2018 Section 6 - Laryngology, Voice Disorders, and Bronchoesophalogy

TABLE II. Pooled Outcomes of Additional Surgery and Decannulation Success for Surgical Procedures.

Additional Surgery, % (95% CI); I 2 %

Decannulation Success, % (95% CI); I 2 %

Technique

Studies

Patients

LTR with anastomosis

17

414

31% (26%-38%); 88%

89% (83%-93%); 0%

Open LTR, expansion graft/laryngoplasty

15

209

38& (31%-46%); 63%

83% (77%-88%); 0%

Endoscopic

5

73

44% (32%-56%); 51%

63% (47%-77%); 0%

CI 5 confidence interval; LTR 5 laryngotracheal reconstruction.

designed as case series, with an ideal global total of 16. Inclusion of comparative studies with noncomparative studies did not affect the mean for MINORS criteria among the studies. The methodological items that were most reported were the clearly stated aim, prospective collection of data, and endpoints appropriate to the aim of the study. However, length of follow-up, inclusion of consecutive patients, unbiased assessment of the study endpoints, rate of patients lost to follow-up, and prospective calculation of study size were most frequently not included. The range for noncomparative studies was five to 12, with a mean of 10. The range of comparative studies was 14 to 17, with a mean of 15. When including the comparative studies with the noncomparative studies, the mean remained 10. DISCUSSION The main findings of this study show that laryngotra- cheal resection with anastomosis decreases additional sur- gery and increases the rate of decannulation when compared to endoscopic procedures and LTR with augmentation/expan- sion grafting. Additionally, patients with idiopathic stenosis are less likely to receive additional surgery compared with those with trauma or intubation/tracheostomy as a cause of

difference between traumatic and intubation/tracheostomy (54% vs. 35%, P 5 .067). Etiology of stenosis, traumatic, intubation/tracheostomy, and idiopathic did have an impact on decannulation rates (78% vs. 88% vs. 63%, P < .001) (Fig. 3). Intubation/tracheostomy had the most successful rate of decannulation compared to trauma (88% vs. 78%, P < .001) and idiopathic (88% vs. 63%, P 5 .003). Trauma had a more successful rate than idiopathic stenosis (78% vs. 63%, P < .001). Of 39 articles, five 28–32 discussed the algorithm for patients receiving treatment, which were based on either Cotton-Meyer scoring or discussion of severity/grade of stenosis. Thirteen articles discussed grade of steno- sis, 1–3,6,8,11,13,14,25,27,28,30,31 and six articles discussed length of stenosis 1,17,24–26,28 as patient variables. There were 35 studies that discussed whether patients received additional surgery. Of these, 22/35 1,3,5–7,11,13,16,17,22–25,28–31,34,35,37–39 studies specified whether the patient had an open or endo- scopic procedure and 13/35 4,8,10,12,14,18–21,26,27,33,36 studies did not specify the further treatment. The level of bias, based on MINORS criteria, 47 did not impact the outcomes for decannulation success or additional surgery. There were 4 comparative studies, designed as cohort studies, with an ideal global total of 24, and 35 noncomparative studies,

Fig. 2. Subgroup analysis by technique. Blue indicates additional surgery, and red indicates decannulation success. LTR 5 laryngotra- cheal reconstruction. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]

Laryngoscope 127: January 2017

Lewis et al.: Surgical Treatment of Adult LTS

277