甲狀腺手術之術中神經監測 -轉譯豬隻模式研究與臨床新應用

Intraoperative neuromonitoring during thyroid surgery- translational porcine model research and clinical new applications

吳哲維(Che-Wei Wu)

指導教授 ： 江豐裕

高雄醫學大學/醫學院/臨床醫學研究所/博士(2014年)

https://doi.org/10.6832/KMU.2014.00049

摘要

目的:
喉返神經(RLN)痲痺是甲狀腺手術後最常見且嚴重的併發症。許多文獻已證實，術中常規目視辨識RLN可以降低神經的受傷的機率。手術中的輔助工具，例如”術中神經監測(IONM)”，近年來已漸漸被接受應用於確認目視資訊，預測術後聲帶功能，並有助於釐清術中RLN受傷的機轉。然而目前由於許多使用上的安全疑慮，操作上的潛在問題，以及缺乏標準化的關係，不但限制了這項新技術的使用，而且訊息的錯誤判讀反而會增加神經受傷的風險。本研究目的是建立一套前瞻性實驗動物(豬隻)模式來探討執行IONM時的安全性以及刺激迷走神經(VN)及RLN時的理想電流強度。我們也利用這套動物模式來探討術中RLN急性受傷時肌電圖(EMG)訊號變化模式並藉此歸納出如何應用IONM來預防神經受傷的可信賴策略。同時我們也將動物研究所得的資訊應用於發展臨床使用IONM時的新策略。
方法:
動物轉譯研究部分我們採用Duroc–Landrace迷你公豬進行實驗。IONM使用NIM系統，所誘發的聲門肌肉群EMG訊號經由特製氣管內管上的表面電極接收。第一階段安全性實驗共採用八隻迷你豬(n=16條VNs及RLNs)。首先測試並紀錄每條神經在不同電流強度刺激下的EMG反應基準值，隨後我們再給予每條神經連續10分鐘的脈動性較高電流強度刺激(3mA, 4Hz)，同時間紀錄EMG訊號及心肺功能所發生的變化並加以分析。第二階段神經受傷實驗採用十五隻迷你豬(n=30條RLNs)進行研究。神經刺激採用VN自動周期性刺激，同時連續性紀錄並對應神經遭受不同受傷模式時即時的EMG訊號變化。
臨床應用研究部分我們納入了440名接受甲狀腺手術的病患，探討執行IONM時採用 ”Nerve Mapping” 的方法來完成 1. 早期定位及辨識RLN (220名病患，333條神經)；以及 2. 在不打開頸動脈鞘下完成VN刺激(220名病患，376條神經)之可行性與可靠性。
結果:
動物研究第一階段實驗我們發現，電流刺激強度與被誘發EMG振幅間呈現正向的「劑量反應曲線」，而且只要使用約1mA的電流刺激VN或RLN即可誘發出最大的EMG振幅反應。同時在經過10分鐘3mA的連續刺激後，我們在EMG訊號上並沒有發現明顯的改變，此外持續的心肺功能監測也沒有發生不良的變化。動物研究第二階段實驗我們發現，當RLN術中受不同程度張力牽拉時，EMG均呈現出漸進式的部分減弱情形 (合併振幅減弱及潛時延長，n=8)，而且當牽拉解除時，EMG訊號會有漸進式恢復表現；然而當RLN術中受到電熱傷害(n=4)、夾壓(n=1)， 以及切斷(n=1)時，EMG訊號則是表現出立即性的大幅減弱或消失，且無法觀察到有漸進式恢復的表現。在進一步探討不同時間神經牽拉後EMG恢復程度的實驗中我們發現(n=16)，若能在神經受牽拉時及早發現EMG訊號減弱，並在訊號消失前解除，則訊號可以達到幾乎完全恢復狀態(98± 3%, n=6)。不過當神經受到牽拉的時間越長或同一條神經再次進行反覆牽拉，則訊號恢復的情形會愈差；若在發生EMG訊號出現消失當時，以及在訊號消失一分鐘後才解除牽拉，則20分鐘後的訊號恢復表現分別只有36± 4% (n=4) 及 15± 2% (n=6)。
臨床研究第一階段我們證實所有RLN均可以成功的經由應用nerve mapping的方法來早期定位與辨識，其中包括87條(26%)困難辨識的神經；而所需的電流強度在其中315條神經(95%)為2mA，另18條神經(5%)為3mA。而且nerve mapping所測到的訊號強度(振幅為932± 436μV)與直接刺激VN (振幅=811± 389μV)或RLN (振幅=1132± 472μV) 時表現相當，且均呈現出清楚且明確的喉部EMG波形。病患手術後出現神經麻痺機率僅0.6%，並無永久性麻痺發生。臨床研究第二階段我們證實所有研究個案術中均可以經由應用nerve mapping的方法，在不打開頸動脈鞘情況下完成VN刺激，同時並沒有造成任何不良反應。此外所有的EMG反應均可以在手術開始後的30分鐘內成功測得，訊號也如同傳統打開頸動脈鞘直接刺激VN時一樣，呈現出清楚且明確的喉部EMG反應。
結論：
經由本篇動物模式研究顯示，執行IONM時使用電流刺激RLN及VN是安全的；1mA電流就足以誘發最大EMG反應，可以當作常規使用的強度以降低發生神經受傷的風險及假訊號的情形。較高強度的電流則可選擇用於間接探查或定位神經的路徑。我們也發現，當術中RLN受牽拉時伴隨著漸進式部分EMG訊號減弱之情形，同時及早在訊號消失前解除神經牽拉可以有極佳的恢復機會。這意味著IONM可以應用於及早發現不正常的EMG變化，同時提醒術者立即更正甲狀腺手術中不當的牽拉以避免RLN的不可逆損傷發生。
經由我們的臨床應用研究也證實，應用間接nerve mapping的方式可以有效的在手術中早期定位及辨識出RLN，大大降低術中神經遭受意外受傷或過度牽扯的風險。並且也可以在不打開頸動脈鞘情況下，完成簡易及安全的VN刺激。這些研究成果將有助於將IONM改良為預防甲狀腺手術中RLN受傷的一項更簡易、安全及可靠工具。

關鍵字

術中神經監測 (IONM) ； 甲狀腺手術 ； 喉返神經(RLN) ； 迷走神經(VN) ； 肌電圖(EMG) ； 聲帶麻痺 ； 前瞻性動物模式 ； 轉譯研究 ； 臨床應用

並列摘要

Objective:
Recurrent laryngeal nerve (RLN) palsy is the most common and serious complication after thyroid operation. Routine visual identification of the RLN has been reported to be associated with lower rates of nerve injury. Surgical adjuncts such as intraoperative neuromonitoring (IONM) are being applied to confirm visual information, predict postoperative vocal cord function, and help in elucidating the mechanism of intraoperative RLN injury. However, several safety concerns, potential pitfalls, and the lack of standardization not only limit the value of this novel technology, but also result in misleading information and conversely, increase the risk of RLN injury. The purpose of this study was to develop a prospective experimental animal (porcine) model to investigate the safety and optimal intensity of electrical vagus nerve (VN) and RLN stimulation during IONM. We also used this model to investigate the electromyographic (EMG) signal pattern during an acute RLN injury and establish reliable strategies to prevent the injury using IONM. The information gaining from the animal model was also applied to develop new clinical strategies during IONM.
Methods:
Duroc–Landrace male piglets were used for animal translational researches. The piglets underwent IONM via nerve integrity monitor (NIM) system, and the electrically evoked electromyography (EMG) was recorded from the vocalis muscles via endotracheal surface electrodes. Eight piglets (n=16 VNs and RLNs) were enrolled in the period-1 safety study. The baseline EMG parameters were measured and continuous pulsatile stimulations (3 mA, 4Hz) were performed on each VN and RLN for 10 minutes. Changes of EMG waveform and cardiopulmonary status were analyzed. Fifteen piglets (30 RLNs) were enrolled in the period-2 nerve injury study. The piglets underwent IONM via automated periodic VN stimulation and had their EMG tracings recorded and correlated with various models of nerve injury.
In the clinical application researches, we enrolled 440 patients who underwent thyroid operations with application of IONM and investigated the feasibility and reliability of using “nerve mapping” technique to early localize and identify the RLN (period-1, 220 patients, 333 RLNs at risk), and to perform VN stimulation without dissection the carotid sheath (period-2, 220 patients, 376 RLNs at risk).
Results
In period-1 animal study, we found that a dose–response curve existed with increasing EMG amplitude as stimulating current was increased, with maximum amplitude elicited on VN and RLN stimulation at around 1 mA. No obvious EMG changes and untoward cardiopulmonary effects were observed after the continuous stimulation with 3mA for 10 minutes. In period-2 animal study, we observed a progressive, partial EMG loss under RLN tractions with different tension (n = 8), and the EMG gradually gained partial recovery after the traction was relieved. Among the nerves injured with electrothermal (n = 4), clamping (n = 1), and transection (n = 1) models, the EMG showed immediate partial or complete loss, and no gradual EMG recovery was observed. Another 16 RLNs were used to investigate the potential of EMG recovery after different extents of RLN traction. We noted the EMG showed nearly full recovery if the traction stress was relieved before the loss of signal (LOS), but the recovery was worse if prolonged or repeated traction was applied. The mean restored amplitudes after the traction was relieved before, during, and after the LOS were 98± 3% (n = 6), 36± 4% (n = 4), and 15± 2% (n = 6), respectively.
In period-1 clinical study, we confirmed that all RLNs, including 87 (26%) nerves that were regarded as difficult to identify, were successfully early localized and identified with nerve mapping. The stimulation level for RLN localization was 2mA in 315 nerves (95%) and 3mA in the other 18 nerves (5%). The signal obtained from RLN mapping (amplitude= 932± 436μV) showed a clear and reliable laryngeal EMG response that was similar to that from direct vagus (amplitude= 811± 389μV) or RLN stimulation (amplitude= 1132± 472μV). The palsy rate was only 0.6% and no permanent palsy occurred. In period-2 clinical study, we confirmed that using nerve mapping technique to perform VN stimulation without nerve exposure was feasible in all cases and did not result in any morbidity. All EMG signals were successfully obtained within 30 minutes of the start of the operation and all showed a clear and reliable laryngeal EMG response that was similar to that from the conventional method in which direct VN exposure for stimulation is applied.
Conclusion
Our animal model showed that electrical RLN and VN stimulation are very safe during IONM. 1mA stimulation is enough to evoke the maximal EMG response and can be selected for routinely use to minimize the potential risk of nerve damage and false results. Higher current can be selected when indirectly mapping or localizing the path of nerve. We also found that traction to RLN showed graded, partial EMG changes; early release of the traction before the EMG has degraded to LOS offers a good chance of EMG recovery. Therefore IONM can be used as a tool for the early detection of adverse EMG changes that may alert surgeons to correct certain maneuvers immediately to prevent irreversible nerve injury during the thyroid operation.
Our clinical application study confirmed that the nerve mapping technique is an effective tool to early localize and identify the RLN that may greatly decrease the risk of nerve inadvertent injury or excessive traction, and to perform simple and safe VN stimulation without dissection the carotid sheath. These results will greatly help the IONM to become a more simple, safe, and reliable tool to prevent RLN injury during thyroid surgery.

並列關鍵字

Intraoperative neuromonitoring (IONM) ； Thyroid surgery ； Recurrent laryngeal nerve (RLN) ； Vagus nerve (VN) ； Electromyography (EMG) ； Vocal cord paralysis ； Prospective animal model ； Translation research ； Clinical application

參考文獻

[1] Kern KA: Medicolegal analysis of errors in diagnosis and treatment of surgical endocrine disease. Surgery 1993, 114:1167-73; discussion 73-4.

[2] Chiang FY, Wang LF, Huang YF, Lee KW, Kuo WR: Recurrent laryngeal nerve palsy after thyroidectomy with routine identification of the recurrent laryngeal nerve. Surgery 2005, 137:342-7.

[3] Jatzko GR, Lisborg PH, Muller MG, Wette VM: Recurrent nerve palsy after thyroid operations--principal nerve identification and a literature review. Surgery 1994, 115:139-44.

[4] Moley JF, Lairmore TC, Doherty GM, Brunt LM, DeBenedetti MK: Preservation of the recurrent laryngeal nerves in thyroid and parathyroid reoperations. Surgery 1999, 126:673-7; discussion 7-9.

[5] Steurer M, Passler C, Denk DM, Schneider B, Niederle B, Bigenzahn W: Advantages of recurrent laryngeal nerve identification in thyroidectomy and parathyroidectomy and the importance of preoperative and postoperative laryngoscopic examination in more than 1000 nerves at risk. Laryngoscope 2002, 112:124-33.