胸腔鏡手術對原發型自發性氣胸病患之治療 —追加Minocycline肋膜沾粘之效果及作用機轉

Video-Assisted Thoracoscopic Surgery for Primary Spontaneous Pneumothorax: Effects and Mechanisms of Additional Minocycline Pleurodesis

陳晉興(Jin-Shing Chen)

指導教授 ： 呂勝春 ； 李元麒

國立臺灣大學/醫學院/臨床醫學研究所/博士(2004年)

https://doi.org/10.6342/NTU.2004.02189

摘要

原發型自發性氣胸通常發生於瘦、高之年輕男性。90%以上的病患可在肺臟發現一個或數個氣泡。第一次發作後若沒有手術或藥物治療，則再次發作的機率為23-50%，第二次發作後，更有高達80%病患會再有第三次發作。因此學者建議第二次氣胸復發或某些特殊情況的第一次氣胸發作病患應考慮手術治療。手術治療的原則，除了切除氣泡外，還必須加上肋膜沾粘的步驟，來減少日後氣胸復發的機率。至於最合適的手術治療方法目前並沒有共識。
原發型自發性氣胸手術治療的研究大致可分為以下重點：一、減少手術的創傷及並促進傷口的美觀。二、減少手術後併發症及氣胸復發的機率。三、肋膜沾粘的效果及作用機轉。以下是我們關於這三方面的研究成果摘要：

一、減少手術的創傷及並促進傷口的美觀
爲減少手術的創傷及加速病患的恢復，目前原發型自發性氣胸手術治療的主流為胸腔鏡手術。但傳統胸腔鏡手術的傷口並非完美，術後仍有約30%的病患會有長期傷口疼痛的現象。爲了進一步減少手術的創傷，我們改良迷你胸腔鏡的手術方法，並應用在原發型自發性氣胸病患的治療。
為了測試改良的迷你胸腔鏡手術是否為治療原發型自發性氣胸之可行方法，於2001年4月到2002年4月間，台大醫院胸腔外科共有63位原發型自發性氣胸病患因同側氣胸復發，對側氣胸復發，或胸管持續漏氣而接受胸腔鏡手術治療。其中28例病患選擇改良式的迷你胸腔鏡手術，另有35例病患選擇傳統之胸腔鏡手術。兩組病患不論在年齡、體重、性別、吸菸比例、手術適應症及手術發現皆無顯著差異。在手術結果方面，兩組病患的手術時間，術後疼痛程度，術後使用止痛藥劑量，術後胸管引流時間，以及住院時間皆類似。病患平均追蹤8個月(介於2到14個月)。根據問卷調查顯示使用改良式迷你胸腔鏡手術的病患，其長期傷口疼痛的比例較少，傷口滿意度較佳，且僅有一例病患有氣胸復發的情況。此一研究結果顯示改良式迷你胸腔鏡手術在技術上是可行的，同時可促進傷口美觀及減少長期疼痛的機率。
為了進一步評估改良式迷你胸腔鏡手術的安全性及長期效果，我們收集了2001年4月到2003年10月間，因原發型自發性氣胸在台大醫院接受改良式迷你胸腔鏡手術的病患。在這段期間共有110位病患接受改良式迷你胸腔鏡手術。病患平均年齡23.5歲，其中102位病患接受單側手術，有8位病患同時接受雙側手術。手術的方式為利用迷你胸腔鏡進行氣泡切除術及壁層肋膜摩擦術。平均手術的時間為76 分鐘，沒有任何病患發生死亡或重大併發症。術後6例病患(5.1%)有長期胸管漏氣情況，胸管拔除後的肋膜塌陷(pleural detachment)則發生於2例病患(2%)。術後平均住院天數為4.2±2.4天。病患術後平均的追蹤時間為16個月(介於2到30個月)，有術後長期傷口疼痛之病患只有16.3%，疼痛都是偶而發生，不需規則服用止痛藥。大部分的病患(73.0%)對傷口皆感到滿意或非常滿意。同側氣胸復發發生於8例改良式迷你胸腔鏡手術(7.1%)，其中有5例發生於手術後6個月內。為了找尋和改良式迷你胸腔鏡手術術後氣胸復發的相關因子，我們利用十個變數來進行Cox氏回歸分析(Cox’ regression model)。這十個變數為：年齡，性別，吸菸狀況，手術適應症，氣泡數目，氣泡位置，單側或雙側手術，是否追加minocycline肋膜沾粘術，使用止痛藥meperidine之劑量，及是否出現併發症等。單變數分析(univariate analysis)中發現當氣泡數目大於或等於三個，同時接受雙側改良式迷你胸腔鏡手術，及術後沒有追加minocycline肋膜沾粘術的病患其氣胸復發之機率明顯偏高(p < 0.05)。Cox氏回歸的多變數分析（multivariate analysis）則顯示同時接受雙側改良式迷你胸腔鏡手術及術後沒有追加minocycline肋膜沾粘術是顯著且獨立的氣胸復發危險因子。當我們利用Kaplan-Meier方法來預估改良式迷你胸腔鏡手術術後的復發機率時，可以發現單側和雙側手術及有沒有追加minocycline肋膜沾粘術的確有明顯的差別。
在此研究中我們發現改良式迷你胸腔鏡手術對於單側的原發型自發性氣胸病患，是一項安全且有效的治療，且術後追加minocycline肋膜沾粘術可以減少氣胸復發的機率。但對於必須接受雙側氣胸手術的病患而言，改良式迷你胸腔鏡手術治療的效果則需再進一步研究。

二、減少手術後併發症及氣胸復發的機率
爲減少氣胸手術後復發的機率，胸腔鏡氣泡切除後通常會追加肋膜沾粘術。肋膜沾粘的方式，可分為化學式沾粘（chemical pleurodesis）及機械式沾粘（mechanical pleurodesis）兩大類。化學式沾粘是指在病患肋膜腔內注射刺激性之藥物，引發壁層肋膜及臟層肋膜之發炎，因而引起兩層肋膜間的沾粘。機械式沾粘則包含肋膜摩擦術（pleural abrasion）及肋膜剝離術（pleurectomy）等。目前胸腔鏡氣胸手術較常使用的作法是氣泡切除後追加壁層肋膜摩擦術，但術後約有5%至10%病患會有長期漏氣情況，同時仍有5%至9%的病患會發生同側氣胸復發。為減少長期漏氣及術後氣胸復發的機率，我們嘗試在胸腔鏡氣泡切除及壁層肋膜摩擦術後追加minocycline肋膜沾粘。
從1992年到2001年，台大醫院共有364位原發型自發性氣胸病患接受傳統胸腔鏡手術。其中，1992年到1994年，共有53位病患在手術後沒有追加任何治療(對照組)。在1994年後，313例病患為減少術後之氣胸復發機率，在術後以minocycline hydrochloride 7mg/kg，由胸管注入肋膜腔中（minocycline組）。在這個回溯性歷史性對照研究中，我們發現兩組病患在身體狀況、手術發現、以及手術時間上皆無明顯差異。注射minocycline後最大的副作用是引發胸痛，但實驗組必須注射止痛藥meperidine的總劑量和對照組並無顯著差異。Minocycline組在手術後其長期漏氣的病患較少（7% vs 18%, p=0.025），病患胸管置放天數較短，同時住院的日期也縮短。術後追蹤的過程中，共有14位病患發生同側氣胸復發。其中minocycline組的復發機率明顯較低(2.9%比9.8%，p＝0.033)。這個研究雖然顯示肋膜腔內注射minocycline可能可以改善胸腔鏡手術治療原發型自發性氣胸的效果，然而其最大的缺點在於此為一歷史性對照之回溯性研究，因此無法排除minocycline組手術結果較好的原因是因為胸腔鏡手術技術進步的結果，而非minocycline真的有效。
為了真正了解胸腔鏡手術後追加minocycline肋膜沾粘的效果，我們在2001年6月到2004年2月進行一項前瞻性的隨機臨床試驗。202位病患因原發型自發性氣胸接受胸腔鏡手術治療。其中60人接受傳統胸腔鏡手術，142人接受改良式迷你胸腔鏡手術。手術方法皆為氣泡切除術及壁層肋膜摩擦術。手術後病人隨機分為兩組。Minocycline組病患〈103位〉在術後皆追加minocycline肋膜沾粘術〈7mg/kg，每位病患約施打300mg或400mg〉。對照組病患〈99位〉則不追加任何治療。手術時間平均為79分鐘。兩組病患不論在年齡、性別、體重、抽菸狀態、手術適應症、手術方式、手術發現、使用自動吻合釘(Endoscopic stapler cartridge)的數目及手術時間皆沒有明顯差別。病患在接受minocycline肋膜沾粘術後並沒有明顯的副作用或併發症。Minocycline治療後病患最大的抱怨是有短暫胸痛的現象，約有70%之病患必須肌肉注射止痛藥meperidine，且meperidine施打的總劑量在minocycline組明顯較高。至於術後胸管置放天數、術後住院日數、以及發生併發症之比例兩組病患則無明顯差別。但使用minocycline似乎有減少長期漏氣的趨勢。在術後長期漏氣的8位病患中，minocycline組只有2位，對照組則有6位(2% v.s. 6%, p = 0.1)，且其中四位需要使用minocycline來減少漏氣的時間。手術病患沒有任何死亡發生。病患術後平均追蹤15個月〈一個月至三十二個月〉，兩組病患恢復正常工作及上課的時間類似，同時長期傷口疼痛的比例也類似。術後肺功能測驗顯示兩組病患的用力肺活量及用力呼氣一秒量並沒有明顯差異。
追蹤時間內共有10位病患發生同側氣胸復發。其中minocycline組只有2位 (2%)，對照組則有8位 (8%)，但仍未達統計學上之顯著差別 (p = 0.055)。然而若將病患分為低危險群〈氣泡數目1或2個〉及高危險群〈氣泡數目3個以上或手術時找不到氣泡〉，則顯示追加minocycline肋膜沾粘術明顯可減少高危險病患術後氣胸復發之機率 (p = 0.03)。此現象利用Kaplan-Meier之方法分析，其結果仍相同。

三、追加Minocycline肋膜沾粘之效果及作用機轉
藉由前兩項研究，我們初步證實術後經由胸管追加minocycline肋膜沾粘確實可以減少氣胸復發的機率。然而其減少氣胸復發的機制則並不完全清楚。因此我們利用紐西蘭白兔作為動物模式，來研究機械式肋膜沾粘術和化學式肋膜沾粘術的肋膜沾粘的機制和效果。我們使用36隻2.5至4.0公斤的紐西蘭白兔進行實驗。兔子隨機分為四組接受胸腔鏡手術：第一組接受壁層肋膜磨擦術(pleural abrasion)，第二組接受minocycline肋膜沾粘術，第三組接受壁層肋膜磨擦術及minocycline肋膜沾粘術，第四組只接受胸腔鏡檢查。術後30天，我們將兔子犧牲，觀察其肋膜腔內沾粘及肋膜發炎的程度。結果顯示使用壁層肋膜摩擦術及minocycline肋膜沾粘術都會引起肋膜沾粘，但效果都不是很好。合併使用壁層肋膜摩擦術及minocycline肋膜沾粘術則會引起明顯之肋膜沾粘，沾粘之效果明顯優於單獨使用壁層肋膜摩擦術或minocycline肋膜沾粘術(p < 0.05)。只接受胸腔鏡檢查的對照組兔子肋膜則幾乎沒有沾粘發生。接著我們將兔子胸壁及肺臟作組織學檢查，發現不論使用壁層肋膜摩擦術、minocycline肋膜沾粘術、或合併使用兩種術式的兔子，其壁層肋膜皆有明顯發炎及纖維化的現象；但就臟層肋膜而言，minocycline肋膜沾粘術或合併使用兩種方式的兔子出現明顯臟層肋膜發炎及纖維化的現象，但壁層肋膜摩擦術的兔子卻沒有〈平均肋膜發炎指數0.5，和對照組類似，但和minocycline肋膜沾粘術或合併使用兩種術式有顯著差別〉。顯示minocycline會同時引起壁層肋膜及臟層肋膜發炎，而壁層肋膜摩擦術只選擇性引起壁層肋膜之發炎。透過本動物實驗，可知胸腔鏡手術後追加minocycline肋膜沾粘術一方面加強肋膜沾粘程度，一方面引發臟層肋膜發炎及纖維化，因而可以減少術後長期漏氣及氣胸復發之機率。
為了進一步尋找預測肋膜沾粘程度的因子，我們分別於胸腔鏡術後第24, 48, 及72小時抽取壁層肋膜摩擦術、minocycline肋膜沾粘術、或合併使用兩種方式兔子的胸水來進行細胞激素的研究。結果發現三組兔子胸水中TGF-β1的含量都會隨著時間的增加而上升，尤其是合併使用兩種方式的兔子胸水更為明顯：72 小時後，合併使用兩種術式的兔子其胸水中的TGF-β1含量明顯超越單獨使用壁層肋膜摩擦術或minocycline肋膜沾粘術的兔子。另外在所有的兔子中，其72小時時胸水中TGF-β1的濃度和30天後肋膜沾粘的程度也有密切相關：即TGF-β1的濃度越高，則以後肋膜沾粘程度越厲害。相反地，胸水的體積、紅血球數目、白血球數目、白血球分類、IL-6 及 IL-8的濃度和30天後肋膜沾粘程度則沒有相關。本研究的結果顯示胸腔鏡手術後72小時胸水中的TGF-β1濃度可能可以作為預測日後肋膜沾粘程度的指標。
為了驗證動物實驗的結果是否發生於實際病患身上，並了解胸腔鏡手術後追加minocycline肋膜沾粘術所引起細胞激素的變化，我們分別於minocycline組及對照組病患中，各抽取21及24位病患術前及術後第一天、第二天的肋膜積水來進行IL-6, IL-8及TGF-

關鍵字

minocycline ； 肋膜沾粘 ； 胸腔鏡手術 ； 氣胸 ； 迷你內視鏡手術 ； TGF-beta

並列摘要

Primary spontaneous pneumothorax most commonly occurs in young, tall, lean males. The most common cause of primary spontaneous pneumothorax is the rupture of an apical subpleural bleb. The estimated recurrence rate is 23-50% after the first episode and increases to 60-80% after the second pneumothorax. Optimal management of this benign disease has been a matter of debate. Traditionally, open thoracotomy has been considered the definite treatment for patients with recurrence. With this surgical approach, the offending bleb is found and removed, then, pleurodesis or pleurectomy is performed to encourage adhesion formation. While open thoracotomy provides the definite treatment in patients with recurrence, physicians are reluctant to refer patients for this treatment because of a long postoperative recovery, pain, and a high rate of complications. The advancement in video-assisted thoracic surgery (VATS) has enabled bullectomy and pleural abrasion through thoracoscopy to become a viable alternative in the treatment of primary spontaneous pneumothorax. However, problems associated with thoracoscopic management of primary spontaneous pneumothorax remained unsolved. First, the postoperative chest pain has not been completely alleviated and the surgical wounds of VATS have not been cosmetically perfect. Second, the recurrence rate of this procedure ranges between 4 % and 10% in most series and is generally higher than open thoracotomy. In this study, novel approaches were attempted to solve these problems.

Part I: Using Modified Needlescopic VATS in Primary Spontaneous Pneumothorax
Recently, the development of technology in the videoscopic field has produced 2-mm and 3-mm needlescopic equipments and instruments for thoracic procedures. Sympathectomy for hyperhidrosis and preoperative diagnostic work-up in general thoracic surgery fields using the needlescopic VATS technique has shown clinically excellent results with less postoperative pain and scarring. The initial reports also suggest that needlescopic VATS is a feasible technique in treating primary spontaneous pneumothorax. While needlescopic surgery may reduce the trauma inflicted on the chest wall, there are still some limitations to its use due to narrow field of vision, lower resolution, and difficulty in handling bulky tissue when compared with conventional VATS. Furthermore, no comparative or prospective studies have been conducted to prove that needlescopic VATS is less painful, equally effective, or more cosmetic. In addition, the patient number was small in these studies and the long-term safety and efficacy of this new technique is unknown.
In this study, we used a modified method and port strategy of needlescopic VATS technique to treat primary spontaneous pneumothorax. We compared the feasibility, complications, and short-term outcomes of this procedure with those of conventional VATS procedures. The safety and efficacy of the needlescopic VATS approach for treating primary spontaneous pneumothorax were also evaluated.

1. Needlescopic VATS is technically feasible and provides superior pain control and cosmetic results
From April 2001 to April 2002, a total of 63 consecutive patients requiring VATS due to recurrent, persistent, or contralateral spontaneous pneumothorax were enrolled in this study. Patients chose to undergo conventional VATS or needlescopic VATS by his or her own choice after the attending surgeon explained the differences between the two procedures. The cost of needlescopic VATS is higher than that of conventional VATS because the national health insurance program in this country will cover only a part of the disposable equipment.
The median age of the patients was 22 years (range: 14 to 48 years; 51 males, 12 females). Of these patients, 15 (23.8%) were smokers. The indications for operation were ipsilateral recurrence in 24 (38.1%), continuous air leaks in 33 (52.4%), and contralateral recurrence in 6 (9.5%). Blebs or bullae were identified in 57 (90.5%) patients. Among these patients, multiple blebs (>3 blebs) were visualized in 20 cases (35.1%), and the most common site for blebs was the upper lobe (86 %). Thirty-four patients chose conventional VATS and 29 chose needlescopic VATS. However, one patient in the needlescopic VATS group was converted to conventional VATS because air leaks could not be found. In the needlescopic VATS group, the initial 9 patients were managed with 2-mm instruments. The 3-mm instruments were used in the remaining 19 patients because one patient had early recurrence caused by a missed bleb. In addition, manipulation of the lung tissue with 2-mm instruments was inconvenient.
The two groups did not differ in age, sex, weight, body surface area, smoking status, side involvement, surgical indications, and operative findings. Time of operation, postoperative pain, request of opioid analgesics, and results of treatment were also comparable in each groups. No operative deaths and no major complications were reported in either of the two groups. Three patients (4.8%) had air leaks lasting longer than five days; these patients were managed conservatively. Pleural detachment was noted in one patient after removal of the chest drainage. Subcutaneous hematoma was noted in one patient in the conventional VATS group.
Patients underwent followe-up for a duration of 2 to 14 months, with a mean follow-up of 8 months. Only one patient in the conventional VATS group was lost to follow-up. Ipsilateral recurrence of pneumothorax occurred in one patient 3 weeks after the 2-mm needlescopic VATS. He underwent a subsequent conventional VATS and a missed bleb was noted at the lower edge of the right middle lobe of the lung.
After surgery, patients returned to work or school very early in the course of this study, with a mean interval of 10.0±2.7 days in the conventional VATS group and 9.9±2.4 days in the needlescopic VATS group. When the groups were compared for residual neuralgia, occasional or persistent pain on the operated side presented in 53% of patients in the conventional VATS group and only in 25% of patients in the needlescopic VATS group (p =0.021). In addition, more patients were very satisfied or satisfied with their scars in the needlescopic VATS group than in the conventional VATS group (86% vs. 75%, p =0.043).

2. Needlescopic VATS is a safe and valid alternative for the treatment of unilateral primary spontaneous pneumothorax
From April 2001 to October 2003, 110 patients with primary spontaneous pneumothorax who had undergone a needlescopic VATS procedure were identified. The median age was 21 years (range: 15 to 48 years; 93 males, 17 females), 28 (25.5%) were smokers. The indications for operation were ipsilateral recurrence in 69 (62.7%), continuous air leaks in 18 (16.4%), contralateral recurrence in 13 (11.8 %), bilateral pneumothorax in 4 (3.6%), associated hemothorax in 3 (2.7%), and uncomplicated first episode in 3 (2.7%). The needlescopic VATS procedure was unilateral in 102 patients. Eight patients had a bilateral procedure owing to bilateral pneumothorax (4 patients) or unilateral pneumothorax with obvious blebs or bullae on the contralateral lung by computed tomographic scanning (4 patients). Among the 118 procedures, blebs or bullae were identified in 111 (94.1%) cases, in which multiple blebs (>3 blebs) were visualized in 51 cases (43.2%). The most common site for blebs was the upper lobe (77.1 %). In three cases, the procedure was converted to conventional VATS because air leaks could not be found. The mean operation time was 75.7 + 23.3 minutes. After surgery, additional minocycline pleurodesis was undertaken in 65 cases (55.1%).
After the operation, the median visual analog scale rating decreased steadily during the first three postoperative days. Meperidine was requested by 71.2 % of the patients, mainly on the first postoperative day. The mean accumulated dosage of injection was 1.4 + 1.4 ampules. No operative deaths and no major complications were noted. Six patients (5.1%) had air leaks lasting longer than 5 days; these patients were managed conservatively. Pleural detachment was noted in two patients after removal of the chest drainage. One patient was treated with needle aspiration and the other was treated by tube thoracostomy. The mean postoperative hospital stay was 4.2 + 2.4 days.
Patients underwent follow-up for a duration between 3 and 32 months, with a mean follow-up of 16 months. Six patients (5.5%) were lost to follow-up. After the operation, patients returned to work or school very early, with a median interval of 7 days. In terms of postoperative residual neuralgia, only 16.3% of patients experienced occasional pain on the operated side. In addition, most patients (73%) were very satisfied or satisfied with their scars.
Ipsilateral recurrence of pneumothorax occurred in 8 of the 118 procedures. Among them, 5 recurrences developed within 6 months after the operation. The recurrent pneumothoraces were managed in several ways. Three patients underwent conventional VATS procedures with pleural abrasion. Two patients were treated with a chest tube. The remaining 3 patients had a small pneumothorax (< 15%) and were managed by observation.
To identify the risk factors associated with ipsilateral recurrence after needlescopic VATS, we entered 10 perioperative variables, including age, gender, smoking history, operative indications, number of blebs, site of blebs, unilateral or bilateral procedure, use of additional minocycline pleurodesis, dose of requested meperidine, and the presence of complications, into univariate analyses using the Cox proportional hazards regression model. Three variables were significant predictors of recurrence (p < 0.05): the number of identified blebs, unilateral or bilateral procedures, and use of additional minocycline pleurodesis. When 1 or 2 blebs were identified, the recurrence rate was 1.7% (1/60). When no bleb or multiple blebs (> 3) were noted, the recurrence rate was 12.0% (7/58) (p = 0.03). Bilateral procedures had a significantly higher rate of recurrence (31.3%, 5/16) compared to unilateral procedures (2.9%, 3/102) (p < 0.01). Recurrence occurred in 3.1% (2/65) of patients who had additional minocycline pleurodesis and in 11.3% (6/53) of those in whom minocycline was not used (p = 0.02). The multivariate analyses by the Cox proportional hazards regression model revealed that a bilateral procedure and no minocycline pleurodesis were significant and independent predictors of ipsilateral recurrence after needlescopic VATS (p < 0.001 and p = 0.019, respectively). The estimated rates of recurrence after the needlescopic VATS procedures were also significantly different between the unilateral and bilateral procedures.

Part II: Effects of Additional Minocycline Pleurodesis after Thoracoscopic Procedures for Primary Spontaneous Pneumothorax
Intrapleural instillation of a chemical irritant (chemical pleurodesis) is an effective way to shorten the duration of air leaks and prevent the recurrence of spontaneous pneumothorax in non-surgical patients. Tetracycline, which was the most commonly used irritant, is no longer available. Minocycline, a derivative of tetracycline, is as effective as tetracycline in inducing pleural fibrosis in rabbits. However, published studies with its use in human subjects are limited, with studies of small cohorts of patients focusing on the effect of diminishing air leaks. The safety and long-term efficacy on preventing recurrence of pneumothorax have never been addressed.
In the present study, we evaluate the effects of additional minocycline pleurodesis in treating primary spontaneous pneumothorax patients after thoracoscopic procedures by a retrospective study and a prospective randomized clinical trial. We also used animal studies to unveil the underlying mechanisms of minocycline pleurodesis. In addition, we tried to identify markers to predict the degrees of pleurodesis.

1. Minocycline pleurodesis: improved outcome and lower rate of recurrence after thoracoscopic treatment for primary spontaneous pneumothorax
Between January 1992 and April 2001, 364 consecutive patients with spontaneous pneumothorax were referred to our thoracic surgical division for VATS surgery. Before April 1994, nothing was administered to the pleural cavity after operation (control group). Since April 1994, additional minocycline pleurodesis was performed for the purpose of decreasing the incidence of prolonged air leaks and ipsilateral recurrence of pneumothorax after operation (minocycline group). The minocycline group consisted of 313 patients; the control group consisted of 51 patients. Their average age of the patients was 25 years (ranging = 13 to 59). Blebs or bullae were identified in 346 (95.1%) patients, and the most common site for blebs was the upper lobe (87 %). The operation duration was 90.3±31.8 minutes (ranging from 40 to 250 minutes). The two groups did not differ in age, sex, weight, smoking status, side involvement, surgical indications, surgical approaches, operative findings, and time of operation. The results of treatment showed that there was no hypersensitivity or adverse reactions for minocycline instillation. Chest pain was a common complaint after minocycline instillation and severe pain that required immediate meperidine injection occurred in 63 patients (20.1%). However, the total amount of meperidine was not significantly different between the two groups. Patients treated with minocycline had significantly shorter periods of chest drainage and postoperative hospital stay. No deaths occurred and no full thoracotomy was needed during and after the treatment. Thirty-one patients (8.5%) had air leaks lasting longer than 5 days. The rate of prolonged air leaks was significantly lower in the minocycline group. Pleural detachment was noted in 6 patients after removal of chest drainage. Hemothorax developed in 2 patients of the minocycline group. In one patient, massive bleeding occurred before instillation of minocycline and mini-thoracotomy was required to stop bleeding. The other patient had hemothorax 10 days after minocycline instillation. He was managed by tube thoracostomy. Wound infection was noted in 2 patients. No patient had pleural empyema in this study.
During the postoperative follow-up, recurrent ipsilateral pneumothorax was noted in 14 of the 364 patients, with a significant lower rate in patients with minocycline instillation (2.9% vs. 9.8%, p = 0.033). Using the Kaplan-Meier method, it appears that patients in the minocycline group had a lower rate of recurrence and that the recurrences took place at a later time (p = 0.036 by log-rank test). However, minocycline did not completely prevent recurrence. The median recurrence time was 4.0 (mean = 6.4) months in the control group and 13.3 months (mean = 17.7) in the minocycline group, with a recurrence rate within 6 months of 7.8% and 0.3% respectively (p = 0.002). To determine whether the difference of recurrence was associated with the learning curve of thoracoscopic technique, we stratified the recurrent patients in the minocycline group according to the first, second, and third groups of 100 patients. We found no obvious temporal trends of decreasing recurrence rate during different operation periods.

2. Prospective, randomized trial shows minocycline pleurodesis reduces recurrence rate after thoracoscopic treatment for primary spontaneous pneumothorax
Between June 2001 and February 2004, 202 patients with spontaneous pneumothorax were enrolled for this trial. Needlescopic VATS was performed in 142 patients (70%) and conventional VATS in 60 patients (30%). The operation procedures consisted of bullectomy and abrasion of the upper half of the parietal pleura. After the operation, patients were randomized to receive additional minocycline pleurodesis (7mg/kg, minocycline group) or not (control group). The minocycline group consisted of 102 patients, and the control group consisted of 99 patients. Their average age was 25 years (ranging from 15 to 50 years). Of these patients, 57 (28%) were smokers. Blebs or bullae were identified in 190 (94%) patients. Among them, 75 patients (37%) had multiple blebs (> 3 blebs). The mean operation duration was 79 minutes (ranging from 40 to 180 minutes). The two groups did not differ in age, sex, weight, smoking status, side involvement, surgical indications, surgical approaches, operative findings, number of endoscopic stapler cartridge, and time of operation.
The results of treatment showed that there was no hypersensitivity or adverse reactions for minocycline instillation. Chest pain was a common complaint after minocycline instillation and around 70% of the patients required meperidine injection. The mean dose of meperidine injection was 2.3 ampules in the minocycline group and 1.4 ampules in the control group (p = 0.002). Patients treated with minocycline had comparable chest drainage duration, postoperative hospital stay, and complication rates as the control group. No deaths occurred and no mini-thoracotomy was needed during and after the treatment. Eight patients (8.5%) had air leaks lasting longer than 5 days. Patients in the minocycline group have trends of a decreased rate of prolonged air leaks (2% vs. 6%, p = 0.1).
The patients underwent follow-up for a mean of 15 months (range = 1-32). After discharge from the hospital, patients in each group had a similar duration to return to work or school (with a mean of 8.3 days in the minocycline group vs. 8.2 days in the control group). Postoperative residual neuralgia and pulmonary function were also comparable in both groups. Patients’ pulmonary function test was performed at least 6 month after the operation and showed that the two groups had comparable forced vital capacity (FVC) and forced expiratory volume in one second (FEV1.0). During the postoperative follow-up, recurrent ipsilateral pneumothorax was noted in 2 patients in the minocycline group and 8 patients in the control group (2 % vs. 8%, p = 0.055). When these patients were stratified into low-risk (bleb number = 1 or 2) and high-risk groups (bleb number = 0 or > 3), minocycline pleurodesis decreased the rate of recurrence from 16% to 2 % in the high-risk group (p = 0.03), but had no obvious effects in the low risk group (2% vs. 2%). The results were similar when the Kaplan-Meier method was used to estimate the rate of recurrence after VATS.

Part III: Effects and Mechanisms of Additional Minocycline Pleurodesis
1. Combined pleural abrasion and minocycline instillation for best pleurodesis
Clinically we have shown that additional minocycline pleurodesis decreases the rate of ipsilateral pneumothorax recurrence in high-risk patients. In addition, minocycline pleurodesis may decrease the rate of prolonged air leaks after thoracoscopic treatment for primary spontaneous pneumothorax. However, the underlying mechanisms remain unknown. We hypothesize that additional minocycline pleurodesis produces better effects of pleurodesis than mechanical pleural abrasion alone, and animal studies were used to test this hypothesis.
Thirty six New Zealand white male rabbits weighing between 2.5 and 4.0 kg were randomized into four groups: 1, thoracoscopic parietal pleural abrasion (abrasion group = 10 rabbits), 2, thoracoscopic instillation of minocycline, 7mg/kg (minocycline group = 10 rabbits), 3, a combination of both methods (combination group =10 rabbits), or 4, thoracoscopic examination alone (controlled group = 6 rabbits). After the operation, a soft silastic catheter was inserted into the pleural space. The pleural effusions were collected at 24, 48, and 72 hours, and then the chest tube was removed. The total volume of pleural fluid aspirated was recorded. Pleural fluid analyses included RBC count, WBC count and classification, and IL-6, IL-8, and TGF-

並列關鍵字

pleurodesis ； thoracoscopy ； pneumothorax ； needlescope ； minocycline ； TGF-beta