這是用戶在 2025-1-8 13:08 為 https://app.immersivetranslate.com/pdf-pro/52b23924-66da-4526-aa7c-2d09f3afe1e0 保存的雙語快照頁面，由 沉浸式翻譯 提供雙語支持。了解如何保存？

Hearing Preservation After Intervention in Vestibular Schwannoma
前庭神經鞘瘤手術後的聽力保護

*Kristen L. Yancey, †Samuel L. Barnett, *Walter Kutz, *Brandon Isaacson, ‡ ‡ ‡\ddagger Zabi Wardak, † † †\dagger Bruce Mickey, and *Jacob B. Hunter
克里斯汀·L·楊西、†塞缪爾·L·巴奈特、*瓦爾特·庫茨、*布蘭登·艾萨克森、 ‡ ‡ ‡\ddagger 扎比·瓦爾達克、 † † †\dagger 布魯斯·米奇和*雅各布·B·亨特。*Department of Otolaryngology-Head and Neck Surgery, University of Texas Southwestern Medical Center; †Department of Neurosurgery, University of Texas Southwestern Medical Center; ‡ ‡ ‡\ddagger Department of Radiation-Oncology, University of Texas Southwestern Medical
耳鼻喉科-頸部外科系, 德克薩斯大學西南醫學中心 神經外科系, 德克薩斯大學西南醫學中心 腫瘤放射治療科, 德克薩斯大學西南醫學中心

Center, Dallas, TX  達拉斯中心, TX

Abstract  摘要

Objective: This study aimed to assess the durability of audiological outcomes after radiation and surgery in the management of vestibular schwannoma. Study Design: Retrospective review. Setting: Tertiary academic center. Patients: Adults with sporadic vestibular schwannoma and serviceable hearing at the time of intervention. Interventions: Gamma Knife, middle cranial fossa, or retrosigmoid approaches. Main Outcome Measures: Pure-tone audiometry and speech discrimination scores. Results: Postintervention serviceable hearing (class A/B) was preserved in 70.4 % 70.4 % 70.4%70.4 \% ( n = 130 n = 130 n=130\mathrm{n}=130; mean follow-up, 3.31 yr; range, 0 − 15.25 yr ) 0 − 15.25 yr ) 0-15.25yr)0-15.25 \mathrm{yr}). Of the 49 patients treated with radiation, 19 ( 39.6 % 39.6 % 39.6%39.6 \% ) had serviceable hearing at last follow-up, compared with 38 ( 46.9 % 38 ( 46.9 % 38(46.9%38(46.9 \% of 81 ) ) )) who underwent retrosigmoid ( n = 36 n = 36 n=36\mathrm{n}=36 [44.4%]) and middle cranial fossa ( n = 45 n = 45 n=45\mathrm{n}=45 [55.6%]) approaches (odds ratio [OR], 1.40; 95% confidence interval [CI], 0.67 − 2.82 ; p = 0.47 0.67 − 2.82 ; p = 0.47 0.67-2.82;p=0.470.67-2.82 ; p=0.47 ). A matched analysis by age, tumor volume, and preintervention hearing ( n = 38 ) ( n = 38 ) (n=38)(\mathrm{n}=38) also found no difference in hearing preservation (HP)
目標：本研究旨在評估在治療前庭神經鞘瘤中使用放射線治療和手術後的聽力結果的持久性。 研究設計：回顧性審查。 設置：三級學術中心。 患者：在治療時具有可服務聽力的散發性前庭神經鞘瘤成人患者。 干預措施：γ刀、中窩及逆 S 形靜脈洞入路方法。 主要結果指標：純音聽力測試和言語識別分數。 結果：手術後可服務聽力（A/B 類）保留在(37.5%)（平均隨訪 3.31 年，範圍 1-6.9 年)。在 49 名接受放射治療的患者中，有 19 名(38.8%)在最後隨訪時保留可服務聽力，與接受逆 S 形靜脈洞入路（44.4%)和中窩（55.6%)方法的 81 名患者相比（比值比 1.40，95%置信區間 0.72-2.72)。根據年齡、腫瘤體積和術前聽力進行配對分析也發現聽力保留沒有差異。

likelihood between surgery and radiation (OR, 2.33; 95% CI, 0.24 − 35.91 ; p = 0.59 0.24 − 35.91 ; p = 0.59 0.24-35.91;p=0.590.24-35.91 ; p=0.59 ). After initial HP, 4 ( 9.5 % 9.5 % 9.5%9.5 \% ) surgical versus 10 ( 37.0 % ) 10 ( 37.0 % ) 10(37.0%)10(37.0 \%) radiated patients subsequently lost residual serviceable (A/B) hearing (OR, 0.18 ; 95 % CI , 0.06 − 0.69 ; p = 0.01 0.18 ; 95 % CI , 0.06 − 0.69 ; p = 0.01 0.18;95%CI,0.06-0.69;p=0.010.18 ; 95 \% \mathrm{CI}, 0.06-0.69 ; p=0.01 ) at a mean 3.74 ± 3.58 3.74 ± 3.58 3.74+-3.583.74 \pm 3.58 and 4.73 ± 3.83 4.73 ± 3.83 4.73+-3.834.73 \pm 3.83 years after surgery and radiation, respectively. Overall, 5 - and 10 -year HP rates (A/B) after initially successful HP surgery were 84.4 and 63.0 % 63.0 % 63.0%63.0 \%, respectively. However, survival estimates declined to 48.9 % 48.9 % 48.9%48.9 \% at 5 years and 32.7 % 32.7 % 32.7%32.7 \% at 10 years when patients with immediate postoperative serviceable hearing loss were also included, which were comparable to radiation-HP rates at 5 and 10 years of 28.0 and 14.2 % 14.2 % 14.2%14.2 \%, respectively ( p = 0.75 p = 0.75 p=0.75p=0.75 ).
手術與輻射療法間的可能性(OR,2.33;95%CI, 0.24 − 35.91 ; p = 0.59 0.24 − 35.91 ; p = 0.59 0.24-35.91;p=0.590.24-35.91 ; p=0.59 )。初次 HP 後,4( 9.5 % 9.5 % 9.5%9.5 \% )接受手術治療與 10 ( 37.0 % ) 10 ( 37.0 % ) 10(37.0%)10(37.0 \%) 接受輻射治療的患者最終失去(A/B)可利用的聽力(OR, 0.18 ; 95 % CI , 0.06 − 0.69 ; p = 0.01 0.18 ; 95 % CI , 0.06 − 0.69 ; p = 0.01 0.18;95%CI,0.06-0.69;p=0.010.18 ; 95 \% \mathrm{CI}, 0.06-0.69 ; p=0.01 ),平均於手術和輻射治療後 3.74 ± 3.58 3.74 ± 3.58 3.74+-3.583.74 \pm 3.58 和 4.73 ± 3.83 4.73 ± 3.83 4.73+-3.834.73 \pm 3.83 年出現。整體而言,初次成功 HP 手術 5 年和 10 年的 HP 率(A/B)分別為 84.4 和 63.0 % 63.0 % 63.0%63.0 \% 。然而,當將立即術後失去可利用聽力的患者納入考慮時,生存率下降至 5 年 48.9 % 48.9 % 48.9%48.9 \% 和 10 年 32.7 % 32.7 % 32.7%32.7 \% ,與輻射-HP 治療 5 年和 10 年分別為 28.0 和 14.2 % 14.2 % 14.2%14.2 \% 的率相當( p = 0.75 p = 0.75 p=0.75p=0.75 )。
Conclusions: After vestibular schwannoma intervention, overall HP was similar between radiated and surgical cohorts. However, when successful, surgical approaches offered more durable hearing outcomes at long-term follow-up.
前庭神經鞘瘤手術後,整體聽力保留相似,但手術治療能提供更持久的聽力效果。
Key Words: Hearing preservation-Middle cranial fossa-Retrosigmoid-Stereotactic radiosurgery-Vestibular schwannoma.
聽力保護-中顱窩-乙狀竇後-立體定向放射外科手術-聽神經瘤。
Otol Neurotol 43:e846-e855, 2022.
耳鼻喉科神經科學 43:e846-e855, 2022.

INTRODUCTION  介紹

The incidence of sporadic vestibular schwannoma (VS) continues to rise in part from increased use of magnetic resonance imaging (1), with a recent systematic review estimating 1 in 500 persons (2). After increased detection, more patients will require long-term management of these sometimes incidentally, diagnosed tumors. A frank discussion of associated risks with each approach (observation, radiation, and surgery) and eliciting priorities from patients
散發性前庭神經鞘腫瘤(VS)的發生率繼續上升,部分原因是磁共振成像的使用增加(1),最近的系統性回顧估計每 500 人中就有 1 人(2)。檢測增多後,更多患者將需要長期管理這些有時偶然診斷的腫瘤。需要與患者坦誠討論各種方法(觀察、輻射和手術)的相關風險,並了解患者的優先事項。
regarding hearing and facial nerve outcomes are the mainstays of modern VS management (3).
有關聽力及面神經結果是現代 VS 管理的主要支柱(3)。
When diagnosed early, patients may have good or even excellent hearing, and there is no consensus on optimal management, presenting challenging paradigms for patients to navigate. Although hearing loss was the most frequently identified greatest morbidity after surgery, reported by 26 % 26 % 26%26 \% of 1,940 surveyed patients (4), hearing preservation (HP) after intervention is associated with improved patient-reported quality of life (5). Even in ideal surgical candidates with small, medial tumors conducive to HP surgical approaches, there is no guarantee. However, with serial observation and radiation, there is uncertainty in the duration of maintaining serviceable hearing. Long-term data are slowly accumulating regarding hearing outcomes with each management approach (6-10), but outcomes beyond 5 ( 11 − 18 ) 5 ( 11 − 18 ) 5(11-18)5(11-18) and 10 ( 18 , 19 ) 10 ( 18 , 19 ) 10(18,19)10(18,19) years remain limited. As VS diagnoses are made at earlier time points in the tumors’ natural history, long-term HP data will aid in counseling patients presenting with serviceable hearing. The aim of this report was to evaluate our center’s hearing outcomes after surgery and radiation for VS management.
當早期診斷時,患者可能有良好甚至卓越的聽力,對於最佳管理方式並沒有共識,給予患者導航時帶來了挑戰性的範式。雖然聽力損失是手術後最常見的最大併發症,由 1,940 名接受調查的患者中的 26 % 26 % 26%26 \% 人報告(4),但手術後的聽力保留與改善患者自我報告的生活質量有關(5)。即使是適合手術的理想候選人,其腫瘤小、位於內耳,適合聽力保留手術方法,也無法保證成功。然而,通過連續觀察和放射治療,維持可用聽力的持續時間存在不確定性。關於每種管理方法的聽力結果,長期數據正在緩慢積累(6-10),但 5 ( 11 − 18 ) 5 ( 11 − 18 ) 5(11-18)5(11-18) 和 10 ( 18 , 19 ) 10 ( 18 , 19 ) 10(18,19)10(18,19) 年以后的結果仍然有限。隨著聽神經瘤診斷在腫瘤自然病史的更早階段,長期聽力保留的數據將有助於為有可用聽力的患者提供諮詢。本報告的目的是評估我們中心手術和放射治療管理聽神經瘤的聽力結果。

MATERIALS AND METHODS  材料與方法

Institutional review board approval was obtained (STU- 112016 − 040 112016 − 040 112016-040112016-040 ). Adult patients ( ≥ 18 yr ≥ 18 yr >= 18yr\geq 18 \mathrm{yr} ) diagnosed with sporadic VS undergoing surgery with an attempted HP approach (retrosigmoid [RS] or middle fossa craniotomy [MCF]) or Gamma Knife stereotactic radiosurgery (GK-SRS) between 2003 and 2019 were reviewed. Patients with neurofibromatosis type 2 were excluded.
機構審查委員會批准(STU- 112016 − 040 112016 − 040 112016-040112016-040 )。診斷為散發型聽神經瘤並於 2003 年至 2019 年接受手術(枕後窗[RS]或中窩開顱手術[MCF])或伽瑪刀立體定向放射治療(GK-SRS)的成年患者 ( ≥ 18 yr ≥ 18 yr >= 18yr\geq 18 \mathrm{yr} )接受了檢討。排除了患有神經纖維瘤 2 型的患者。
Data collected included patient demographics, preintervention and postintervention pure-tone audiometry and word recognition scores (WRS), and tumor characteristics. Tumor volumes were measured by serially segmenting axial slices from the last preintervention gadolinium-enhanced T1-weighted magnetic resonance imaging series using Slicer v.4.11.20210226, an open-source software for medical imaging informatics (20).
收集的數據包括患者的人口統計學特徵、介入前和介入後純音聽力測試和單詞辨認分數(WRS)以及腫瘤特徵。腫瘤體積通過使用開源醫學影像信息學軟件 Slicer v.4.11.20210226 分割最後一次介入前鈣化 T1 加權磁共振成像系列的軸向切片來測量。
Observation, radiation, and surgery are discussed in all cases, engaging in joint decision making with patients. Elderly and middle-aged patients with growing tumors and significant, medical comorbidities are more likely to undergo radiation at our center. Younger patients with primarily intracanalicular tumors and serviceable hearing typically undergo observation versus MCF depending on patient preference. An RS approach is offered to patients desiring surgery with serviceable hearing and larger tumors ( 1.5 − 1.5 − 1.5-1.5- to 2.0 − cm 2.0 − cm 2.0-cm2.0-\mathrm{cm} maximum axial dimension) and/or primarily cisternal components.
觀察、放射療法和手術是在所有病例中討論的,並與病患進行共同決策。患有不斷增長的腫瘤和重大醫療併發症的老年和中年患者更可能在我們的中心接受放射療法。主要具有顳內聲道腫瘤和可用聽力的較年輕患者,通常會根據患者的偏好選擇觀察或經乳突到中窩的手術。我們會向那些渴望接受手術、具有可用聽力和較大腫瘤(最大橫向尺寸介於 1.5 − 1.5 − 1.5-1.5- 2.0 − cm 2.0 − cm 2.0-cm2.0-\mathrm{cm}
Per center protocols, patients undergo GK-SRS with the Perfexion and Icon units, initially with a rigid-fixation Leksell stereotactic frame (Elekta Instruments, Atlanta, GA) and since 2017 with a thermoplastic mask. Tumors were treated to a single dose of 12 to 13 Gy at the 50 % 50 % 50%50 \% isodose line.
根據中心協議,患者接受 Perfexion 和 Icon 單位的 GK-SRS,最初使用剛性固定 Leksell 立體定向框架(Elekta Instruments,亞特蘭大,GA),自 2017 年起使用熱塑性面罩。腫瘤被治療為單劑量 12 至 13 Gy 的 50 % 50 % 50%50 \% 等劑量線。

Hearing Outcomes  聽力成果

Patients were categorized by hearing according to the Academy of Otolaryngology-Head and Neck Surgery (AAO-HNS) (21) and Gardner-Robertson (GR) classes (22), and by WRS alone (23) to facilitate comparisons with the existing body of literature. Pure-tone air conduction averages ( PTA AAOHNS PTA AAOHNS PTA_(AAOHNS)\mathrm{PTA}_{\mathrm{AAOHNS}} ) were calculated using the average of unaided thresholds at 500, 1,000, 2,000, 3 , 000 Hz 3 , 000 Hz 3,000Hz3,000 \mathrm{~Hz} for AAO-HNS classification and to create scattergrams
根據耳鼻喉-頭頸外科學會(AAO-HNS)(21)和 Gardner-Robertson(GR)等級(22)以及單純語音識別率(23),將患者依聽力狀況進行分類,以便與現有文獻進行比較。純音導聽閾值平均值( PTA AAOHNS PTA AAOHNS PTA_(AAOHNS)\mathrm{PTA}_{\mathrm{AAOHNS}} )使用 500、1,000、2,000 3 , 000 Hz 3 , 000 Hz 3,000Hz3,000 \mathrm{~Hz} 等頻率的未矯正閾值平均值計算,用於 AAO-HNS 分類和製作散點圖。
(24) in accordance with the 2012 consensus guidelines (25). Thresholds at 2,000 and 4 , 000 Hz 4 , 000 Hz 4,000Hz4,000 \mathrm{~Hz} were averaged when 3 , 000 Hz 3 , 000 Hz 3,000Hz3,000 \mathrm{~Hz} was not available. Pure-tone averages ( PTA GR PTA GR PTA_(GR)\mathrm{PTA}_{\mathrm{GR}} ) were also calculated as the average of unaided thresholds at 500 , 1 , 000 500 , 1 , 000 500,1,000500,1,000, and 2 , 000 Hz 2 , 000 Hz 2,000Hz2,000 \mathrm{~Hz} for GR classes. Serviceable hearing was defined as AAO-HNS class A or B, or GR I or II, corresponding to PTA of 50 dB or less and WRS of 50 % 50 % 50%50 \% or greater. The least restrictive classification of serviceable hearing was used for inclusion criteria.
(24) 根據 2012 共識指南(25)。當 3 , 000 Hz 3 , 000 Hz 3,000Hz3,000 \mathrm{~Hz} 不可用時,在 2,000 和 4 , 000 Hz 4 , 000 Hz 4,000Hz4,000 \mathrm{~Hz} 的閾值取平均。純音平均(< code2>)也計算為 500 , 1 , 000 500 , 1 , 000 500,1,000500,1,000 和 2 , 000 Hz 2 , 000 Hz 2,000Hz2,000 \mathrm{~Hz} GR 類的無輔助閾值的平均。可服務聽力定義為 AAO-HNS A 或 B 類,或 GR I 或 II 類,相當於 PTA 50 分貝或更低,WRS 為 50 % 50 % 50%50 \% 或更高。用最不嚴格的可服務聽力分類作為納入標準。
To account for age-related decline in hearing over time, A) PTAs and B) WRS scores were corrected using the contralateral ear by the following adjustments (26):
依照(26)的調整: A)純音聽力閾值 B)語詞識別分數
A) Ipsilateral Corrected PTA X X _(X)_{X} at Time " X " = = == Ipsilateral PTA X X _(X)_{X} - (Contralateral PTA - Contralateral PTA preintervention )
甲) 側同側校正後純音聽力 X X _(X)_{X} 在時間 " X " = = == 側同側純音聽力 X X _(X)_{X} - (對側純音聽力 - 對側介入前純音聽力)
B) Ipsilateral Corrected WRS X WRS X WRS_(X)\mathrm{WRS}_{\mathrm{X}} at Time " X " = = == Ipsilateral WRS X + ( X + _(X)+(:}_{\mathrm{X}}+\left(\right. Contralateral WRS preintervention − WRS preintervention  − WRS_("preintervention ")-\mathrm{WRS}_{\text {preintervention }}- Contralateral WRS X X _(X)_{\mathrm{X}} )
相同側已修正 WRS X WRS X WRS_(X)\mathrm{WRS}_{\mathrm{X}} 於時間「X」 = = == 相同側 WRS X + ( X + _(X)+(:}_{\mathrm{X}}+\left(\right. 對側 WRS preintervention − WRS preintervention  − WRS_("preintervention ")-\mathrm{WRS}_{\text {preintervention }}- 對側 WRS X X _(X)_{\mathrm{X}}
As a result, only hearing loss that occurred beyond any contralateral hearing decline was used to determine postintervention hearing classes.
因此,只有在任何對側聽力下降之外發生的聽力損失才用於確定術後聽力類別。

Statistical Analysis  統計分析

Statistical analysis was performed using Prism 8.3 (GraphPad Software Inc., La Jolla, CA). Fisher’s exact and t t tt tests were used to compare categorical and continuous variables between radiation and operative cohorts, respectively. Fisher’s exact tests were used to compare the overall likelihood of progressing to nonserviceable hearing by tumor characteristics and intervention modality. Kaplan-Meier (KM) curves were created to estimate rates of maintaining serviceable hearing over time. The level of statistical significance was set at p < 0.05 p < 0.05 p < 0.05p<0.05.
統計分析使用 Prism 8.3 (GraphPad Software Inc., La Jolla, CA)進行。Fisher's exact 和 t t tt 檢定用於比較放射線和手術隊列之間的類別和連續變量。Fisher's exact 檢定用於比較腫瘤特徵和干預方式對惡化至不可用聽力的總體可能性。創建 Kaplan-Meier (KM)曲線以估計隨時間維持可用聽力的比率。統計學顯著性水平設定為 p < 0.05 p < 0.05 p < 0.05p<0.05 。

RESULTS  結果

One hundred thirty patients were included for review (Table 1). Approximately two-thirds ( n = 81 ) ( n = 81 ) (n=81)(\mathrm{n}=81) of subjects underwent surgery, corresponding to 45 MCF and 36 RS approaches. The remaining 49 patients received radiation.
一百三十名病患被納入審查(表 1)。大約三分之二 ( n = 81 ) ( n = 81 ) (n=81)(\mathrm{n}=81) 的受試者接受了手術,其中 45 例為 MCF 方法,36 例為 RS 方法。其餘 49 名病患接受了放射治療。
TABLE 1. Summary of patient characteristics
表 1. 病患特性摘要
All  全部 Intervention  干預 OR 95% CI p p pp
Surgery  手術 Radiation  輻射
No. subjects (%)  無。科目（%） 130 81 (62.3) 49 (37.7)
Age at intervention, mean ± ± +-\pm SD, yr
年齡於介入時,平均 ± ± +-\pm 標準差,年		 50.97 ± 13.43 50.97 ± 13.43 50.97+-13.4350.97 \pm 13.43 46.58 ± 12.64 46.58 ± 12.64 46.58+-12.6446.58 \pm 12.64 59.64 ± 10.48 59.64 ± 10.48 59.64+-10.4859.64 \pm 10.48 - - < 0.01 < 0.01 < 0.01<0.01
Female sex, n (%)
女性性別, n (%)		 68 (52.3) 43 (53) 25 (51.0)
Preintervention hearing  預先介入聽證
WRS, mean ± ± +-\pm SD, %
物料需求標準,標準差,百分比		 92.54 ± 11.01 92.54 ± 11.01 92.54+-11.0192.54 \pm 11.01 94.1 ± 9.2 94.1 ± 9.2 94.1+-9.294.1 \pm 9.2 89.8 ± 13.2 89.8 ± 13.2 89.8+-13.289.8 \pm 13.2 - - 0.03
AAO-HNS PTA a a ^(a){ }^{a}, mean ± ± +-\pm SD (dB)
美國耳鼻喉科學會聽力餘量百分比,平均標準差(分貝)		 27.43 ± 12.43 27.43 ± 12.43 27.43+-12.4327.43 \pm 12.43 25.04 ± 11.87 25.04 ± 11.87 25.04+-11.8725.04 \pm 11.87 31.70 ± 12.24 31.70 ± 12.24 31.70+-12.2431.70 \pm 12.24 - - < 0.01 < 0.01 < 0.01<0.01
GR PTA b b ^(b){ }^{b}, mean ± ± +-\pm SD, dB
格雷耳培訓協會 b b ^(b){ }^{b} ，即 ± ± +-\pm 標準差，分貝		 24.67 ± 11.96 24.67 ± 11.96 24.67+-11.9624.67 \pm 11.96 22.98 ± 11.68 22.98 ± 11.68 22.98+-11.6822.98 \pm 11.68 27.53 ± 11.86 27.53 ± 11.86 27.53+-11.8627.53 \pm 11.86 - - 0.03
Tumor characteristics c c  ^("c "){ }^{\text {c }}
腫瘤特徵
Volume, mean ± ± +-\pm SD, cm 3 cm 3 cm^(3)\mathrm{cm}^{3}
體積、平均值 ± ± +-\pm 標準差、 cm 3 cm 3 cm^(3)\mathrm{cm}^{3} 		0.81 ± 1.00 0.81 ± 1.00 0.81+-1.000.81 \pm 1.00 0.59 ± 0.77 0.59 ± 0.77 0.59+-0.770.59 \pm 0.77 1.10 ± 1.19 1.10 ± 1.19 1.10+-1.191.10 \pm 1.19 - - < 0.01 < 0.01 < 0.01<0.01
Maximum linear dimension, cm
最大線性尺寸，公分		 1.24 ± 0.46 1.24 ± 0.46 1.24+-0.461.24 \pm 0.46 1.14 ± 0.45 1.14 ± 0.45 1.14+-0.451.14 \pm 0.45 1.40 ± 0.43 1.40 ± 0.43 1.40+-0.431.40 \pm 0.43 - - < 0.01 < 0.01 < 0.01<0.01
Intracanalicular only  內腔中 27 (25.2) 21 (34.4) 6 (13.0) 3.59 1.29-9.69 0.01
Fundal cap  胎底薄膜 77 (72.7) 48 (79.0) 29 (63.0) 2.16 0.93-4.84 0.09
All Intervention OR 95% CI p Surgery Radiation No. subjects (%) 130 81 (62.3) 49 (37.7) Age at intervention, mean +- SD, yr 50.97+-13.43 46.58+-12.64 59.64+-10.48 - - < 0.01 Female sex, n (%) 68 (52.3) 43 (53) 25 (51.0) Preintervention hearing WRS, mean +- SD, % 92.54+-11.01 94.1+-9.2 89.8+-13.2 - - 0.03 AAO-HNS PTA ^(a), mean +- SD (dB) 27.43+-12.43 25.04+-11.87 31.70+-12.24 - - < 0.01 GR PTA ^(b), mean +- SD, dB 24.67+-11.96 22.98+-11.68 27.53+-11.86 - - 0.03 Tumor characteristics ^("c ") Volume, mean +- SD, cm^(3) 0.81+-1.00 0.59+-0.77 1.10+-1.19 - - < 0.01 Maximum linear dimension, cm 1.24+-0.46 1.14+-0.45 1.40+-0.43 - - < 0.01 Intracanalicular only 27 (25.2) 21 (34.4) 6 (13.0) 3.59 1.29-9.69 0.01 Fundal cap 77 (72.7) 48 (79.0) 29 (63.0) 2.16 0.93-4.84 0.09| | All | Intervention | | OR | 95% CI | $p$ | | :---: | :---: | :---: | :---: | :---: | :---: | :---: | | | | Surgery | Radiation | | | | | No. subjects (%) | 130 | 81 (62.3) | 49 (37.7) | | | | | Age at intervention, mean $\pm$ SD, yr | $50.97 \pm 13.43$ | $46.58 \pm 12.64$ | $59.64 \pm 10.48$ | - | - | $<0.01$ | | Female sex, n (%) | 68 (52.3) | 43 (53) | 25 (51.0) | | | | | Preintervention hearing | | | | | | | | WRS, mean $\pm$ SD, % | $92.54 \pm 11.01$ | $94.1 \pm 9.2$ | $89.8 \pm 13.2$ | - | - | 0.03 | | AAO-HNS PTA ${ }^{a}$, mean $\pm$ SD (dB) | $27.43 \pm 12.43$ | $25.04 \pm 11.87$ | $31.70 \pm 12.24$ | - | - | $<0.01$ | | GR PTA ${ }^{b}$, mean $\pm$ SD, dB | $24.67 \pm 11.96$ | $22.98 \pm 11.68$ | $27.53 \pm 11.86$ | - | - | 0.03 | | Tumor characteristics ${ }^{\text {c }}$ | | | | | | | | Volume, mean $\pm$ SD, $\mathrm{cm}^{3}$ | $0.81 \pm 1.00$ | $0.59 \pm 0.77$ | $1.10 \pm 1.19$ | - | - | $<0.01$ | | Maximum linear dimension, cm | $1.24 \pm 0.46$ | $1.14 \pm 0.45$ | $1.40 \pm 0.43$ | - | - | $<0.01$ | | Intracanalicular only | 27 (25.2) | 21 (34.4) | 6 (13.0) | 3.59 | 1.29-9.69 | 0.01 | | Fundal cap | 77 (72.7) | 48 (79.0) | 29 (63.0) | 2.16 | 0.93-4.84 | 0.09 |
a a ^(a){ }^{a} PTA calculated as the average of air conduction thresholds at 500 , 1 , 000 , 2 , 000 500 , 1 , 000 , 2 , 000 500,1,000,2,000500,1,000,2,000, and 3,000 . The value 120 dB was substituted when the level of hearing loss was beyond the limits of the audiometer.
a a ^(a){ }^{a} PTA 經計算得出,為在 500 , 1 , 000 , 2 , 000 500 , 1 , 000 , 2 , 000 500,1,000,2,000500,1,000,2,000 及 3,000Hz 的氣導閾值的平均值。當聽力損失超過聽力計測量範圍時,以 120dB 代替。
b b ^(b){ }^{b} PTA calculated as the average of air conduction thresholds at 500 , 1 , 000 , 2 , 000 500 , 1 , 000 , 2 , 000 500,1,000,2,000500,1,000,2,000. and 120 dB was substituted when the level of hearing loss was beyond the limits of the audiometer.
聽導性閥值平均值
c c  ^("c "){ }^{\text {c }} There were 61 scans available for primary review for surgery patients and 46 for radiated patients.
有 61 次手術患者的掃描可用於主要審查,以及 46 次接受輻照治療患者的掃描。
AAO-HNS indicates American Academy of Otolaryngology-Head and Neck Surgery; CI, confidence interval; GR, Gardner-Robertson; OR, odds ratio; PTA, pure-tone average; SD, standard deviation; WRS, word recognition score.
美國耳鼻喉頭頸外科學會;信賴區間;Gardner-Robertson;優勢比;純音平均;標準差;語音識別分數。
Mean audiometric follow-up was 3.31 ± 3.40 3.31 ± 3.40 3.31+-3.403.31 \pm 3.40 years for all patients who maintained serviceable hearing at last follow-up. For patients with initial HP after intervention ( n = 73 n = 73 n=73\mathrm{n}=73 ), 24 had additional audiometric follow-up of at least 5 years ( n = 13 n = 13 n=13\mathrm{n}=13 surgery, n = 11 n = 11 n=11\mathrm{n}=11 radiation). For radiated patients with HP, mean audiometric follow-up was 3.40 ± 3.00 3.40 ± 3.00 3.40+-3.003.40 \pm 3.00 years. In the surgical cohort, mean audiometric follow-up was 3.27 ± 3.61 3.27 ± 3.61 3.27+-3.613.27 \pm 3.61 years for HP patients. For patients who had lost serviceable hearing at last follow-up, the mean time of deterioration to nonserviceable hearing was 3.07 ± 2.87 3.07 ± 2.87 3.07+-2.873.07 \pm 2.87 years for radiated patients and 5.32 ± 3.57 5.32 ± 3.57 5.32+-3.575.32 \pm 3.57 years for surgical patients ( p < 0.16 ) ( p < 0.16 ) (p < 0.16)(p<0.16).
平均聽力評估跟進時間為所有最終維持有用聽力的患者 3.31 ± 3.40 3.31 ± 3.40 3.31+-3.403.31 \pm 3.40 年。對於術後出現初次 HP 的患者( n = 73 n = 73 n=73\mathrm{n}=73 ),24 名患者至少有 5 年的聽力評估跟進( n = 13 n = 13 n=13\mathrm{n}=13 手術, n = 11 n = 11 n=11\mathrm{n}=11 放射療法)。對於放射療法組 HP 患者,平均聽力評估跟進時間為 3.40 ± 3.00 3.40 ± 3.00 3.40+-3.003.40 \pm 3.00 年。在手術組,HP 患者的平均聽力評估跟進時間為 3.27 ± 3.61 3.27 ± 3.61 3.27+-3.613.27 \pm 3.61 年。對於最終失去有用聽力的患者,放射療法組病人惡化至無法維持有用聽力的平均時間為 3.07 ± 2.87 3.07 ± 2.87 3.07+-2.873.07 \pm 2.87 年,手術組為 5.32 ± 3.57 5.32 ± 3.57 5.32+-3.575.32 \pm 3.57 年 ( p < 0.16 ) ( p < 0.16 ) (p < 0.16)(p<0.16) 。
Of the 42 ( 51.9 % 51.9 % 51.9%51.9 \% ) surgical patients with initial class A/B ( n = 43 GR I / II n = 43 GR I / II n=43GRI//II\mathrm{n}=43 \mathrm{GR} \mathrm{I} / \mathrm{II} ) hearing immediately postoperatively, 4 (9.5%) later fell to class C ( n = 3 ) C ( n = 3 ) C(n=3)C(n=3) or D ( n = 1 ) D ( n = 1 ) D(n=1)D(n=1) over an average of 3.74 years (range, 0.69 − 8.40 yr 0.69 − 8.40 yr 0.69-8.40yr0.69-8.40 \mathrm{yr} ), and three declined to GR III over an average of 4.76 years (range, 1.20 − 8.40 yr 1.20 − 8.40 yr 1.20-8.40yr1.20-8.40 \mathrm{yr}; Fig. 1). One patient lost hearing in the setting of subsequent growth of residual tumor and was later radiated. Twenty-seven of 49 ( 55.1 % ) 49 ( 55.1 % ) 49(55.1%)49(55.1 \%) radiated patients initially had class A/B after Gamma Knife, and 10 of 27 (37.0%) later declined to class C ( n = 1 ) C ( n = 1 ) C(n=1)C(n=1) and D ( n = 9 ) D ( n = 9 ) D(n=9)D(n=9), at an average of 4.73 years later (range, 1.22 − 14.13 yr 1.22 − 14.13 yr 1.22-14.13yr1.22-14.13 \mathrm{yr} ). By GR classes, 11 of 30 ( 36.7 % ) 30 ( 36.7 % ) 30(36.7%)30(36.7 \%) initially GR I/II radiated patients later declined to GR III ( n = 9 ) ( n = 9 ) (n=9)(\mathrm{n}=9) and GR V ( n = 2 ) ( n = 2 ) (n=2)(\mathrm{n}=2). Among all patients who initially had serviceable hearing after intervention (surgery and radiation), there were greater of odds of later transitioning to nonserviceable hearing among radiated patients (odds ratio [OR ] Class C/D , 5.59 [ 95 % ] Class C/D  , 5.59 [ 95 % ]_("Class C/D "),5.59[95%]_{\text {Class C/D }}, 5.59[95 \% confidence interval { CI } , 1.45 − 17.52 ; p = 0.01 ] ; OR GR { CI } , 1.45 − 17.52 ; p = 0.01 ] ; OR GR {CI},1.45-17.52;p=0.01];OR_(GR)\{\mathrm{CI}\}, 1.45-17.52 ; p=0.01] ; \mathrm{OR}_{\mathrm{GR}} III-v, 7.53 [ 95 % CI , 1.82 − 26.77 ; p < 0.01 ] 95 % CI , 1.82 − 26.77 ; p < 0.01 ] 95%CI,1.82-26.77;p < 0.01]95 \% \mathrm{CI}, 1.82-26.77 ; p<0.01] ). This finding suggests that when an intervention is initially successful in preserving
在最初術後出現 A/B 級( n = 43 GR I / II n = 43 GR I / II n=43GRI//II\mathrm{n}=43 \mathrm{GR} \mathrm{I} / \mathrm{II} )聽力的 42 例( 51.9 % 51.9 % 51.9%51.9 \% )手術患者中,4 例(9.5%)在平均 3.74 年(範圍為 0.69 − 8.40 yr 0.69 − 8.40 yr 0.69-8.40yr0.69-8.40 \mathrm{yr} )內下降至 C ( n = 3 ) C ( n = 3 ) C(n=3)C(n=3) 或 D ( n = 1 ) D ( n = 1 ) D(n=1)D(n=1) 級,另外 3 例在平均 4.76 年(範圍為 1.20 − 8.40 yr 1.20 − 8.40 yr 1.20-8.40yr1.20-8.40 \mathrm{yr} )內下降至 GR III 級(圖 1)。1 例患者在隨後腫瘤殘留生長的情況下喪失聽力,後來接受了放射治療。27 例接受伽馬刀治療的患者最初達到 A/B 級聽力,其中 10 例(37.0%)在平均 4.73 年後(範圍為 1.22 − 14.13 yr 1.22 − 14.13 yr 1.22-14.13yr1.22-14.13 \mathrm{yr} )下降至 C ( n = 1 ) C ( n = 1 ) C(n=1)C(n=1) 和 D ( n = 9 ) D ( n = 9 ) D(n=9)D(n=9) 級。在最初 GR I/II 級的 30 ( 36.7 % ) 30 ( 36.7 % ) 30(36.7%)30(36.7 \%) 例接受放射治療的患者中,11 例後來下降至 GR III ( n = 9 ) ( n = 9 ) (n=9)(\mathrm{n}=9) 和 GR V ( n = 2 ) ( n = 2 ) (n=2)(\mathrm{n}=2) 級。在所有術後最初有可用聽力的患者中,接受放射治療的患者後來轉向非可用聽力的機率更高(OR ] Class C/D , 5.59 [ 95 % ] Class C/D  , 5.59 [ 95 % ]_("Class C/D "),5.59[95%]_{\text {Class C/D }}, 5.59[95 \% 95%CI { CI } , 1.45 − 17.52 ; p = 0.01 ] ; OR GR { CI } , 1.45 − 17.52 ; p = 0.01 ] ; OR GR {CI},1.45-17.52;p=0.01];OR_(GR)\{\mathrm{CI}\}, 1.45-17.52 ; p=0.01] ; \mathrm{OR}_{\mathrm{GR}} 為 7.53 95 % CI , 1.82 − 26.77 ; p < 0.01 ] 95 % CI , 1.82 − 26.77 ; p < 0.01 ] 95%CI,1.82-26.77;p < 0.01]95 \% \mathrm{CI}, 1.82-26.77 ; p<0.01] )。這一研究結果表明,當一種治療方法最初成功保留了聽力時,放射治療可能會增加後續聽力下降的風險。
hearing, the subsequent likelihood of maintaining it is greater with surgery.
聽力, 手術後保持聽力的機率更高。
KM survival analyses also suggested different risks for progression to nonserviceable hearing between radiation and surgical cohorts, as the survival curves intersected. This intersection indicates that the hazard ratios were not consistent over time between groups, with more surgical patients losing serviceable hearing immediately postoperatively compared with the gradual transition to nonserviceable hearing classes among radiation patients. There were no differences in survivorship by log-rank tests between hearing classifications for either the surgery ( p = 0.64 p = 0.64 p=0.64p=0.64 ) or radiation ( p = 0.55 p = 0.55 p=0.55p=0.55 ) cohorts (data not displayed).
放射治療與手術隊列之間,將服務性聽力退化為無服務性聽力的風險不同,生存分析曲線有交叉。此交叉表示危險比在不同時間點之間不一致,手術患者立即失去服務性聽力,而放射治療患者則逐漸過渡到無服務性聽力。手術( p = 0.64 p = 0.64 p=0.64p=0.64 )或放射治療( p = 0.55 p = 0.55 p=0.55p=0.55 )隊列的任何聽力分級都沒有生存差異(未顯示數據)。
At last follow-up, 46.9 % ( n = 38 46.9 % ( n = 38 46.9%(n=3846.9 \%(n=38 of 81 ) ) )) of surgical patients maintained serviceable hearing with class A / B A / B A//B\mathrm{A} / \mathrm{B} and 48 % 48 % 48%48 \% ( n = 42 n = 42 n=42\mathrm{n}=42 of 81 ) with GR I/II (Table 2). By WRS classes, 46 ( 56.8 % 56.8 % 56.8%56.8 \% ) surgical patients had WRS of 50 % 50 % 50%50 \% or greater at last follow-up compared with 19 (38.8%) of 49 radiated patients ( p = 0.07 p = 0.07 p=0.07p=0.07 ). An MCF approach was more likely to result in HP ( n = 29 n = 29 n=29\mathrm{n}=29 of 45 [64.4%]) at last audiometric evaluation compared with an RS ( n = 9 n = 9 n=9\mathrm{n}=9 of 36 [25.0%]) approach (OR, 5.44; 95 % 95 % 95%95 \% CI, 1.98-14.07; p < 0.01 p < 0.01 p < 0.01p<0.01 ). Nineteen (38.8%) radiated patients maintained class A/B hearing, and 21 (43.8%) were GR I/II at last follow-up, with similar odds of HP as surgery ( p ≥ 0.36 p ≥ 0.36 p >= 0.36p \geq 0.36 ). Scatterplots of preintervention to postintervention PTA AAOHNS PTA AAOHNS PTA_(AAOHNS)\mathrm{PTA}_{\mathrm{AAOHNS}} and WRS are displayed in Figure 2.
81 名手術患者中， 46.9 % ( n = 38 46.9 % ( n = 38 46.9%(n=3846.9 \%(n=38 人保持可用聽力，達到 A / B A / B A//B\mathrm{A} / \mathrm{B} 級和 48 % 48 % 48%48 \% 級（佔 81 人中 n = 42 n = 42 n=42\mathrm{n}=42 人）。按 WRS 等級分類，與 49 名接受放射治療的患者（佔 p = 0.07 p = 0.07 p=0.07p=0.07 人）相比，最終追蹤時 56.8 % 56.8 % 56.8%56.8 \% 名手術患者 WRS 達到 50 % 50 % 50%50 \% 或更高。採用 MCF 手術方式的患者中， n = 29 n = 29 n=29\mathrm{n}=29 人（佔 45 人的 64.4%）在最後聽力評估中達到 HP 水平，而採用 RS 手術方式的患者中，則有 n = 9 n = 9 n=9\mathrm{n}=9 人（佔 36 人的 25.0%）達到 HP 水平（OR, 5.44; 95 % 95 % 95%95 \% CI, 1.98-14.07; p < 0.01 p < 0.01 p < 0.01p<0.01 ）。最終追蹤時， p ≥ 0.36 p ≥ 0.36 p >= 0.36p \geq 0.36 名接受放射治療的患者保持 A/B 級聽力，21 名（佔 43.8%）達到 GR I/II 級。圖 2 中顯示了治療前後 PTA 和 WRS 的散點圖。

Preintervention Hearing, Age, and Tumor Size
預介入聽力、年齡和腫瘤大小

Having preintervention class A hearing was associated with higher odds of maintaining serviceable hearing for both surgical (OR, 3.51; 95% CI, 1.31-8.84; p = 0.02 p = 0.02 p=0.02p=0.02 )
進行手術前具有 A 類聽力的人，保持可用聽力的機率較高(OR, 3.51; 95% CI, 1.31-8.84; p = 0.02 p = 0.02 p=0.02p=0.02 )。
No. at risk  無風險
GR 43 25 13 3 1
AAOHNS 42 22 12 3 1
No. at risk GR 43 25 13 3 1 AAOHNS 42 22 12 3 1| No. at risk | | | | | | | :--- | :--- | :--- | :--- | :--- | :--- | | GR | 43 | 25 | 13 | 3 | 1 | | AAOHNS | 42 | 22 | 12 | 3 | 1 |
FIG. 1. Durability of hearing preservation after initially successful hearing preservation surgical approach. Patients who lost hearing during or immediately after surgery were excluded from the model ( n = 39 ) ( n = 39 ) (n=39)(\mathrm{n}=39). Vertical dashes indicate censored patients, marking duration of follow-up beyond which no further audiometric data were available.
圖 1. 初次成功聽力保留手術後聽力保留的持久性。在手術中或手術後立即失去聽力的患者被排除在模型之外。垂直虛線表示受檢者被中止, 標記超出無法獲得進一步聽力學數據的隨訪期間。
Otology & Neurotology, Vol. 43, No. 8, 2022
耳科學和神經耳科學，第 43 卷，第 8 期，2022 年
TABLE 2. Hearing classification at last audiometric follow-up for patients undergoing (A) surgery ( n = 81 n = 81 n=81n=81 ) or (B) radiation ( n = 49 n = 49 n=49n=49 )
表 2. 接受 (A) 手術 ( n = 81 n = 81 n=81n=81 ) 或 (B) 輻射治療 ( n = 49 n = 49 n=49n=49 ) 患者在最後一次聽力檢查的分類
(A) Surgery  手術 Last Postoperative Audiogram
最後術後聽力檢查		 (B) Radiation  輻射 Last Postradiation Audiogram
末期放射後聽力圖		 OR 95% CI p p pp
Preoperative  術前 AAO-HNS a a  ^("a "){ }^{\text {a }}  美國耳鼻喉-頭頸外科學會 Preradiation  前輻射 AAO-HNS a a  ^("a "){ }^{\text {a }}  美國耳鼻喉頭頸外科學會
AAO-HNS Class  美國耳鼻喉學會分類 A B C D All, %  全部, % NS AAO-HNS Class  美國耳鼻喉學會分類 A B C D All, %  全部, % NS
A 18 13 1 23 55 (67.9) 24 A 4 9 0 9 22 (45.8) 9
B 2 5 6 13 26 (32.1) 19 B 0 5 6 13 25 (52.1) b b  ^("b "){ }^{\text {b }} 19
C - - - - 0 (0) - C 0 0 0 2 2 (4.2) 2
D - - - - 0 (0) - D - - - - 0 (0) -
All, %  全部, % 20 (24.7) 18 (22.2) 7 (8.6) 36 (44.4) 81 (100) 43 (53.1) All, %  全部, % 4 (8.3) 14 (29.2) 6 (12.5) 24 (50.0) 48 (98) 30 (62.5) 1.46 0.70-3.03 0.36
G R c G R c GR^(c)\mathbf{G R}^{c} GR c c  ^("c "){ }^{\text {c }}  代碼 0
GR Class  班級 I II III IV-V All, %  全部, % NS GR Class  班級 I II III IV-V All, %  全部, % NS
I 25 8 5 22 60 (74.1) 27 I 6 6 6 5 23 (47.9) 11
11 3 36 4 8 21 (25.9) 12 11 1 8 11 5 26 (54.2) b b  ^("b "){ }^{\text {b }} 16
III-IV - - - - 0 (0) - III-IV - - - - 0 (0) -
V - - - - 0 (0) - V - - - - 0 (0) -
All, %  全部, % 28 (34.6) 44 (54.3) 9 (11.1) 30 (37.0) 81 (100) 39 (48.2) All, %  全部, % 7 (14.6) 14 (29.2) 17 (35.4) 10 (20.8) 48 (98) 27 (57.4) 1.39 0.67-2.75 0.47
WRS d d  ^("d "){ }^{\text {d }}  資料運作服務 d d  ^("d "){ }^{\text {d }} WRS d d  ^("d "){ }^{\text {d }}  網絡資源服務 d d  ^("d "){ }^{\text {d }}
WRS Class  課堂學習 I II III IV All, %  全部, % 550% WRS Class  課堂學習 I II III IV All, %  全部, % <50%  50%
I 40 5 5 28 78 (96.3) 33 I 19 6 11 7 43 (87.8) 18
II 0 1 0 2 3 (3.7) 2 II 0 0 3 3 6 (12.2) 12
III - - - - 0 (0) - III - - - - 0 (0) -
IV - - - - 0 (0) - IV - - - - 0 (0) -
All, %  全部, % 40 (49.4) 6 (7.4) 5 (6.2) 30 (37.0) 81 (100) 35 (43.2) All, %  全部, % 19 (38.8) 6 (12.2) 14 (28.6) 10 (20.4) 49 (100) 30 (61.2) 2.08 1.00-4.22 0.07
(A) Surgery Last Postoperative Audiogram (B) Radiation Last Postradiation Audiogram OR 95% CI p Preoperative AAO-HNS ^("a ") Preradiation AAO-HNS ^("a ") AAO-HNS Class A B C D All, % NS AAO-HNS Class A B C D All, % NS A 18 13 1 23 55 (67.9) 24 A 4 9 0 9 22 (45.8) 9 B 2 5 6 13 26 (32.1) 19 B 0 5 6 13 25 (52.1) ^("b ") 19 C - - - - 0 (0) - C 0 0 0 2 2 (4.2) 2 D - - - - 0 (0) - D - - - - 0 (0) - All, % 20 (24.7) 18 (22.2) 7 (8.6) 36 (44.4) 81 (100) 43 (53.1) All, % 4 (8.3) 14 (29.2) 6 (12.5) 24 (50.0) 48 (98) 30 (62.5) 1.46 0.70-3.03 0.36 GR^(c) GR ^("c ") GR Class I II III IV-V All, % NS GR Class I II III IV-V All, % NS I 25 8 5 22 60 (74.1) 27 I 6 6 6 5 23 (47.9) 11 11 3 36 4 8 21 (25.9) 12 11 1 8 11 5 26 (54.2) ^("b ") 16 III-IV - - - - 0 (0) - III-IV - - - - 0 (0) - V - - - - 0 (0) - V - - - - 0 (0) - All, % 28 (34.6) 44 (54.3) 9 (11.1) 30 (37.0) 81 (100) 39 (48.2) All, % 7 (14.6) 14 (29.2) 17 (35.4) 10 (20.8) 48 (98) 27 (57.4) 1.39 0.67-2.75 0.47 WRS ^("d ") WRS ^("d ") WRS Class I II III IV All, % 550% WRS Class I II III IV All, % <50% I 40 5 5 28 78 (96.3) 33 I 19 6 11 7 43 (87.8) 18 II 0 1 0 2 3 (3.7) 2 II 0 0 3 3 6 (12.2) 12 III - - - - 0 (0) - III - - - - 0 (0) - IV - - - - 0 (0) - IV - - - - 0 (0) - All, % 40 (49.4) 6 (7.4) 5 (6.2) 30 (37.0) 81 (100) 35 (43.2) All, % 19 (38.8) 6 (12.2) 14 (28.6) 10 (20.4) 49 (100) 30 (61.2) 2.08 1.00-4.22 0.07| (A) Surgery | Last Postoperative Audiogram | | | | | | (B) Radiation | Last Postradiation Audiogram | | | | | | OR | 95% CI | $p$ | | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | | Preoperative | AAO-HNS ${ }^{\text {a }}$ | | | | | | Preradiation | AAO-HNS ${ }^{\text {a }}$ | | | | | | | | | | AAO-HNS Class | A | B | C | D | All, % | NS | AAO-HNS Class | A | B | C | D | All, % | NS | | | | | A | 18 | 13 | 1 | 23 | 55 (67.9) | 24 | A | 4 | 9 | 0 | 9 | 22 (45.8) | 9 | | | | | B | 2 | 5 | 6 | 13 | 26 (32.1) | 19 | B | 0 | 5 | 6 | 13 | 25 (52.1) ${ }^{\text {b }}$ | 19 | | | | | C | - | - | - | - | 0 (0) | - | C | 0 | 0 | 0 | 2 | 2 (4.2) | 2 | | | | | D | - | - | - | - | 0 (0) | - | D | - | - | - | - | 0 (0) | - | | | | | All, % | 20 (24.7) | 18 (22.2) | 7 (8.6) | 36 (44.4) | 81 (100) | 43 (53.1) | All, % | 4 (8.3) | 14 (29.2) | 6 (12.5) | 24 (50.0) | 48 (98) | 30 (62.5) | 1.46 | 0.70-3.03 | 0.36 | | | | | $\mathbf{G R}^{c}$ | | | | | | | GR ${ }^{\text {c }}$ | | | | | | | | GR Class | I | II | III | IV-V | All, % | NS | GR Class | I | II | III | IV-V | All, % | NS | | | | | I | 25 | 8 | 5 | 22 | 60 (74.1) | 27 | I | 6 | 6 | 6 | 5 | 23 (47.9) | 11 | | | | | 11 | 3 | 36 | 4 | 8 | 21 (25.9) | 12 | 11 | 1 | 8 | 11 | 5 | 26 (54.2) ${ }^{\text {b }}$ | 16 | | | | | III-IV | - | - | - | - | 0 (0) | - | III-IV | - | - | - | - | 0 (0) | - | | | | | V | - | - | - | - | 0 (0) | - | V | - | - | - | - | 0 (0) | - | | | | | All, % | 28 (34.6) | 44 (54.3) | 9 (11.1) | 30 (37.0) | 81 (100) | 39 (48.2) | All, % | 7 (14.6) | 14 (29.2) | 17 (35.4) | 10 (20.8) | 48 (98) | 27 (57.4) | 1.39 | 0.67-2.75 | 0.47 | | | | | WRS ${ }^{\text {d }}$ | | | | | | | WRS ${ }^{\text {d }}$ | | | | | | | | WRS Class | I | II | III | IV | All, % | 550% | WRS Class | I | II | III | IV | All, % | <50% | | | | | I | 40 | 5 | 5 | 28 | 78 (96.3) | 33 | I | 19 | 6 | 11 | 7 | 43 (87.8) | 18 | | | | | II | 0 | 1 | 0 | 2 | 3 (3.7) | 2 | II | 0 | 0 | 3 | 3 | 6 (12.2) | 12 | | | | | III | - | - | - | - | 0 (0) | - | III | - | - | - | - | 0 (0) | - | | | | | IV | - | - | - | - | 0 (0) | - | IV | - | - | - | - | 0 (0) | - | | | | | All, % | 40 (49.4) | 6 (7.4) | 5 (6.2) | 30 (37.0) | 81 (100) | 35 (43.2) | All, % | 19 (38.8) | 6 (12.2) | 14 (28.6) | 10 (20.4) | 49 (100) | 30 (61.2) | 2.08 | 1.00-4.22 | 0.07 |
There were no statistically significant differences in the odds of maintaining serviceable hearing or WRS of 50 % 50 % 50%50 \% or greater between cohorts by Fisher’s exact tests. Mean and median time to last audiometric follow-up were 3.3 and 2.2 years (range, 0.1 − 15.3 yr 0.1 − 15.3 yr 0.1-15.3yr0.1-15.3 \mathrm{yr} ) in patients with serviceable hearing at last follow-up, respectively. Mean and median time of deterioration to nonserviceable hearing were 3.4 and 2.8 years
在 Fisher's 精確檢驗中,維持可用聽力或語音辨識能力 50 % 50 % 50%50 \% 或以上的幾率在各組之間沒有統計上顯著的差異。在最後一次聽力檢查中有可用聽力的患者,平均和中位數的最後追蹤時間分別為 3.3 年和 2.2 年( 0.1 − 15.3 yr 0.1 − 15.3 yr 0.1-15.3yr0.1-15.3 \mathrm{yr} )。聽力惡化至不可用的平均和中位數時間為 3.4 年和 2.8 年。
a a _(a){ }_{a} (range, 0.3 − 14.1 0.3 − 14.1 0.3-14.10.3-14.1 yr), respectively. HNS hearing classes: A, PTA ≤ 30 dB ; ≤ 30 dB ; <= 30dB;\leq 30 \mathrm{~dB} ; WRS ≥ 70 % ; ≥ 70 % ; >= 70%;\geq 70 \% ; B, PTA 31 − 50 dB 31 − 50 dB 31-50dB31-50 \mathrm{~dB}; WRS ≥ 50 % ; ≥ 50 % ; >= 50%;\geq 50 \% ; C, PTA ≥ 51 dB ; ≥ 51 dB ; >= 51dB;\geq 51 \mathrm{~dB} ; WRS ≥ 50 % ; ≥ 50 % ; >= 50%;\geq 50 \% ; D, any PTA; WRS ≤ 49 % ≤ 49 % <= 49%\leq 49 \%. PTA calculated as the average of air conduction thresholds at 500 , 1 , 000 , 2 , 000 500 , 1 , 000 , 2 , 000 500,1,000,2,000500,1,000,2,000, and 3 , 000 Hz 3 , 000 Hz 3,000Hz3,000 \mathrm{~Hz}. The value 120 dB was substituted when the level of hearing loss was beyond the limits of the audiometer.
a a _(a){ }_{a} (範圍, 0.3 − 14.1 0.3 − 14.1 0.3-14.10.3-14.1 年)。HNS 聽力類別: A, PTA ≤ 30 dB ; ≤ 30 dB ; <= 30dB;\leq 30 \mathrm{~dB} ; WRS ≥ 70 % ; ≥ 70 % ; >= 70%;\geq 70 \% ; B, PTA 31 − 50 dB 31 − 50 dB 31-50dB31-50 \mathrm{~dB} ; WRS ≥ 50 % ; ≥ 50 % ; >= 50%;\geq 50 \% ; C, PTA ≥ 51 dB ; ≥ 51 dB ; >= 51dB;\geq 51 \mathrm{~dB} ; WRS ≥ 50 % ; ≥ 50 % ; >= 50%;\geq 50 \% ; D, 任何 PTA; WRS ≤ 49 % ≤ 49 % <= 49%\leq 49 \% 。PTA 計算為 500 , 1 , 000 , 2 , 000 500 , 1 , 000 , 2 , 000 500,1,000,2,000500,1,000,2,000 和 3 , 000 Hz 3 , 000 Hz 3,000Hz3,000 \mathrm{~Hz} 氣導閾值的平均值。當聽力損失超出聽力計的限度時,以 120 dB 代替。
GR hearing classes: I, PTA ≤ 30 dB ≤ 30 dB <= 30dB\leq 30 \mathrm{~dB} and WRS ≥ 70 % ≥ 70 % >= 70%\geq 70 \%; II, PTA 31 − 50 dB 31 − 50 dB 31-50dB31-50 \mathrm{~dB} and WRS 50 − 69 % 50 − 69 % 50-69%50-69 \%; III, PTA 51 − 90 dB 51 − 90 dB 51-90dB51-90 \mathrm{~dB} and WRS 5 − 49 % 5 − 49 % 5-49%5-49 \%; IV, PTA ≥ 91 dB ≥ 91 dB >= 91dB\geq 91 \mathrm{~dB} and WRS 1-4% (there were no patients who fell within this category); V, no response. PTA calculated as the average of air conduction thresholds at 500 , 1 , 000 500 , 1 , 000 500,1,000500,1,000, and 2 , 000 Hz 2 , 000 Hz 2,000Hz2,000 \mathrm{~Hz}. The value 120 dB was substituted when the level of hearing loss was beyond testing capacity.
聽力檢查類別：I 類，PTA ≤ 30 dB ≤ 30 dB <= 30dB\leq 30 \mathrm{~dB} 和 WRS ≥ 70 % ≥ 70 % >= 70%\geq 70 \% ；II 類，PTA 31 − 50 dB 31 − 50 dB 31-50dB31-50 \mathrm{~dB} 和 WRS 50 − 69 % 50 − 69 % 50-69%50-69 \% ；III 類，PTA 51 − 90 dB 51 − 90 dB 51-90dB51-90 \mathrm{~dB} 和 WRS 5 − 49 % 5 − 49 % 5-49%5-49 \% ；IV 類，PTA ≥ 91 dB ≥ 91 dB >= 91dB\geq 91 \mathrm{~dB} 和 WRS 1-4%（此類別無患者）；V 類，無反應。PTA 計算為空氣傳導閾值在 500 , 1 , 000 500 , 1 , 000 500,1,000500,1,000 和 2 , 000 Hz 2 , 000 Hz 2,000Hz2,000 \mathrm{~Hz} 的平均值。當聽力損失超出測試能力時，採用 120 dB 作為替代值。
AAO-HNS indicates American Academy of Otolaryngology-Head and Neck Surgery; GR, Gardner-Robertson; NS, nonserviceable (defined as AAO-HNS class C or D, or GD III-IV; PTA, pure-tone average; WRS, word recognition score.
美國耳鼻喉頭頸外科學會;格納-羅伯遜;非醫療服務(定義為美國耳鼻喉頭頸外科學會 C 或 D 類,或格納-羅伯遜 III-IV 級);純音平均;語詞識別分數。
FIG. 2. Scattergrams for preintervention and postintervention hearing at last audiometric follow-up in patients who underwent surgery ( A A AA ) or radiation (B). The left-side diagrams display pure-tone averages (PTA) on the y y yy axis and word recognition scores (WRS) on the x x xx axis. The changes in PTA and WRS correspond the y y yy and x x xx axes, respectively, on the diagrams to the right. The numbers indicate how many subjects correspond to a given grid’s representative PTA and WRS.
圖 2。接受手術(A)或放射治療(B)的患者,在最後一次聽力測試的治療前後聽力分散圖。左側圖表顯示純音平均值(PTA)在 x 軸,語音辨識分數(WRS)在 y 軸。右側圖表分別顯示 PTA 和 WRS 的變化。數字表示有多少受試者的 PTA 和 WRS 對應於相同的網格。
and radiation cohorts (OR, 5.49; 95% CI, 1.38-19.60; p = 0.01 p = 0.01 p=0.01p=0.01 ). There was no statistically significant difference in age at intervention between patients with and without HP, for either surgery ( p = 0.13 ) ( p = 0.13 ) (p=0.13)(p=0.13) or radiation cohorts ( p = 0.22 ) ( p = 0.22 ) (p=0.22)(p=0.22).
放射性組(OR, 5.49; 95% CI, 1.38-19.60; p = 0.01 p = 0.01 p=0.01p=0.01 )。患有或未患有 HP 的患者在手術 ( p = 0.13 ) ( p = 0.13 ) (p=0.13)(p=0.13) 或放射性組 ( p = 0.22 ) ( p = 0.22 ) (p=0.22)(p=0.22) 之間的干預年齡差異沒有統計學意義。
Tumor volumes between patients with and without serviceable hearing immediately postoperatively were 0.50 ± 0.74 0.50 ± 0.74 0.50+-0.740.50 \pm 0.74 and 0.72 ± 0.78 cm 3 0.72 ± 0.78 cm 3 0.72+-0.78cm^(3)0.72 \pm 0.78 \mathrm{~cm}^{3}, respectively ( p = 0.18 p = 0.18 p=0.18p=0.18 ). Similarly, there was no statistically significant difference in tumor volumes between radiation patients with and without serviceable hearing at first audiometric follow-up ( 1.03 ± 0.90 1.03 ± 0.90 1.03+-0.901.03 \pm 0.90 versus 1.19 ± 1.51 cm 3 ; p = 0.66 1.19 ± 1.51 cm 3 ; p = 0.66 1.19+-1.51cm^(3);p=0.661.19 \pm 1.51 \mathrm{~cm}^{3} ; p=0.66 ). The presence of a fundal fluid cap in surgical patients did not confer an HP advantage ( p > 0.99 p > 0.99 p > 0.99p>0.99 ).
手術後立即,有和無可聽覺功能的病人腫瘤體積分別為 0.50 ± 0.74 0.50 ± 0.74 0.50+-0.740.50 \pm 0.74 和 0.72 ± 0.78 cm 3 0.72 ± 0.78 cm 3 0.72+-0.78cm^(3)0.72 \pm 0.78 \mathrm{~cm}^{3} 。同樣地,接受放射治療的病人在首次聽力跟踪時,有和無可聽覺功能的腫瘤體積沒有統計學上的顯著差異( 1.03 ± 0.90 1.03 ± 0.90 1.03+-0.901.03 \pm 0.90 比 1.19 ± 1.51 cm 3 ; p = 0.66 1.19 ± 1.51 cm 3 ; p = 0.66 1.19+-1.51cm^(3);p=0.661.19 \pm 1.51 \mathrm{~cm}^{3} ; p=0.66 )。手術病人存在基底液體囊並不能提供 HP 優勢( p > 0.99 p > 0.99 p > 0.99p>0.99 )。
Eight GK-SRS patients were not offered surgery because of age greater than 70 years ( n = 7 ) ( n = 7 ) (n=7)(n=7) or large size ( n = 1 ) ( n = 1 ) (n=1)(n=1). On average, patients treated with radiation were older (mean difference, 11.62 yr ; p < 0.01 11.62 yr ; p < 0.01 11.62yr;p < 0.0111.62 \mathrm{yr} ; p<0.01 ) and had larger tumors (mean difference, 0.51 cm 3 ; p < 0.01 0.51 cm 3 ; p < 0.01 0.51cm^(3);p < 0.010.51 \mathrm{~cm}^{3} ; p<0.01 ) and worse hearing by WRS (mean difference, 4.3 % ; p = 0.03 4.3 % ; p = 0.03 4.3%;p=0.034.3 \% ; p=0.03 ) and PTA (mean difference PTA AAOHNS, 6.7 dB ; p < 0.01 6.7 dB ; p < 0.01 6.7dB;p < 0.016.7 \mathrm{~dB} ; p<0.01 ) at the time of intervention, evaluated at an average of 3.11 ± 2.83 3.11 ± 2.83 3.11+-2.833.11 \pm 2.83 months before treatment. Given that preintervention hearing, age, and tumor size have been associated with the likelihood of HP, a subgroup analysis was conducted. Surgical ( n = 19 ) ( n = 19 ) (n=19)(\mathrm{n}=19) and radiation ( n = 19 ) ( n = 19 ) (n=19)(\mathrm{n}=19) cohorts were matched by preintervention
八位 GK-SRS 患者因年齡大於 70 歲 ( n = 7 ) ( n = 7 ) (n=7)(n=7) 或腫瘤較大 ( n = 1 ) ( n = 1 ) (n=1)(n=1) 而未接受手術。接受放射治療的患者平均年齡更大(平均差異 11.62 yr ; p < 0.01 11.62 yr ; p < 0.01 11.62yr;p < 0.0111.62 \mathrm{yr} ; p<0.01 )、腫瘤更大(平均差異 0.51 cm 3 ; p < 0.01 0.51 cm 3 ; p < 0.01 0.51cm^(3);p < 0.010.51 \mathrm{~cm}^{3} ; p<0.01 )以及 WRS(平均差異 4.3 % ; p = 0.03 4.3 % ; p = 0.03 4.3%;p=0.034.3 \% ; p=0.03 )和 PTA(平均差異 PTA AAOHNS、 6.7 dB ; p < 0.01 6.7 dB ; p < 0.01 6.7dB;p < 0.016.7 \mathrm{~dB} ; p<0.01 )聽力更差,在治療前平均 3.11 ± 2.83 3.11 ± 2.83 3.11+-2.833.11 \pm 2.83 個月進行評估。鑒於治療前的聽力、年齡和腫瘤大小與 HP 發生率有關,因此進行了亞組分析。手術 ( n = 19 ) ( n = 19 ) (n=19)(\mathrm{n}=19) 和放射治療 ( n = 19 ) ( n = 19 ) (n=19)(\mathrm{n}=19) 組根據治療前的特徵進行了配對。
hearing (WRS within 10 % 10 % 10%10 \%, PTA within 10 dB ), tumor size (within 0.5 cm 3 0.5 cm 3 0.5cm^(3)0.5 \mathrm{~cm}^{3} ), and age at intervention (within 5 years). t t tt Tests confirmed that there were no differences in these preintervention parameters between cohorts or duration of follow-up. There were no statistically significant differences in the proportion of patients maintaining serviceable hearing or WRS of 50 % 50 % 50%50 \% or greater at long-term follow-up between surgery and radiation (Table 3).
聽力(WRS 位於 10 % 10 % 10%10 \% ，PTA 位於 10 dB 內)、腫瘤大小(位於 0.5 cm 3 0.5 cm 3 0.5cm^(3)0.5 \mathrm{~cm}^{3} )和干預年齡(位於 5 年內)。 t t tt 測試確認這些干預前參數在隊列之間或跟進期間沒有差異。在長期跟進中,在維持可用聽力或 WRS 50 % 50 % 50%50 \% 或更高方面,手術和放射療法之間沒有統計學上顯著的差異(表 3)。

DISCUSSION  討論

A growing body of literature on long-term VS hearing outcomes will aid in patient counseling, as loss to follow-up and lack of class I evidence have precluded consensus regarding optimal management for patients presenting with good hearing and small ( < 2 cm < 2 cm < 2cm<2 \mathrm{~cm} ) tumors. Previous meta-analyses have encountered several challenges in accounting for considerable heterogeneity within the literature. These include the following: 1) the use of different classifications to describe hearing outcomes, 2) shrinking sample sizes at longer follow-up intervals, 3) varying radiation modalities with different dosimetry and fractionation protocols, and 4) teasing out contributions from other variables ( 12 , 27 , 28 ) ( 12 , 27 , 28 ) (12,27,28)(12,27,28). The most robust predictors of maintaining serviceable hearing (class A / B , GR I / II A / B , GR I / II A//B,GRI//II\mathrm{A} / \mathrm{B}, \mathrm{GR} \mathrm{I} / \mathrm{II} ) are preintervention
聽力保護的長期結果文獻正在不斷增加,這有助於病人輔導,因為失訪率高和缺乏一級證據,使人們難以對良好聽力和小型(< 3 cm)腫瘤的患者達成共識的最佳管理方式。先前的 meta 分析在處理文獻中的大異質性方面遇到了幾個挑戰。這些挑戰包括:1)使用不同的分類來描述聽力結果,2)隨著隨訪時間的延長,樣本量不斷縮減,3)採用不同的輻射手段,劑量學和分割方案也各不相同,4)區分其他變數的貢獻。保持可使用聽力(第二類)的最可靠預測因素是治療前
TABLE 3. Hearing classifications for patients undergoing surgery ( n = 9 n = 9 n=9n=9 via middle cranial fossa; n = 10 n = 10 n=10n=10 retrosigmoid) and Gamma Knife radiation ( n = 19 n = 19 n=19n=19 ) matched by age, tumor size,
聽力分類表 3。患者接受手術 ( n = 9 n = 9 n=9n=9 中窩顱底; n = 10 n = 10 n=10n=10 乙狀峽後路徑) 和 Gamma 刀放射療法 ( n = 19 n = 19 n=19n=19 ) 的患者,按年齡、腫瘤大小相匹配
Surgery  手術 Radiation  輻射 OR ( 95 % CI 95 % CI 95%CI95 \% \mathrm{CI} )  或 ( 95 % CI 95 % CI 95%CI95 \% \mathrm{CI} ) p p pp
Preoperative  術前 1 yr  1 年 3 yr  3 年 5 yr  5 年 10 yr  10 年 Pre-GK  前國民教育 1 yr  一年 3 yr  3 年 5 yr  5 年 10 yr  10 年
Age at intervention, mean ± ± +-\pm SD, yr
年齡於介入時,平均 ± ± +-\pm 標準差,年		 56.21 ± 8.03 56.21 ± 8.03 56.21+-8.0356.21 \pm 8.03 - - - - 56.42 ± 10.64 56.42 ± 10.64 56.42+-10.6456.42 \pm 10.64 - - - - - 0.95
Volume, mean ± SD , cm 3 ± SD , cm 3 +-SD,cm^(3)\pm \mathrm{SD}, \mathrm{cm}^{3}
體積，平均值		 0.74 ± 0.86 Lost a 0.74 ± 0.86  Lost  a {:[0.74+-0.86],[" Lost "^(a)]:}\begin{gathered} 0.74 \pm 0.86 \\ \text { Lost }^{a} \end{gathered} - 4 ( 21.1 ) 4 ( 21.1 ) 4(21.1)4(21.1) 5 ( 26.3 ) 5 ( 26.3 ) 5(26.3)5(26.3) 7 (36.8) 7  (36.8)  7" (36.8) "7 \text { (36.8) } 0.80 ± 0.79 0.80 ± 0.79 0.80+-0.790.80 \pm 0.79 - 2 ( 10.5 ) ― 2 ( 10.5 ) ¯ 2( bar(10.5))2(\overline{10.5)} 4 ( 21.1 ) ― 4 ( 21.1 ) ¯ 4 bar((21.1))4 \overline{(21.1)} 5 ( 26.3 ) 5 ( 26.3 ) 5(26.3)5(26.3) - 0.81
AAO-HNS, n (%)  美國耳鼻喉頭頸外科學會, n (%)
A 12 (63.2) 4 (21.1) 3 (15.8) 2 (10.5) 0 (0) 11 (57.9) 6 (31.6) 2 (10.5) 0 (0) 0 (0) - -
B 7 (36.8) 5 (26.3) 2 (10.5) 1 (5.3) 2 (10.5) 8 (42.1) 5 (26.3) 4 (21.1) 2 (10.5) 1 (5.3) - -
C 0 (0) 1 (5.3) 1 (5.3) 1 (5.3) 1 (5.3) 0 (0) 2 (10.5) 3 (15.8) 1 (5.3) 1 (5.3) - -
D 0 (0) 9 (47.4) 9 (47.4) 10 (52.6) 9 (47.4) 0 (0) 6 (31.6) 8 (42.1) 12 (63.2) 12 (63.2) - -
Serviceable  可用的 19 (100) 9 (47.4) 5 (26.3) 3 (15.8) 2 (10.5) 19 (100) 11 (57.9) 6 (31.6) 2 (10.5) 1 (5.3) 2.33 (0.24-35.91) 0.59
PTA b PTA b PTA^(b)\mathrm{PTA}^{b}, mean ± SD , dB ± SD , dB +-SD,dB\pm \mathrm{SD}, \mathrm{dB}
PTA b PTA b PTA^(b)\mathrm{PTA}^{b} , 意味著 ± SD , dB ± SD , dB +-SD,dB\pm \mathrm{SD}, \mathrm{dB} 		25.92 ± 14.14 25.92 ± 14.14 25.92+-14.1425.92 \pm 14.14 27.36 ± 12.39 27.36 ± 12.39 27.36+-12.3927.36 \pm 12.39 - 0.75
Gardner-Robertson, n (%)
高德納-羅伯遜, n (%)
I 13 (68.4) 5 (26.3) 3 (15.8) 2 (10.5) 0 (0) 12 (63.2) 6 (31.6) 2 (10.5) 0 (0) 0 (0) - -
II 6 (31.6) 5 (26.3) 2 (10.5) 1 (5.3) 2 (10.5) 7 (36.8) 7 (36.8) 4 (21.1) 2 (10.5) 1 (5.3) - -
III 0 (0) 1 (5.3) 2 (10.5) 2 (10.5) 2 (10.5) 0 (0) 4 (21.1) 7 (36.8) 7 (36.8) 7 (36.8) - -
IV 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) - -
V 0 (0) 8 (42.1) 8 (42.1) 8 (42.1) 8 (42.1) 0 (0) 2 (10.5) 4 (21.1) 6 (31.6) 6 (31.6) - -
Serviceable  可供使用的 19 (100) 10 (52.6) 5 (26.3) 3 (15.8) 2 (10.5) 19 (100) 13 (68.4) 6 (31.6) 2 (10.5) 1 (5.3) 2.33 (0.24-35.91) 0.59
PTA c PTA c PTA^(c)\mathrm{PTA}^{c}, mean ± SD , dB ± SD , dB +-SD,dB\pm \mathrm{SD}, \mathrm{dB}
PTA c PTA c PTA^(c)\mathrm{PTA}^{c} , 表示 ± SD , dB ± SD , dB +-SD,dB\pm \mathrm{SD}, \mathrm{dB} 		23.27 ± 13.40 23.27 ± 13.40 23.27+-13.4023.27 \pm 13.40 22.88 ± 11.50 22.88 ± 11.50 22.88+-11.5022.88 \pm 11.50 - 0.92
WRS d , n ( % ) WRS d , n ( % ) WRS^(d),n(%)\mathrm{WRS}^{d}, \mathrm{n}(\%)
I 17 (89.5) 8 (42.1) 5 (26.3) 3 (15.8) 2 (10.5) 18 (94.7) 11 (57.9) 7 (36.8) 2 (10.5) 0 (0) - -
II 2 (10.2) 2 (10.2) 1 (5.3) 0 (0) 1 (5.3) 1 (5.3) 2 (10.2) 2 (10.2) 1 (5.3) 1 (5.3) - -
III 0 (0) 1 (5.3) 1 (5.3) 2 (10.2) 1 (5.3) 0 (0) 4 (21.1) 4 (21.1) 6 (31.6) 7 (36.8) - -
IV 0 (0) 8 (42.1) 8 (42.1) 8 (42.1) 8 (42.1) 0 (0) 2 (10.2) 4 (21.1) 6 (31.6) 6 (31.6) - -
N ≥ 50 % N ≥ 50 % N >= 50%\mathrm{N} \geq 50 \% 19 (100) 7 (36.8) 6 (31.6) 3 (15.8) 3 (15.8) 19 (100) 13 (68.4) 9 (47.4) 3 (15.8) 2 (10.5) 1.77 (0.31-11.17) 0.65
Mean ± ± +-\pm SD, %
平均值 ± ± +-\pm 標準差，百分比		 92.6 ± 12.5 92.6 ± 12.5 92.6+-12.592.6 \pm 12.5 94.3 ± 8.9 94.3 ± 8.9 94.3+-8.994.3 \pm 8.9 - 0.63
Surgery Radiation OR ( 95%CI ) p Preoperative 1 yr 3 yr 5 yr 10 yr Pre-GK 1 yr 3 yr 5 yr 10 yr Age at intervention, mean +- SD, yr 56.21+-8.03 - - - - 56.42+-10.64 - - - - - 0.95 Volume, mean +-SD,cm^(3) "0.74+-0.86 Lost ^(a)" - 4(21.1) 5(26.3) 7" (36.8) " 0.80+-0.79 - 2( bar(10.5)) 4 bar((21.1)) 5(26.3) - 0.81 AAO-HNS, n (%) A 12 (63.2) 4 (21.1) 3 (15.8) 2 (10.5) 0 (0) 11 (57.9) 6 (31.6) 2 (10.5) 0 (0) 0 (0) - - B 7 (36.8) 5 (26.3) 2 (10.5) 1 (5.3) 2 (10.5) 8 (42.1) 5 (26.3) 4 (21.1) 2 (10.5) 1 (5.3) - - C 0 (0) 1 (5.3) 1 (5.3) 1 (5.3) 1 (5.3) 0 (0) 2 (10.5) 3 (15.8) 1 (5.3) 1 (5.3) - - D 0 (0) 9 (47.4) 9 (47.4) 10 (52.6) 9 (47.4) 0 (0) 6 (31.6) 8 (42.1) 12 (63.2) 12 (63.2) - - Serviceable 19 (100) 9 (47.4) 5 (26.3) 3 (15.8) 2 (10.5) 19 (100) 11 (57.9) 6 (31.6) 2 (10.5) 1 (5.3) 2.33 (0.24-35.91) 0.59 PTA^(b), mean +-SD,dB 25.92+-14.14 27.36+-12.39 - 0.75 Gardner-Robertson, n (%) I 13 (68.4) 5 (26.3) 3 (15.8) 2 (10.5) 0 (0) 12 (63.2) 6 (31.6) 2 (10.5) 0 (0) 0 (0) - - II 6 (31.6) 5 (26.3) 2 (10.5) 1 (5.3) 2 (10.5) 7 (36.8) 7 (36.8) 4 (21.1) 2 (10.5) 1 (5.3) - - III 0 (0) 1 (5.3) 2 (10.5) 2 (10.5) 2 (10.5) 0 (0) 4 (21.1) 7 (36.8) 7 (36.8) 7 (36.8) - - IV 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) - - V 0 (0) 8 (42.1) 8 (42.1) 8 (42.1) 8 (42.1) 0 (0) 2 (10.5) 4 (21.1) 6 (31.6) 6 (31.6) - - Serviceable 19 (100) 10 (52.6) 5 (26.3) 3 (15.8) 2 (10.5) 19 (100) 13 (68.4) 6 (31.6) 2 (10.5) 1 (5.3) 2.33 (0.24-35.91) 0.59 PTA^(c), mean +-SD,dB 23.27+-13.40 22.88+-11.50 - 0.92 WRS^(d),n(%) I 17 (89.5) 8 (42.1) 5 (26.3) 3 (15.8) 2 (10.5) 18 (94.7) 11 (57.9) 7 (36.8) 2 (10.5) 0 (0) - - II 2 (10.2) 2 (10.2) 1 (5.3) 0 (0) 1 (5.3) 1 (5.3) 2 (10.2) 2 (10.2) 1 (5.3) 1 (5.3) - - III 0 (0) 1 (5.3) 1 (5.3) 2 (10.2) 1 (5.3) 0 (0) 4 (21.1) 4 (21.1) 6 (31.6) 7 (36.8) - - IV 0 (0) 8 (42.1) 8 (42.1) 8 (42.1) 8 (42.1) 0 (0) 2 (10.2) 4 (21.1) 6 (31.6) 6 (31.6) - - N >= 50% 19 (100) 7 (36.8) 6 (31.6) 3 (15.8) 3 (15.8) 19 (100) 13 (68.4) 9 (47.4) 3 (15.8) 2 (10.5) 1.77 (0.31-11.17) 0.65 Mean +- SD, % 92.6+-12.5 94.3+-8.9 - 0.63| | Surgery | | | | | Radiation | | | | | OR ( $95 \% \mathrm{CI}$ ) | $p$ | | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | | | Preoperative | 1 yr | 3 yr | 5 yr | 10 yr | Pre-GK | 1 yr | 3 yr | 5 yr | 10 yr | | | | Age at intervention, mean $\pm$ SD, yr | $56.21 \pm 8.03$ | - | - | - | - | $56.42 \pm 10.64$ | - | - | - | - | - | 0.95 | | Volume, mean $\pm \mathrm{SD}, \mathrm{cm}^{3}$ | $\begin{gathered} 0.74 \pm 0.86 \\ \text { Lost }^{a} \end{gathered}$ | - | $4(21.1)$ | $5(26.3)$ | $7 \text { (36.8) }$ | $0.80 \pm 0.79$ | - | $2(\overline{10.5)}$ | $4 \overline{(21.1)}$ | $5(26.3)$ | - | 0.81 | | AAO-HNS, n (%) | | | | | | | | | | | | | | A | 12 (63.2) | 4 (21.1) | 3 (15.8) | 2 (10.5) | 0 (0) | 11 (57.9) | 6 (31.6) | 2 (10.5) | 0 (0) | 0 (0) | - | - | | B | 7 (36.8) | 5 (26.3) | 2 (10.5) | 1 (5.3) | 2 (10.5) | 8 (42.1) | 5 (26.3) | 4 (21.1) | 2 (10.5) | 1 (5.3) | - | - | | C | 0 (0) | 1 (5.3) | 1 (5.3) | 1 (5.3) | 1 (5.3) | 0 (0) | 2 (10.5) | 3 (15.8) | 1 (5.3) | 1 (5.3) | - | - | | D | 0 (0) | 9 (47.4) | 9 (47.4) | 10 (52.6) | 9 (47.4) | 0 (0) | 6 (31.6) | 8 (42.1) | 12 (63.2) | 12 (63.2) | - | - | | Serviceable | 19 (100) | 9 (47.4) | 5 (26.3) | 3 (15.8) | 2 (10.5) | 19 (100) | 11 (57.9) | 6 (31.6) | 2 (10.5) | 1 (5.3) | 2.33 (0.24-35.91) | 0.59 | | $\mathrm{PTA}^{b}$, mean $\pm \mathrm{SD}, \mathrm{dB}$ | $25.92 \pm 14.14$ | | | | | $27.36 \pm 12.39$ | | | | | - | 0.75 | | Gardner-Robertson, n (%) | | | | | | | | | | | | | | I | 13 (68.4) | 5 (26.3) | 3 (15.8) | 2 (10.5) | 0 (0) | 12 (63.2) | 6 (31.6) | 2 (10.5) | 0 (0) | 0 (0) | - | - | | II | 6 (31.6) | 5 (26.3) | 2 (10.5) | 1 (5.3) | 2 (10.5) | 7 (36.8) | 7 (36.8) | 4 (21.1) | 2 (10.5) | 1 (5.3) | - | - | | III | 0 (0) | 1 (5.3) | 2 (10.5) | 2 (10.5) | 2 (10.5) | 0 (0) | 4 (21.1) | 7 (36.8) | 7 (36.8) | 7 (36.8) | - | - | | IV | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | - | - | | V | 0 (0) | 8 (42.1) | 8 (42.1) | 8 (42.1) | 8 (42.1) | 0 (0) | 2 (10.5) | 4 (21.1) | 6 (31.6) | 6 (31.6) | - | - | | Serviceable | 19 (100) | 10 (52.6) | 5 (26.3) | 3 (15.8) | 2 (10.5) | 19 (100) | 13 (68.4) | 6 (31.6) | 2 (10.5) | 1 (5.3) | 2.33 (0.24-35.91) | 0.59 | | $\mathrm{PTA}^{c}$, mean $\pm \mathrm{SD}, \mathrm{dB}$ | $23.27 \pm 13.40$ | | | | | $22.88 \pm 11.50$ | | | | | - | 0.92 | | $\mathrm{WRS}^{d}, \mathrm{n}(\%)$ | | | | | | | | | | | | | | I | 17 (89.5) | 8 (42.1) | 5 (26.3) | 3 (15.8) | 2 (10.5) | 18 (94.7) | 11 (57.9) | 7 (36.8) | 2 (10.5) | 0 (0) | - | - | | II | 2 (10.2) | 2 (10.2) | 1 (5.3) | 0 (0) | 1 (5.3) | 1 (5.3) | 2 (10.2) | 2 (10.2) | 1 (5.3) | 1 (5.3) | - | - | | III | 0 (0) | 1 (5.3) | 1 (5.3) | 2 (10.2) | 1 (5.3) | 0 (0) | 4 (21.1) | 4 (21.1) | 6 (31.6) | 7 (36.8) | - | - | | IV | 0 (0) | 8 (42.1) | 8 (42.1) | 8 (42.1) | 8 (42.1) | 0 (0) | 2 (10.2) | 4 (21.1) | 6 (31.6) | 6 (31.6) | - | - | | $\mathrm{N} \geq 50 \%$ | 19 (100) | 7 (36.8) | 6 (31.6) | 3 (15.8) | 3 (15.8) | 19 (100) | 13 (68.4) | 9 (47.4) | 3 (15.8) | 2 (10.5) | 1.77 (0.31-11.17) | 0.65 | | Mean $\pm$ SD, % | $92.6 \pm 12.5$ | | | | | $94.3 \pm 8.9$ | | | | | - | 0.63 |
Median time to last audiometric follow-up was 2.75 years (interquartile range, 2.37 yr ) in patients with serviceable hearing at last follow-up and was comparable between radiated and surgical patients ( p > 0.99 p > 0.99 p > 0.99p>0.99 ). Median time of deterioration to nonserviceable hearing was 1.20 years (interquartile range, 3.10 yr ) for all patients. There was no significant difference in the odds of maintaining serviceable hearing (AAO-HNS A / B A / B A//B\mathrm{A} / \mathrm{B}, GR I/II) or WRS ≥ 50 % ≥ 50 % >= 50%\geq 50 \% at long-term follow-up between radiation and surgical cohorts.
被服務聽力保持最後追蹤時間的中位數為 2.75 年(四分位數範圍,2.37 年),並且在輻射和手術患者之間是可比較的。所有患者惡化到無法服務的聽力的中位數時間為 1.20 年(四分位數範圍,3.10 年)。在長期追蹤期間,維持可服務聽力(AAO-HNS、GR I/II)或 WRS 的機會在放射和手術隊列之間沒有顯著差異。
b b ^(b){ }^{b} AAO-HNS hearing classes: A, PTA ≤ 30 dB ; ≤ 30 dB ; <= 30dB;\leq 30 \mathrm{~dB} ; WRS ≥ 70 % ; ≥ 70 % ; >= 70%;\geq 70 \% ; B, PTA 31 − 50 dB ; 31 − 50 dB ; 31-50dB;31-50 \mathrm{~dB} ; WRS ≥ 50 % ; ≥ 50 % ; >= 50%;\geq 50 \% ; C, PTA ≥ 51 dB ; ≥ 51 dB ; >= 51dB;\geq 51 \mathrm{~dB} ; WRS ≥ 50 % ; ≥ 50 % ; >= 50%;\geq 50 \% ; D, any PTA; WRS ≤ 49 % ≤ 49 % <= 49%\leq 49 \%. PTA calculated as the average of air conduction thresholds 500 , 100 , 2 , 000 500 , 100 , 2 , 000 500,100,2,000500,100,2,000, and 3 , 000 Hz 3 , 000 Hz 3,000Hz3,000 \mathrm{~Hz}. The value 12 dB . calculated as the average of air conduction thresholds at 500 , 1 , 000 500 , 1 , 000 500,1,000500,1,000, and 2 , 000 Hz 2 , 000 Hz 2,000Hz2,000 \mathrm{~Hz}. The value 120 dB was substituted when the level of hearing loss was beyond testing capacity.
b b ^(b){ }^{b} AAO-HNS 聽力等級:A, PTA ≤ 30 dB ; ≤ 30 dB ; <= 30dB;\leq 30 \mathrm{~dB} ; WRS ≥ 70 % ; ≥ 70 % ; >= 70%;\geq 70 \% ; B, PTA 31 − 50 dB ; 31 − 50 dB ; 31-50dB;31-50 \mathrm{~dB} ; WRS ≥ 50 % ; ≥ 50 % ; >= 50%;\geq 50 \% ; C, PTA ≥ 51 dB ; ≥ 51 dB ; >= 51dB;\geq 51 \mathrm{~dB} ; WRS ≥ 50 % ; ≥ 50 % ; >= 50%;\geq 50 \% ; D, 任意 PTA; WRS ≤ 49 % ≤ 49 % <= 49%\leq 49 \% 。 PTA 以氣導閾值的平均值計算 500 , 100 , 2 , 000 500 , 100 , 2 , 000 500,100,2,000500,100,2,000 , 和 3 , 000 Hz 3 , 000 Hz 3,000Hz3,000 \mathrm{~Hz} 。 值為 12 dB 。 以氣導閾值的平均值計算 500 , 1 , 000 500 , 1 , 000 500,1,000500,1,000 , 和 2 , 000 Hz 2 , 000 Hz 2,000Hz2,000 \mathrm{~Hz} 。 若聽力損失超出測試範圍,則以 120 dB 代替。
AAO-HNS indicates American Academy of Otolaryngology-Head and Neck Surgery; CI, confidence interval; GK, Gamma Knife; PTA, pure-tone average; SD, standard deviation; WRS, word recognition score.
美國耳鼻喉頭頸外科學院;CI,信賴區間;GK,伽瑪刀;PTA,純音平均;SD,標準差;WRS,語音辨識分數。
hearing ( 5 , 7 , 9 , 23 , 26 , 29 − 32 5 , 7 , 9 , 23 , 26 , 29 − 32 5,7,9,23,26,29-325,7,9,23,26,29-32 ), smaller tumor size (31-33) or volumes (29), lower marginal tumor ( ≤ 12 Gy ≤ 12 Gy <= 12Gy\leq 12 \mathrm{~Gy} ) and cochlear dose ( < 4 Gy < 4 Gy < 4Gy<4 \mathrm{~Gy} ) (34-39) in radiated patients, and the presence of a fluid fundal cap ( 40 , 41 ) ( 40 , 41 ) (40,41)(40,41) in surgical candidates (12). Following Gamma Knife, poorer baseline hearing not only affects the likelihood of retaining serviceable hearing but is also associated with a steeper rate of decline (42).
放射療法患者中聽力( 5 , 7 , 9 , 23 , 26 , 29 − 32 5 , 7 , 9 , 23 , 26 , 29 − 32 5,7,9,23,26,29-325,7,9,23,26,29-32 )、較小腫瘤大小(31-33)或體積(29)、較低邊緣腫瘤( ≤ 12 Gy ≤ 12 Gy <= 12Gy\leq 12 \mathrm{~Gy} )和蝸牛劑量( < 4 Gy < 4 Gy < 4Gy<4 \mathrm{~Gy} )(34-39)以及手術候選人出現積潤性視蓋( ( 40 , 41 ) ( 40 , 41 ) (40,41)(40,41) )(12)。在伽碼刀手術後,基線較差的聽力不僅影響保留可服務聽力的可能性,也與更陡峭的下降率有關(42)。
We only included patients with serviceable hearing preintervention, small ( < 2 cm < 2 cm < 2cm<2 \mathrm{~cm} in any linear dimension) tumors, and low tumor doses (12-13 Gy) in surgical and radiation patients, respectively. Maximum cochlear dose did, however, exceed 4 Gy in 26 patients, with a mean maximum dose of 9.0 ± 4.1 Gy 9.0 ± 4.1 Gy 9.0+-4.1Gy9.0 \pm 4.1 \mathrm{~Gy} (range, 3.5-16.0 Gy) and an overall mean dose of 4.0 ± 2.0 Gy 4.0 ± 2.0 Gy 4.0+-2.0Gy4.0 \pm 2.0 \mathrm{~Gy} (range, 2.1-10.5 Gy).
我們只包括在手術和放射治療患者中具有可維修聽力、小型(任何線性尺寸 < 2 cm < 2 cm < 2cm<2 \mathrm{~cm} )腫瘤和低劑量腫瘤(分別為 12-13 Gy)的患者。然而,在 26 名患者中,最大耳蝸劑量超過 4 Gy,平均最大劑量為 9.0 ± 4.1 Gy 9.0 ± 4.1 Gy 9.0+-4.1Gy9.0 \pm 4.1 \mathrm{~Gy} (範圍為 3.5-16.0 Gy),總平均劑量為 4.0 ± 2.0 Gy 4.0 ± 2.0 Gy 4.0+-2.0Gy4.0 \pm 2.0 \mathrm{~Gy} (範圍為 2.1-10.5 Gy)。
KM estimates of HP did not differ between radiation and surgery, and a subgroup-matched analysis (by age, tumor volume, and preintervention hearing) also did not reveal statistically significant differences between modalities. However, the odds of maintaining hearing were higher for patients with immediate postintervention serviceable hearing in the surgical cohort. The KM survival curves also intersected between surgical and radiation cohorts, suggesting unequal hazard rates over time, thereby violating the assumption of proportional hazards for these models.
放射線治療和手術之間的 KM 估計 HP 沒有差異,亞組配對分析(按年齡、腫瘤體積和干預前聽力)也沒有發現方式之間有統計學上的顯著差異。然而,對於術後立即有可用聽力的患者,手術組保聽率更高。KM 生存曲線在手術組和放射線組之間也有交叉,這表明隨時間的危險率不等,從而違反了這些模型的比例危險性假設。
As others have reported, we found that a greater proportion of surgical and radiation patients with better preintervention hearing maintained serviceable hearing at last follow-up ( 7 , 9 , 29 ) ( 7 , 9 , 29 ) (7,9,29)(7,9,29). We did not find a statistically significant difference in odds of HP by tumor volume, but there was a trend toward smaller volumes among surgical patients maintaining serviceable hearing. The presence of a fundal cap was also not informative of HP in surgical patients ( p > 0.99 p > 0.99 p > 0.99p>0.99 ) but was absent in relatively few patients ( n = 13 ) ( n = 13 ) (n=13)(\mathrm{n}=13).
手術及輻射治療患者在最後追蹤中保持可用聽力的比例較高。我們沒有發現腫瘤體積和聽力保留率之間有統計學上顯著的差異,但是接受手術治療而保留可用聽力的患者,腫瘤體積有較小的趨勢。基底膜蓋的存在也不能預測手術患者的聽力保留情況,但是相對很少患者沒有基底膜蓋。

Review of Contemporary Literature
當代文學評論

Surgery: Immediate and 5- and 10-Year Results
手術：即時、5 年及 10 年結果

Our 68.9 % ( n = 31 68.9 % ( n = 31 68.9%(n=3168.9 \%(n=31 of 45 ) immediate postoperative HP (class A/B) rates after MCF and 30.6 % ( n = 11 30.6 % ( n = 11 30.6%(n=1130.6 \%(n=11 of 36 ) ) )) with an RS approach fall within the range of previously reported values. Initial short-term HP rates (class A/B) following an MCF range between 47 % ( n = 181 47 % ( n = 181 47%(n=18147 \%(n=181 of 389 ) ( 43 ) ) ( 43 ) )(43))(43) and 82 % 82 % 82%82 \% ( n = 78 n = 78 n=78\mathrm{n}=78 of 95 ) (17) in a large series ( n ≥ 78 n ≥ 78 n >= 78\mathrm{n} \geq 78 with preoperative
我們 45 例)術後立即 HP(A/B 類型)的費率在 MCF 和 36 例以 RS 方法治療的結果,在先前報告的範圍內。初期短期 HP 費率(A/B 類型)在 MCF 後介於 389 和 95 之間(17),在一大型系列研究(合手術前
class A/B) (15) versus 28.5 % ( n = 54 28.5 % ( n = 54 28.5%(n=5428.5 \%(\mathrm{n}=54 of 189) (9) to 79 % 79 % 79%79 \% ( n = 23 n = 23 n=23\mathrm{n}=23 of 29) (44) with an RS approach. In a 2010 meta-analysis of 49 articles ( n = 998 n = 998 n=998\mathrm{n}=998 with initial class A/B, GR I/II), the overall HP rate was 52 % 52 % 52%52 \% ( 63 % 63 % 63%63 \% with MCF versus 47 % 47 % 47%47 \% with RS, p < 0.001 p < 0.001 p < 0.001p<0.001; follow-up, 0.5 − 7 yr 0.5 − 7 yr 0.5-7yr0.5-7 \mathrm{yr} ), similar to our overall surgical HP rate of 46.9 % ( n = 38 46.9 % ( n = 38 46.9%(n=3846.9 \%(n=38 of 81 ) ( 33 ) ) ( 33 ) )(33))(33).
A 類/B 類)（15）與 28.5 % ( n = 54 28.5 % ( n = 54 28.5%(n=5428.5 \%(\mathrm{n}=54 的 189）（9）相比， 79 % 79 % 79%79 \% （ n = 23 n = 23 n=23\mathrm{n}=23 的 29）（44），採用 RS 方法。在 2010 年對 49 篇文章（ n = 998 n = 998 n=998\mathrm{n}=998 初始 A 類/B 類,GR I/II）的元分析中，總體 HP 率為 52 % 52 % 52%52 \% （ 63 % 63 % 63%63 \% 與 MCF 相比， 47 % 47 % 47%47 \% 與 RS 相比， p < 0.001 p < 0.001 p < 0.001p<0.001 ; 随访, 0.5 − 7 yr 0.5 − 7 yr 0.5-7yr0.5-7 \mathrm{yr} ），與我們的總體手術 HP 率 46.9 % ( n = 38 46.9 % ( n = 38 46.9%(n=3846.9 \%(n=38 的 81 ) ( 33 ) ) ( 33 ) )(33))(33) 相似。
Over time, relatively few surgical patients ( n = 4 n = 4 n=4\mathrm{n}=4 of 42 [ 9.5 % ] 9.5 % ] 9.5%]9.5 \%] ) progressed to nonserviceable classes after initial HP, similar to Hilton et al. (15), who reported a late deterioration in 11 % ( n = 5 11 % ( n = 5 11%(n=511 \%(\mathrm{n}=5; mean follow-up, 4 yr ) ) )). Our estimated HP rate of 84.4 % 84.4 % 84.4%84.4 \% for class A/B and 99.2 % 99.2 % 99.2%99.2 \% for GR I/II after successful HP surgery at 5 years by KM analysis ( n = 13 n = 13 n=13\mathrm{n}=13 still at risk; Fig. 1) is similar to previous published rates. Three studies with at least 5 -year followup in MCF patients with initially successful HP found that class A / B A / B A//B\mathrm{A} / \mathrm{B} was maintained at 5 years in 70 % 70 % 70%70 \% ( n = 16 n = 16 n=16\mathrm{n}=16 of 23 ) (14), 75 % ( n = 12 75 % ( n = 12 75%(n=1275 \%(\mathrm{n}=12 of 16 ) (16), and 84 % 84 % 84%84 \% ( n = 27 ( n = 27 (n=27(\mathrm{n}=27 of 32 ) ( 17 ) ) ( 17 ) )(17))(17). In a review of RS patients with initially preserved class A / B A / B A//B\mathrm{A} / \mathrm{B} hearing, 87 % ( n = 47 87 % ( n = 47 87%(n=4787 \%(\mathrm{n}=47 of 54 ) ) )) maintained A / B A / B A//BA / B at a minimum of 6 years follow-up (9). Notably, when including patients with immediate postoperative anacusis into the KM analysis, 5-year HP rates drop to 48.9 to 55.3 % 55.3 % 55.3%55.3 \% (Fig. 3).
對於連續的時間,相對來說很少有手術病患(超過 42 [ 9.5 % ] 9.5 % ] 9.5%]9.5 \%] ]的 n = 4 n = 4 n=4\mathrm{n}=4 ) 在初步的 HP 手術後進展到無法服務的等級,這與 Hilton 等人(15)的報告類似,他們指出 11 % ( n = 5 11 % ( n = 5 11%(n=511 \%(\mathrm{n}=5 的嚴重化;平均追蹤時間為 4 年 ) ) )) 。我們經由 KM 分析估算的 HP 手術 5 年成功率為 84.4 % 84.4 % 84.4%84.4 \% 類 A/B 和 99.2 % 99.2 % 99.2%99.2 \% 類 GR I/II (圖 1 中 n = 13 n = 13 n=13\mathrm{n}=13 仍在風險群)與先前發表的率相似。有三項研究針對具有初步成功 HP 手術的 MCF 病患,至少追蹤 5 年,發現類 A / B A / B A//B\mathrm{A} / \mathrm{B} 維持率為 70 % 70 % 70%70 \% (23 例中的 n = 16 n = 16 n=16\mathrm{n}=16 )、 75 % ( n = 12 75 % ( n = 12 75%(n=1275 \%(\mathrm{n}=12 (16 例中的 84 % 84 % 84%84 \% ) 和 ( n = 27 ( n = 27 (n=27(\mathrm{n}=27 (32 例中的 ) ( 17 ) ) ( 17 ) )(17))(17) )。在一項 RS 病患研究中,初期保存的類 A / B A / B A//B\mathrm{A} / \mathrm{B} 聽力,在至少 6 年的追蹤期間, 87 % ( n = 47 87 % ( n = 47 87%(n=4787 \%(\mathrm{n}=47 例(共 ) ) )) )維持了該聽力(9)。值得注意的是,當將具有手術後即刻失聰的病患納入 KM 分析時,5 年 HP 成功率下降到 48.9 至 55.3 % 55.3 % 55.3%55.3 \% (圖 3)。
At 10 years after initially successful HP surgery, KM estimates of maintaining HP were 63.0 and 67.6 % 67.6 % 67.6%67.6 \% for class A/B and GR I/II, respectively (Fig. 1). Similarly, Hilton et al. found that 72 % 72 % 72%72 \% of patients with initially successful HP via MCF ( n = 51 n = 51 n=51\mathrm{n}=51 of 78) maintained serviceable hearing (class A/B) by KM analysis (15). Out of 23 patients with initial HP (class A/B) following an RS approach, Chee et al. (44) found that 65 % ( n = 15 ) 65 % ( n = 15 ) 65%(n=15)65 \%(n=15) of these patients maintained class A / B A / B A//B\mathrm{A} / \mathrm{B} at a mean follow-up of 9.4 years, resulting in 52 % 52 % 52%52 \% with 10 -year HP, accounting for those who also lost hearing during surgery. With both surgical approaches, Carlson et al. (12) estimated an overall 10-year HP rate of 43 % 43 % 43%43 \%, including patients with immediate postoperative hearing loss, which is comparable to our estimate of 32.7 to 35 % 35 % 35%35 \% when all surgical patients were included in the survival model (Fig. 3). Although the durability of successfully preserved hearing following either surgical approach is supported by our results and the literature, the overall long-term HP rate is therefore modest when also considering those who experienced immediate hearing loss with surgery.
在手術成功 10 年後,維持聽力保存的科氏估計值分別為類 A/B 組 63.0%和 GR I/II 組 67.6 % 67.6 % 67.6%67.6 \% %(圖 1)。同樣地,Hilton 等人發現經 MCF 手術最初成功的患者中, 72 % 72 % 72%72 \% %仍保持可用聽力(類 A/B)(15)。在 23 名最初使用 RS 手術獲得類 A/B 聽力的患者中,Chee 等人(44)發現 65 % ( n = 15 ) 65 % ( n = 15 ) 65%(n=15)65 \%(n=15) %的患者在平均 9.4 年的追蹤中保持類 A / B A / B A//B\mathrm{A} / \mathrm{B} 聽力,這包括了在手術過程中失去聽力的患者。Carlson 等人(12)估計採用兩種手術方式的總體 10 年聽力保存率為 43 % 43 % 43%43 \% %,包括出現術後立即聽力喪失的患者,這與我們在生存模型中包括所有手術患者時估計的 32.7%至 35 % 35 % 35%35 \% %相當(圖 3)。儘管無論採取哪種手術方式,成功保護聽力的持久性都得到我們的結果和相關文獻的支持,但總的長期聽力保存率在考慮了術中立即出現聽力喪失的患者後仍相對較低。
FIG. 3. Overall hearing preservation after intervention. Kaplan-Meier estimates for maintaining serviceable hearing over time defined by Gardner-Robertson class I/II (A), AAO-HNS classes A/B (B), and WRS of 50 % 50 % 50%50 \% or greater ( C C CC ). Patients who lost hearing immediately postoperatively were assigned a value of 0.01 years to be kept in the model. There was one Gamma Knife patient (pretreatment class B/II) who only had speech discrimination scores available after radiation. AAO-HNS indicates Academy of Otolaryngology-Head and Neck Surgery; WRS, word recognition score.
圖. 3. 干預後整體聽力保存。維持可用聽力的 Kaplan-Meier 估計隨時間的變化,定義為 Gardner-Robertson 第 I/II 級(A)、AAO-HNS 第 A/B 級(B)和 WRS ≥0%(C)。 立即術後丟失聽力的病人被分配 0.01 年的值來保持在模型中。有一個 Gamma 刀患者(術前 B/II 級)只有在接受輻射後才有語音辨識分數。AAO-HNS 是耳鼻咽喉科-頭頸外科學院;WRS 是語音識別分數。

Stereotactic Radiosurgery: Initial and 5- and 10-Year Results
立體定向放射外科手術：初始和 5 年及 10 年的結果

Summarizing radiation outcomes is complicated by evolving protocols with respect to decreasing marginal tumor and cochlear doses, as well as the use of different modalities (GK, linear accelerator radiotherapy, Proton, etc.) (45). To reflect current dosing protocols and facilitate comparison with our center, the following studies also utilized single-fraction Gamma Knife and mean/median tumor marginal doses of 13 Gy or less in patients with serviceable hearing (GR I, II) preintervention.
總結放射線結果受到不斷變化的協議的影響,這些協議旨在降低邊緣腫瘤和耳蝸劑量,以及不同療法的使用(伽瑪刀、直線加速器放射治療、質子等)(45)。為了反映當前的給藥方案並方便與我們中心的比較,以下研究還使用了單次分劑伽瑪刀和 13 Gy 或更低的腫瘤邊緣平均/中位劑量,用於有活動聽力(GR I、II)的患者術前。
Our short-term radiation-HP (class A, B/GR I, II) KM estimates of 84.4 / 86.6 % 84.4 / 86.6 % 84.4//86.6%84.4 / 86.6 \% at 1 year and 65.6/67.7% at 2 years are consistent with the existing literature’s high initial HP rates (Fig. 3). One review of 246 patients reported that 79 % 79 % 79%79 \% ( n = 218 n = 218 n=218\mathrm{n}=218 ) had GR I and II hearing with a median follow-up of 2 years (46). Via KM analysis, an estimated 83 % ( n = 33 83 % ( n = 33 83%(n=3383 \%(n=33 still at risk) will maintain class A / B A / B A//B\mathrm{A} / \mathrm{B} at 2 years (47).
我們短期輻射 HP(A 級,B/GR I,II)KM 估算在 1 年時為 84.4 / 86.6 % 84.4 / 86.6 % 84.4//86.6%84.4 / 86.6 \% ,在 2 年時為 65.6/67.7%,與現有文獻中高初期 HP 率一致(圖 3)。 一項 246 位患者的回顧發現, 79 % 79 % 79%79 \% ( n = 218 n = 218 n=218\mathrm{n}=218 )在中位數 2 年的隨訪中有 GR I 和 II 聽力(46)。 通過 KM 分析,估計 83 % ( n = 33 83 % ( n = 33 83%(n=3383 \%(n=33 仍處於風險中將在 2 年時保持 A / B A / B A//B\mathrm{A} / \mathrm{B} 類(47)。
At 5 years status after radiation, KM estimates of maintaining serviceable hearing (class A, B/GR-I, II) was 28.0/ 27.3 % 27.3 % 27.3%27.3 \% (Fig. 3). Our results were lower than 5-year estimates reported elsewhere: 48 % 48 % 48%48 \% with class A/B ( n = 21 n = 21 n=21\mathrm{n}=21 of 44; median follow-up, 9.3 yr ) ( 26 ) ) ( 26 ) )(26))(26) and 53 % ( n = 132 53 % ( n = 132 53%(n=13253 \%(\mathrm{n}=132; median follow-up, 5.1 yr ) (48) to 69 % ( n = 307 69 % ( n = 307 69%(n=30769 \%(\mathrm{n}=307; median follow-up, 5.5 yr ) (29) with GR I/II hearing. This discrepancy may be attributable to differences in our patient and/ or tumor characteristics. For example, 28 % 28 % 28%28 \% of our patients had GR I hearing and 37 % 37 % 37%37 \% were younger than 60 years at the time of radiation compared with 76 % 76 % 76%76 \% with GR I and 49 % 49 % 49%49 \% being younger than 60 years in the sample described by Johnson et al. (29). Regarding tumor size, 74 % 74 % 74%74 \% of our radiated patients had volumes less than 1.2 cm 3 1.2 cm 3 1.2cm^(3)1.2 \mathrm{~cm}^{3} compared with 82 % 82 % 82%82 \% reported by Johnson et al. Smaller, medial tumors with minimal cisternal extension undergoing radiation may experience less radiation injury to the cochlea and cochlear nerve fibers at the brainstem.
在接受放射治療 5 年後的狀態下,K-M 方法估算保持可服務聽力(A 類、B/GR-I、II 類)的比率為 28.0%。我們的結果低於其他報告的 5 年估算:A/B 類(44%,中位數隨訪 9.3 年)和 GR I/II 類(29%,中位數隨訪 5.5 年)。這種差異可能歸因於我們的患者和/或腫瘤特徵的差異。例如,我們有 60%的患者有 GR I 聽力,而 59%的患者小於 60 歲接受放射治療,與 Johnson 等人的樣本(47%GR I 和 69%小於 60 歲)相比。就腫瘤大小而言,我們的放射治療患者有 75%體積小於 3.5 cm3,而 Johnson 等人報告為 60%。接受放射治療的較小、內側的腫瘤,對於耳蝸和腦幹的耳蝸神經纖維造成的放射損傷可能較小。
Our 10-year KM estimates of serviceable hearing (class A, B/I, II) of 14.2 / 18 % 14.2 / 18 % 14.2//18%14.2 / 18 \% were also lower than previous estimates of 23 % ( n = 4 23 % ( n = 4 23%(n=423 \%(\mathrm{n}=4 still at risk; median follow-up, 9.3 yr ) (26) maintaining class A / B A / B A//B\mathrm{A} / \mathrm{B} and 34 % 34 % 34%34 \% ( n = 132 n = 132 n=132\mathrm{n}=132; median follow-up, 5.1 yr ) (48) to 52 % ( n = 307 52 % ( n = 307 52%(n=30752 \%(\mathrm{n}=307; median follow-up, 5.5 yr ) with GR I and II (29).
我們的 10 年 KM 估計可服務聽力(A 類、B/I 類、II 類)為 14.2 / 18 % 14.2 / 18 % 14.2//18%14.2 / 18 \% 低於先前對 23 % ( n = 4 23 % ( n = 4 23%(n=423 \%(\mathrm{n}=4 仍有風險的估計;中位數追蹤時間為 9.3 年)(26)維持 A / B A / B A//B\mathrm{A} / \mathrm{B} 和 34 % 34 % 34%34 \% 類( n = 132 n = 132 n=132\mathrm{n}=132 ;中位數追蹤時間為 5.1 年)(48)至 52 % ( n = 307 52 % ( n = 307 52%(n=30752 \%(\mathrm{n}=307 ;中位數追蹤時間為 5.5 年)I 及 II 級 GR(29)。
Moreover, Johnson et al. (29) found that maintaining serviceable (GR I, II) hearing was 92 % 92 % 92%92 \% at 10 years ( n = 10 n = 10 n=10\mathrm{n}=10 still at risk) in younger ( < 45 yr < 45 yr < 45yr<45 \mathrm{yr} ) patients with GR I hearing and tumor size less than 1.2 cm 3 1.2 cm 3 1.2cm^(3)1.2 \mathrm{~cm}^{3} at the time of GK-SRS. Only one of our patients met these specifications preintervention but maintained GR II hearing at last follow-up, 5.5 years later.
此外,約翰遜等人(29)發現,在年輕( < 45 yr < 45 yr < 45yr<45 \mathrm{yr} )患者中,GR I 聽力和腫瘤大小小於 1.2 cm 3 1.2 cm 3 1.2cm^(3)1.2 \mathrm{~cm}^{3} 的情況下,在接受 GK-SRS 治療 10 年後,仍保持可接受(GR I,II)的聽力( n = 10 n = 10 n=10\mathrm{n}=10 仍有風險)。我們的患者中,僅有一人符合這些條件,但在最後一次追蹤中,在 5.5 年後仍保持 GR II 聽力。
Because of the association between postradiation long-term serviceable hearing and better preintervention hearing, some centers advocate for early, upfront radiation (48-50) even in the absence of hearing loss or tumor growth. However, there are several considerations to address with this approach. First, most patients are expected to live several decades beyond their initial diagnosis, as the majority are diagnosed in the fifth and sixth decades of life (51). The point at which SRS-induced hearing deterioration plateaus, if ever, is still unknown, with limited available data demonstrating
由於後放療長期可用聽力與更好的預先干預聽力之間的關聯,一些中心支持早期、提前放療(48-50),即便沒有聽力損失或腫瘤增長。然而,這種方法還有幾個需要考慮的因素。首先,大多數患者在最初診斷後預計會活幾十年,因為大多數人在生命的第五和第六十年被診斷出來(51)。如果真的會出現,SRS 誘導的聽力惡化何時會達到平台期仍然未知,可用數據有限。
further decline beyond 10 years (26). Furthermore, the risk for malignant transformation in radiated tumors is still debated in the setting of limited follow-up of at least 10 years ( 52 , 53 ) ( 52 , 53 ) (52,53)(52,53). Follow-up greater than 15 to 20 years is considered an adequate duration to reliability estimate incidence of malignant degeneration (54). Third, a considerable proportion of tumors may never grow or not grow for many years. In a large observational cohort ( n = 2 , 312 n = 2 , 312 n=2,312\mathrm{n}=2,312 ), only 21 % 21 % 21%21 \% of intracanalicular and 37 % 37 % 37%37 \% of cisternal tumors grew (defined as > 2 mm > 2 mm > 2mm>2 \mathrm{~mm} ) in 5 years after diagnosis (mean follow-up, 7.3 yr) (55). A prospective study is underway to compare upfront GK-SRS and observation (56), which may reveal a relative advantage of one management approach for patients presenting with normal hearing.
10 年以上流行病學數據下降(26)。此外,照射腫瘤惡性轉化風險在至少 10 年的有限追蹤期內仍有爭議 ( 52 , 53 ) ( 52 , 53 ) (52,53)(52,53) 。15 至 20 年以上的跟蹤被視為足夠的時間可靠估計惡性變性發生率(54)。第三,相當比例的腫瘤可能永遠不會增長或多年不增長。在一個大規模的觀察性隊列中( n = 2 , 312 n = 2 , 312 n=2,312\mathrm{n}=2,312 ),只有 21 % 21 % 21%21 \% 的管壁內腫瘤和 37 % 37 % 37%37 \% 的蛛網膜下腔腫瘤在診斷後 5 年內有增長(定義為 > 2 mm > 2 mm > 2mm>2 \mathrm{~mm} )(55)。一項前瞻性研究正在進行,以比較預先放射科刀和觀察(56),這可能會揭示一種管理方法相對於有正常聽力的患者的優勢。

Observation: Short-Term and 5- and 10-Year Outcomes
觀察:短期和 5 年以及 10 年的結果

Although not evaluated by this study, others have followed hearing in observed patients. Similar to radiated and surgical VS cohorts, baseline hearing is one of the strongest prognosticators for progression to nonserviceable hearing, in addition to tumor growth during observation (12). At 2 years, 67 % 67 % 67%67 \% of 409 patients maintained serviceable hearing (GR I, II) (57). Notably, 60 % ( n = 29 ) 60 % ( n = 29 ) 60%(n=29)60 \%(\mathrm{n}=29) of patients with greater than 2 mm of growth during observation remained GR I. At 5 years, Stangerup et al. found that 74% of 178 maintained class A hearing (58). Ferri et al. (59) found that 73 % 73 % 73%73 \% ( n = 41 n = 41 n=41\mathrm{n}=41 of 56 ; mean follow-up, 4.8 yr ) maintained class A / B A / B A//BA / B, and a minority ( 36 % 36 % 36%36 \% ) of tumors grew, with no growth occurring beyond 6 years of observation. After 10 years, Stangerup et al. (58) reported that 46 % 46 % 46%46 \% of 178 observed patients maintained class A, and 52 % 52 % 52%52 \% of 491 patients had WRS of 70 % 70 % 70%70 \% or greater. Specifically in intracanalicular tumors, Kirchmann et al. (60) reported 34 % 34 % 34%34 \% HP (class A/B, n = 73 n = 73 n=73\mathrm{n}=73 ). KM 10-year HP estimates (class A / B ) A / B ) A//B)A / B) range from 44 % ( n = 466 ; n = 10 44 % ( n = 466 ; n = 10 44%(n=466;n=1044 \%(n=466 ; n=10 still at risk) by Hunter et al. (61) to 49 % ( n = 213 ; n = 8 49 % ( n = 213 ; n = 8 49%(n=213;n=849 \%(n=213 ; n=8 still at risk) (62).
儘管未由本研究評估,但其他人已觀察患者的聽力情況。與輻射和手術 VS 隊列類似,基線聽力是導致聽力喪失惡化的最強預後指標,此外還包括觀察期內腫瘤的增長(12)。在 2 年後,409 名患者中 67 % 67 % 67%67 \% 保持可服務聽力(GR I, II)(57)。值得注意的是,在觀察期內增長超過 2 mm 的患者中, 60 % ( n = 29 ) 60 % ( n = 29 ) 60%(n=29)60 \%(\mathrm{n}=29) 仍維持 GR I。在 5 年後,Stangerup et al.發現 178 名患者中 74%維持 A 級聽力(58)。Ferri et al.(59)發現 73 % 73 % 73%73 \% ( n = 41 n = 41 n=41\mathrm{n}=41 名,平均隨訪 4.8 年)維持 A / B A / B A//BA / B 級,只有少數( 36 % 36 % 36%36 \% )腫瘤出現增長,但在觀察 6 年後無進一步增長。在 10 年後,Stangerup et al.(58)報告 178 名觀察患者中 46 % 46 % 46%46 \% 維持 A 級,491 名患者中 52 % 52 % 52%52 \% 的 WRS 達 70 % 70 % 70%70 \% 或以上。具體對於內耳道內的腫瘤,Kirchmann et al.(60)報告 34 % 34 % 34%34 \% HP(A/B 級, n = 73 n = 73 n=73\mathrm{n}=73 )。Hunter et al.(61)的 KM 10 年 HP 估計(A/B 級)為 44 % ( n = 466 ; n = 10 44 % ( n = 466 ; n = 10 44%(n=466;n=1044 \%(n=466 ; n=10 至 49 % ( n = 213 ; n = 8 49 % ( n = 213 ; n = 8 49%(n=213;n=849 \%(n=213 ; n=8 仍在風險組(62)。

Direct Comparisons Between Observation, Radiation, and Surgery
直接比較觀察、輻射和手術

In the most comprehensive HP literature review to date, Carlson et al. (12) found similar long-term outcomes with small-to-medium tumors among the three management paradigms. At 2 years, there was a high probability ( > 75 − 100 % > 75 − 100 % > 75-100%>75-100 \% ) of HP in observed or radiated patients with initially good (class A/GR I) or serviceable hearing. After resection, HP probability was moderately high ( > 50 − 75 % > 50 − 75 % > 50-75%>50-75 \% ) if patients had good hearing preoperatively versus moderately low ( > 25 − 50 % > 25 − 50 % > 25-50%>25-50 \% ) if only serviceable (12). At 5 years, HP probability was moderately high in observed and radiated patients with good or serviceable initial hearing and in surgical patients with good preoperative hearing (versus moderately low if serviceable preoperative hearing) (12). At 10 years, HP probability was moderately low for observed patients with initially serviceable hearing and for both radiated and surgical patients with serviceable or good baseline hearing (12). Insufficient data were available to provide 10-year projections for observed patients with good hearing at presentation.
在迄今為止最全面的 HP 文獻回顧中，Carlson 等人(12)發現三種管理模式中小腫瘤的長期結果相似。在 2 年後，觀察或接受放射治療的患者如果初始有良好(A/GR I 級)或可接受的聽力，則很可能出現 HP ( > 75 − 100 % > 75 − 100 % > 75-100%>75-100 \% )。在切除術後，如果患者術前聽力良好，則 HP 的可能性適度較高( > 50 − 75 % > 50 − 75 % > 50-75%>50-75 \% )；而如果只有可接受的聽力，則 HP 的可能性適度較低( > 25 − 50 % > 25 − 50 % > 25-50%>25-50 \% )(12)。在 5 年後，觀察和接受放射治療的患者如果初始聽力良好或可接受，以及接受手術的患者如果術前聽力良好，HP 的可能性適度較高(而如果術前只有可接受的聽力，則適度較低)(12)。在 10 年後，對於初始聽力可接受的觀察組患者以及聽力可接受或良好的放射治療和手術組患者，HP 的可能性適度較低(12)。缺乏足夠數據來預測初始聽力良好的觀察組患者在 10 年後的情況。
  1. Address correspondence and reprint requests to Kristen L. Yancey, M.D., Department of Otolaryngology-Head and Neck Surgery, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX 75390. E-mail: kristen.yancey@utsouthwestern.edu
    向克里斯汀·L·楊西, M.D.提出通訊和重印請求, 地址: 德克薩斯州西南醫學中心奧拉林腦頭頸外科部, 5323 Harry Hines 大道, 達拉斯, TX 75390. 電子郵件: kristen.yancey@utsouthwestern.edu
    No funding was provided for this work. Two authors (B.I., W.K.) are consultants for Alcon and Medtronic.
    没有提供资金支持此工作。两位作者(B.I.，W.K.)是 Alcon 和 Medtronic 的顾问。
    Institutional review board approval: STU 112016-040, UT Southwestern Medical Center.
    機構審查委員會批准:STU 112016-040, UT Southwestern 醫療中心。
    DOI: 10.1097/MAO.0000000000003618
    數位識別碼：10.1097/MAO.0000000000003618