76.7% vs 48.5%, p=0.02). Preoperative chemoradiation was a significant risk factor for major erosion (22.5% vs 4.3%, p=0.05) but not for overall complications. Patients with major erosions had longer stent duration compared to those without (92 vs 36 days: IQR 25-52.3). Erosion rates were not affected by stent diameter (20% ≥20mm vs 10.7% <20mm, p=0.32) or indication.


Placement of covered stents across cervical esophagogastric anastomoses is associated with high complication rates. Although most complications are minor, 10% are major erosions, and this number is even higher after chemoradiation. Esophageal stents are relatively effective at controlling leaks in the acute setting but do not appear effective at healing strictures. With a healing rate of only 25% and a complication rate of 80%, the risk clearly outweighs the benefit of stenting anastomotic strictures following esophagectomy. Regardless of indication, caution should be used when stent duration exceeds 2-3 months due to the risk of erosion.

1. Lerut T, et al. Anastomotic complications after esophagectomy. Dig Surg 2002;19:92– 8.

« Return to SAGES 2015 abstract archive

Reset A Lost Password