Significantly reducing post-tonsillectomy haemorrhage requiring surgery by suturing the faucial pillars: a retrospective analysis.

BACKGROUND: The tonsillectomy is one of the most frequently performed surgical procedures. Given the comparatively frequent postsurgical bleeding associated with this procedure, particular attention has been paid to reduction of the postoperative bleeding rate. In 2006, we introduced routine suturin...

Full description

Bibliographic Details
Main Authors: Götz Senska, Hilal Schröder, Carolin Pütter, Philipp Dost
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2012-01-01
Series:PLoS ONE
Online Access:http://europepmc.org/articles/PMC3485309?pdf=render
Description
Summary:BACKGROUND: The tonsillectomy is one of the most frequently performed surgical procedures. Given the comparatively frequent postsurgical bleeding associated with this procedure, particular attention has been paid to reduction of the postoperative bleeding rate. In 2006, we introduced routine suturing of the faucial pillars at our clinic to reduce postoperative haemorrhage. METHODS: Two groups from the years 2003-2005 (n = 1000) and 2007-2009 (n = 1000) have been compared. We included all patients who had an elective tonsillectomy due to a benign, non-acute inflammatory tonsil illness. In the years 2007-2009, we additionally sutured the faucial pillars after completing haemostasis. For primary haemostasis we used suture ligation and bipolar diathermy. RESULTS: The rate of bleeding requiring second surgery for haemostasis was 3.6% in 2003-2005 but only 2.0% in 2007-2009 (absolute risk reduction 1.6% (95% CI 0.22%-2.45%, p = 0.04)). The median surgery time-including adenoidectomy and paracentesis surgery-increased from 25 to 31 minutes (p<0.01). CONCLUSIONS: We have been able to substantiate that suturing of the faucial pillars nearly halves the rate of postoperative haemorrhage. Surgery takes 8 minutes longer on average. Bleeding occurs later, mostly after 24 h. The limitations of this study relate to its retrospective character and all the potential biases related to observational studies.
ISSN:1932-6203