Supplement Abstracts Submitted for the 70th Annual Scientific Meeting of the American College of Gastroenterology: ESOPHAGUS
Purpose: The purposes of this study were to determine if published studies of dilation of benign esophageal strictures with Savary dilators or balloons follow the rule of 3 (that no more than 3 dilators be passed after resistance is encountered −2 mm effective dilation), and to compare the reported complication rates between studies.
Methods: A systemic review of all papers describing esophageal dilation of benign esophageal strictures using Savary or balloon catheters was conducted. Exclusion criteria were inability to determine dilation protocol or complications, rings, webs, malignant strictures, achalasia, steroid injection, stents, or pediatric patients. Information on stricture type, dilation protocol, and complications was abstracted.
Results: 23 articles described balloon dilation in a total of 771 patients. None of these described dilation protocols which were consistent with the rule of 3 (2 mm dilation). A single study dilated patients by only 3 mm at a time, while most fully dilated patients to at least 15 or 20 mm in the initial session. 3 perforations were reported; no relationship between dilation protocol and complications could be established.
9 articles described Savary dilation in a total of 502 patients. 5 of these articles described a protocol more aggressive than the rule of 3 would dictate, though few dilated fully during the first session. 5 perforations were reported, 4 of these were guidewire perforations in a single study of strictures due to caustic ingestion. All perforations occured in studies dilating more aggressively that the rule of 3 would dictate.
Conclusions: Despite the widely quoted rule of 3, most authors elected to treat patients more rapidly that this would dictate with both Savary dilators and balloons. No randomized controlled trials of this maxim exist, nor would they be feasable, given the rarity of complications. The apparently higher perforation rate with rapid Savary dilation is driven entirely by guidewire perforations in one series. If we accept that these are not a function of the number of dilators used in a single session, then no relationship between dilation protocol and perforation rate is seen with either Savary dilators or balloons. The overall small number of patients described and multiple possible confounders make it difficult to draw strong conclusions from this study; however, it is apparent that no evidence for increased safety with the rule of 3 exists for Savary or balloon dilation of benign strictures.