Experimental subglottic stenosis: Histopathologic and bronchoscopic comparison of electrosurgical, cryosurgical, and laser resection

Abstract

Controversy continues over the optimal method of resecting subglottic stenosis. Electrosurgery, cryosurgery and laser resection have all had some clinical success. We sought to compare these treatment methods in experimentally-created subglottic stenosis. Subglottic strictures were created transbronchoscopically in newborn lambs by electrocoagulating a rim of mucosa. Four weeks later, all had 20 to 90% occlusion documented bronchoscopically. Control animals were found to have dense submucosal fibrosis in the cricoid area. Remaining experimental animals were treated with: (1) electrosurgical resection; (2) cryosurgical treatment; (3) Neodymium-yag (Nd:Yag) laser at 20 watts; (4) Nd:Yag laser with 40 watts; or (5) carbon dioxide laser at 10 watts. Animals from each group were sacrificed at 5 and 30 days postresection and the tracheas were examined grossly and microscopically. Animals sacrificed at 30 days postresection were bronchoscoped at weekly intervals. Cryosurgical and electrosurgical resection resulted in 1-3 mm of thermal damage adjacent to the margin of resection at 5 days post-treatment. Mucosal ingrowth and healing were rapid and there was no development of stenosis or damage to underlying structures. Use of the Nd:Yag laser resulted in 10-20 mm of surrounding damage in animals sacrificed at 5 days. At 30 days posttreatment, there was mucosal healing but extreme underlying damage and redevelopment of stenosis at the level of treatment. The carbon dioxide laser-treated animals showed thermal damage present up to 2 mm in depth from resected areas at 5 days post-treatment. Mucosal regeneration proceeded rapidly. This study indicates that electrosurgical and cryosurgical resection for experimental subglottic stenosis create minimal thermal damage and are associated with comparably acceptable rates of mucosal ingrowth and healing. The carbon dioxide laser created comparable to slightly decreased thermal damage but suffers at present from lack of optimal visualization of the subglottic lesion.