Primary versus modified secondary closure techniques for persistent tracheocutaneous fistula in pediatric patients

Sung Joon Park, Sun A. Han, Tack Kyun Kwon, Myung Whun Sung, Seong Keun Kwon

Research output: Contribution to journalArticlepeer-review


Purpose: This study aimed at compating two closure techniques for tracheocutaneous fistulas (TCFs) in pediatric patients. Methods: A total of 106 consecutive pediatric patients who underwent closure of a persistent TCF between April 2007 and February 2021 at a tertiary pediatric hospital were evaluated, and 103 pediatric patients aged between 12 months and 18 years were included. The clinical characteristics, perioperative outcomes, and postoperative outcomes were compared between TCF closure by primary closure (Group 1) and a modified secondary healing technique (Group 2). Results: Of the 103 patients, 58 were classified into Group 1, and 45 into Group 2. The mean age at tracheostomy and TCF closure was significantly younger in Group 2, and the interval between decannulation to TCF closure was significantly shorter in Group 2. Procedural time and hospital stay were significantly shorter in Group 2 than Group 1. Group 2 had a significantly lower complication rate, need for revision surgery, and recannulation rate than Group 1. Conclusions: Modified secondary healing was more efficient in terms of procedural time and hospital stay, and safer (i.e., fewer complications). It is an effective surgical technique for closing a persistent TCF in younger patients more quickly after decannulation compared to primary closure.

Original languageEnglish
Pages (from-to)123-131
Number of pages9
JournalPediatric Surgery International
Issue number1
StatePublished - Jan 2022


  • Decannulation
  • Pediatric tracheostomy
  • Persistent tracheocutaneous fistula
  • Primary closure
  • Secondary healing


Dive into the research topics of 'Primary versus modified secondary closure techniques for persistent tracheocutaneous fistula in pediatric patients'. Together they form a unique fingerprint.

Cite this