CHICAGO – Nonsurgical approaches are beginning to dominate the management of acute esophageal perforations.
An analysis of 81 consecutive acute esophageal perforation cases between June 1989 and March 2009 revealed that nonsurgical management jumped from 0% during the first 4 years of the study to 75% in the last 4 years (P value less than .001).
The average length of stay declined significantly over the same period, from 26 days to 20 days (P less than .001), while complications trended downward from 50% vs. 33%, Dr. Michal Hubka reported on behalf of lead author Dr. Madhan Kumar Kuppusamy and their colleagues at Virginia Mason Medical Center in Seattle.
In all, 33 patients were managed nonoperatively and 48 surgically. Primary repair was the most common surgical approach (34 cases). Nonsurgical treatments included endoscopic stenting (11 cases), drainage procedures including mediastinal drainage (13 cases), total parenteral nutrition (7 cases), Dobhoff feeding tube (5 cases), gastrostomy (5 cases), endoscopic repair with clips or glue (3 cases), and feeding jejunostomy (3 cases).
"Nonoperative treatment options are increasing and surgeons must be able to apply these techniques to improve outcomes," Dr. Hubka said at the annual meeting of the Western Surgical Association.
Hybrid-type management was performed in 21% of patients, and most often took the form of endoscopic stents or drainage at the time of open or thorascopic drainage or decortication.
The nonoperative group was less likely than the operative group to experience pneumonia (4 patients vs. 7 patients) and dysrhythmias (4 patients vs. 11 patients), but more likely to experience persistent leak at the 14th day (3 vs. 2), stent migration (3 vs. 0), sepsis (1 vs. 0), and renal failure (1 vs. 0), Dr. Hubka said. Deep vein thrombosis occurred in one patient in each group.
Two patients managed medically died vs. one treated surgically (6% vs. 2%), for an overall mortality rate of 3.7%. A historical comparison of nine other studies involving nonoperative management of esophageal perforations presented by Dr. Hubka showed mortality rates reaching a high of 24% between 1973 and 1993 and a low of 3.8% between 1990 and 2001.
One of those nine studies identified a stepwise increase in mortality as time from perforation to diagnosis increased, with 5% of 75 patients dying with an immediate diagnosis vs. 14% with a diagnosis within 24 hours and 44% if the diagnosis occurred after 24 hours (Eur. J. Cardiothorac. Surg. 2003;23:799-804).
In all, 57 patients in the current analysis were treated within 24 hours and 24 were treated after 24 hours. Length of stay was significantly shorter in the early-treatment group at 15.6 days vs. 29.4 days in the late-treatment group; however, complications (20 vs. 11) and mortality (1 vs. 2) were similar, Dr. Hubka said.
"Time to diagnosis continues to be important; however, management in an experienced center facile with all current management techniques is the major issue affecting outcomes," he said.
The percentage of cases referred to the tertiary referral center was 50% from 1989 to 1992 and 79% from 2005 to 2009. Referred patients were significantly more likely to be treated more than 24 hours after perforation.
The improvement in outcomes is likely related to increasing diversity of treatment techniques and management in specialty centers, Dr. Hubka said.
Invited discussant Dr. Jeffrey Peters from the University of Rochester (N. Y.) Medical Center said, "What you heard was an increasing chorus of a paradigm change, if you will, that’s sort of paradoxical to most of us – that someone with a hole in their esophagus does better if you don’t operate on them. I still struggle trying not to do that when patients present in the emergency room with these issues."
Still, he described the improvement in outcomes as true progress for patients. Dr. Peters noted that the etiology of acute perforations has changed over time, with most now iatrogenic, and thus the benefit of early treatment may not be as critical as in years past. He said referral to a tertiary center is important, but that the paper did not prove a causal effect.
Based on the findings, Dr. Peters asked when surgeons should operate, how the size of the injury and presence of underlying disease should be taken into account in treatment decisions, and when surgery should be considered if nonoperative therapy fails.
Dr. Hubka said patients with larger esophageal tears or injuries and moderate mediastinal pleural contamination who can tolerate surgery are the ones who proceed to the operating room. He suggested that the study’s operative rate would likely have been higher if patients with perforations due to neoplasia or cancer had been included and that the presence of such underlying disease would surely push them toward operative management in clinical practice. Finally, if a patient becomes unstable or their level of contamination increases despite nonsurgical management, they would proceed to surgery and decontamination.