2018 Section 5 - Rhinology and Allergic Disorders

E NDOSCOPIC SURGERY FOR SINONASAL ADENOCARCINOMA

TABLE 2. Patients’ characteristics among surgical groups

Characteristics

Endoscopic surgery

Endoscopic-assisted surgery

Open approaches

p value

Male:female ratio Age (mean 6 SD) ITAC:non-ITAC ratio

360:30

20:3

543:317

< .01 < .01 < .01 < .01 .11

64.3 6 13.3

59.7 6 12.4

61.1 6 12.8

242:35

11:5

612:100

No. of patients with wood dust exposure (%)

292 (67.8) 52 (12.8) 169 (41.4) 83 (20.3) 68 (16.7)

8 (25.8) 1 (3.7) 2 (7.4) 4 (14.8) 8 (29.6) 12 (44.4)

656 (51.7)

No. of T1 (%) No. of T2 (%) No. of T3 (%) No. of T4a (%) No. of T4b (%) No. of N0 (%) No. of N1 (%) No. of N2a (%) No. of N2b (%) No. of N2c (%)

56 (7.1)

244 (31)

178 (22.7) 188 (23.9) 120 (15.3) 603 (98.4)

36 (8.8)

253 (98.8)

2 (100)

.25

3 (1.2)

0 (0) 0 (0) 0 (0) 0 (0)

4 (0.7) 2 (0.3) 3 (0.5) 1 (0.2)

0 (0) 0 (0) 0 (0)

Abbreviation: ITAC, intestinal-type adenocarcinoma.

tions in sinonasal adenocarcinomas are numerous, includ- ing lack of facial incisions, excellent visualization and illumination of the surgical site, minimal trauma, shorter hospital stay, and lower costs. 57 However, any treatment in sinonasal malignancies must be primarily judged by its efficacy. In that respect, our study is not conclusive. This is indeed a pooled analysis of patient data, rather than a comparative meta-analysis. This is inevitable, as, up to now, there are no comparative studies (and certainly no randomized controlled trials) comparing endoscopic with external approaches for adenocarcinomas. The main problem in comparing different interventions is Simpson’s paradox (ie, the effect of case mix); in our case, the proportion of patients with T1/T2 versus T3/T4 tumors in the endoscopic versus the open approaches, which could erroneously lead to false conclusions regard- ing their efficacy. It is true that smaller tumors are more likely to be treated endoscopically, and this is indeed what we found within our data. However, comparing results by T classi- fication, we found that endoscopic management was asso- ciated with better OS and DFS across almost all tumor classifications.

For what concerns the outcome and survival, statistics varied among the articles, and, in some studies, data were not amenable for meta-analysis. 24,25,30–32,34 The mean follow-up time was 51.9 6 45.8 months (range, 1–360 months), with 46.4 6 37.6 months (range, 2–180 months) in the endoscopic surgery group and 53.6 6 47.6 months (range, 1–360 months) in the open approach group ( p 5 .09). Within the available data, 536 failure events were reported: 424 local (31.5%), 15 regional (1.1%), and 97 distant (7.2%) failures, resulting in crude disease-free survival (DFS) of 60.7% and local recurrence free sur- vival of 67.6%. The crude DFS, locoregional relapse-free survival, and overall survival (OS) according to T classifi- cation between endoscopic surgery and open approach groups is shown in Table 3. Univariate and multivariate logistic regression analysis demonstrated that advanced T classification and open approach are statistically related to a high rate of major complications (odds ratio [OR] 5 6.1; p < .01 and OR 5 3.5; p < .01, respectively). Univariate and multivariate Cox regression model showed that advanced T classification and open approach are stat- istically related to the high rate of local relapses (see Table 4). Regarding the OS, the univariate Cox regression model highlighted the relationship among non-ITAC, advanced T classification, and bad prognosis, albeit the statistical significance persisted only for advanced T clas- sification and open approach in a multivariate model (see Table 5). DISCUSSION Endoscopic surgery is increasingly and effectively used for sinonasal inflammatory diseases, even showing intra- cranial extension, and benign tumors 56 ; nevertheless, for malignant tumors, this approach is in its relative infancy. A comprehensive analysis of the existing evidence would help to serve as a barometer for the state-of-the-art and to suggest future directions. Given that different types of tumors have several treatment survival implications for patients, we focused this analysis on the surgical man- agement of sinonasal adenocarcinomas comparing out- comes between endoscopic surgery and traditional open approaches. The potential benefits of endoscopic resec-

FIGURE 3. Hospital stay according to the surgical techniques ( p < .01).

HEAD & NECK—DOI 10.1002/HED APRIL 2016

245