xRead - January 2023

TABLE I. Postoperative Clinical and Audiometric Outcomes for Case-Series.

P Value* (Short vs. Long-Term)

Audiometric/Clinical Outcomes

Short-Term

Long-Term

8.2 6 7.5 dB n 5 13 P < 0.001

Long: 15.3 6 11.3 dB n 5 10 P < 0.001

Postoperative mean ABG

0.255

Postoperative ABG 20 dB Postoperative ABG 10 dB

85% n 5 13 62% n 5 13

60% n 5 10 40% n 5 10

0.34

0.41

20.0 6 12.9 dB n 5 13 P < 0.001

32.9 6 13.0 dB n 5 10 P 5 0.063

0.020

Postoperative SRT

< 0.001

Partial/complete restenosis

0% n 5 21 0% n 5 21 9.5% n 5 21

64% n 5 14 7.1% n 5 14

Complete restenosis

0.40

Revision surgery

n/a

n/a

* P value refers to analysis of variance of repeated measures with post-hoc analysis. Bolded values meet a significance threshold of 0.05. P values within the short- and long-term columns are in comparison to preoperative values. ABG 5 air-bone gap; dB 5 decibel; long-term 5 > 2 years; short-term 5 < 2 years; SRT 5 speech recognition threshold.

45.2 months). Significantly more patients had some degree (partial to complete) of canal restenosis at long term (64%) follow-up when compared to short-term (0%), P < 0.001. Those patients with partial restenosis had significantly worse hearing outcomes at last follow-up (mean postoperative ABG 19.0 dB) versus those patients with normal EACs at last follow-up (mean ABG 4.9 dB, P 5 0.005). Only one patient (7.1%) experienced com plete restenosis; this occurred at long-term follow-up. The overall rate of revision surgery was 9.5% (2 of 21). One patient was revised for complete canal restenosis, whereas the other patient developed a perforation and moderate restenosis of the EAC. Systematic Review and Meta-Analysis. Initial search revealed 260 unique articles; after applying the inclusion and exclusion criteria, ultimately 11 articles were included for analysis. 5,8,10,13–20 Including our insti tution’s cohort, a total of 169 patients (62 males and 96 females; one study 20 of 11 patients did not specify gen der) were identified. Audiometric data from these stud ies were divided into short-term and long-term follow-up (Table II). Pooled preoperative ABG (29.3 6 9.7 dB, 95% CI 27.0–31.6) improved significantly both during short term (11.4 6 8.0 dB, 95% CI 8.3–14.5, P < 0.0001) and long-term (14.3 6 9.6 dB, 95% CI 11.6–16.9, P 5 0.0004) (Fig. 2). No significant deterioration in ABG was noted when comparing short-term to long-term data ( P 5 0.194). Surgical intervention successfully decreased the ABG to 20 db in 89.2% (95% CI 81.8%–94.3%) of patients during short-term follow-up, although only 65.4% (95% CI 55.9%–74.1%) maintained this during long-term follow-up ( P 5 0.010) (Fig. 3a of b). In addi tion, a 10 dB ABG closure was noted in 65.2% (95% CI 47.4%–81.0%) during short-term follow-up but in only 36.3% (95% CI 27.0%–46.4%) at long-term follow-up ( P 5 0.04). The long-term complete restenosis rate

dermatologic condition (lichen planus); eight (50%) patients had prior ear surgery, including contralateral canal surgery (n 5 4) and prior ipsilateral tympano plasty (n 5 3); and one patient was undergoing revision surgery (n 5 1) for prior ipsilateral canaloplasty not per formed by the senior author ( P . R . L ). Five (31.3%) patients underwent surgery for bilateral disease. All ears were dry at the time of surgery. One (4.8%) ear that had undergone a prior tympanoplasty had findings of a mid dle ear cholesteatoma intraoperatively. Mean follow-up time of the study population was 45.5 6 42.7 months. Postoperative Audiometric Outcomes. The mean preoperative ABG and SRT were 27.7 6 7.5 dB and 39.5 6 12.3 dB, respectively. Complete postoperative short term audiometric data was available for 13 ears, where as long-term data was available for 10 ears (Table I). Postoperative ABG was significantly improved compared to preoperative values at both short- and long-term fol low-ups ( P < 0.001 for both). No difference was noted between short-term and long-term ABG (8.2 6 7.5 dB and 15.3 6 11.3 dB, respectively, P 5 0.255). Surgical success, defined as an ABG 20 dB, was achieved in the majority of patients (11 of 13, 84.6% at short-term and 6 of 10, 60% at long-term). Whereas achievement of ABGs of 10 dB and 20 dB diminished over time, these differences were not significant when comparing short- and long-term rates ( P 5 0.41, P 5 0.34, respec tively). Speech reception thresholds were significantly improved only at short-term follow-up but failed to reach significance at long-term follow-up ( P 5 0.063). Speech reception threshold was noted to worsen from short-term to long-term follow-up ( P 5 0.020). Postoperative Clinical Outcomes. Although all 21 ears had short-term clinical data, 14 ears had long term clinical data available for review. Mean follow-up for long-term cohort was approximately 7 years (84.8 6

Laryngoscope 127: February 2017

Keller et al.: Retrospective Review and Meta-Analysis 491