HSC Section 8_April 2017

Fig. 2. Comparison of facial nerve out- comes between study and control subjects by extent of resection

near-total resection when required in order to preserve facial nerve integrity. 6 In this updated series, 79.5% received GTR and 20.5% received partial resection. Over- all, 65% of patients had postoperative HB grade I or II function; 57% following gross total resection compared to 92% following partial resection. Anatomical continuity of the facial nerve was lost in 10 (13.7%) cases. A control group was not included in this latter analysis. In 2012, Gerganov et al. also found a much higher rate of poor facial nerve function (43% HB gr 3–6) after salvage surgery in patients who had prior radiation com- pared to nonradiated controls (30% HB gr 3–6). Gross total resection was performed in all but one control case. 9 In 2013, Husseini et al. compared surgical out- comes of 15 patients who had prior radiation therapy (13 sporadic, 2 NF2) to a control group consisting of 15 nonradiated VSs matched according to age, tumor size, and surgical approach. 8 Overall, 13 (87%) previously irradiated VSs were managed with gross total resection. The facial nerve was anatomically preserved in 14 of 15 cases, but only four of 13 (31%) patients who had HB grade I or II function preoperatively maintained HB I or II function following salvage surgery. None of the patients in any of these series experienced recurrence following salvage surgery, regardless of extent of

nerve outcome (HB I-II) between study and control sub- jects (Table V).

DISCUSSION We present a large multicenter case-control study comparing outcomes between postradiated sporadic VS and nonradiated control subjects. Among 37 sporadic VSs that failed primary SRS, our data demonstrate that approximately 77% of patients with normal preoperative facial nerve function retained good (HB I or II) function after salvage surgery, and GTR can be achieved in at least half of the cases. Overall, the rate of long-term facial nerve paresis was similar to a size-matched con- trol population of primary VS; however, the rate of less- than-complete resection was significantly higher in the postradiation group. In 2005, Friedman et al. similarly reviewed a series of 38 patients (28 sporadic, 10 NF2) who received sal- vage microsurgery following failed primary radiation therapy compared to a historical nonradiated control group matched according to age and tumor size. 7 Similar to the current study, they found that the rate of gross total resection was lower in the irradiated group (78.9% vs. 97.4%); however, in contrast they found that facial nerve outcomes were poorer than age- and size-matched nonradiated controls (37% vs. 70% HB I or II). The latter discrepancy may be partly explained by several differen- ces between study populations. In our study, the major- ity of patients (95%) underwent Gamma Knife (Elekta Instruments AB) radiosurgery after the year 2000 using current low-dose treatment parameters. In contrast, the study from the House Ear Clinic spanned the years between 1985 and 2004 and included heterogeneous radiation delivery: gamma knife, linear accelerator (LINAC), and proton beam. Their median length of follow-up was 15 months compared to 26 months. Finally, 10 of the 38 patients had NF2. Patients with NF2-associated VS are known to have poorer postopera- tive facial nerve outcomes compared to patients with sporadic VS. 15 In 2011, Friedman et al. reported an updated institutional experience, including 17 NF2 patients and 56 sporadic VSs, advocating for partial or

TABLE V. Multivariate Logistic Regression Analysis Investigating Independent Associations With Good Postoperative Facial Nerve Outcome (HB I-II).

Variable

Odds Ratio (95% CI)

P Value

Study vs. control subjects

0.61 (0.18–2.04)

0.424

Presurgical tumor size

0.69 (0.33–1.45)

0.332

Age

0.94 (0.87–1.01) 0.18 (0.02–1.37)

0.079 0.097

Extent of resection GTR vs. STR Extent of resection NTR vs. STR

0.67 (0.07–6.67)

0.730

Length of follow-up after salvage

0.92 (0.74–1.14)

0.436

CI 5 confidence interval; GTR 5 gross total resection; HB 5 House- Brackmann; NTR 5 near total resection; STR 5 subtotal resection.

Wise et al.: Surgical Salvage for Recurrent VS

187