Published ahead of print on December 15, 2005, doi:10.1164/rccm.200509-1414OC
© 2006 American Thoracic Society doi: 10.1164/rccm.200509-1414OC
Additional Minocycline Pleurodesis after Thoracoscopic Surgery for Primary Spontaneous PneumothoraxDivision of Thoracic Surgery, Department of Surgery, and Department of Traumatology, National Taiwan University Hospital and National Taiwan University College of Medicine, Taipei, Taiwan Correspondence and requests for reprints should be addressed to Yung-Chie Lee, M.D., Department of Surgery, National Taiwan University Hospital, No. 7, Chung Shan South Road, Taipei 10016, Taiwan. E-mail: wuj{at}ha.mc.ntu.edu.tw
Rationale: Ipsilateral recurrence rates of spontaneous pneumothorax after video-assisted thoracoscopic surgery are higher than rates after open thoracotomy. Objectives: This study was conducted to determine whether additional minocycline pleurodesis would be effective in diminishing recurrence after video-assisted thoracoscopic surgery treatment of primary spontaneous pneumothorax. Methods: Between June 2001 and February 2004, 202 patients with primary spontaneous pneumothorax were treated by conventional or needlescopic video-assisted thoracoscopic surgery. The procedures included resection of blebs and mechanical pleurodesis by scrubbing the parietal pleura. After the operation, patients were randomly assigned to additional minocycline pleurodesis (103 patients) or to observation (99 patients). Main Results: Patients in the minocycline group had higher intensity chest pain and required a higher accumulated dose of meperidine. Short-term results showed that the two groups had comparable chest drainage duration, postoperative hospital stay, and complication rates. Patients in the minocycline group demonstrated a trend of decreased rate of prolonged air leaks (1.9 vs. 6.1%, p = 0.100). After a mean follow-up of 29 mo (12–47 mo), recurrent ipsilateral pneumothorax was noted in two patients in the minocycline group and eight patients in the observation group (p = 0.044 by the Kaplan-Meier method and log-rank test). Postoperative long-term residual chest pain and pulmonary function were comparable in both groups. Conclusions: Although associated with intense immediate chest pain, additional minocycline pleurodesis is a safe and convenient procedure that can reduce the rate of ipsilateral recurrence after thoracoscopic treatment for primary spontaneous pneumothorax.
Key Words: minocycline • pleurodesis • primary spontaneous pneumothorax • thoracoscopy Primary spontaneous pneumothorax occurs most commonly in young, tall, lean males (1, 2). The etiology of this disease remains unclear although some mutations have been noted in patients with familial spontaneous pneumothorax (3, 4). The estimated recurrence rate is 23 to 50% after the first episode and increases to 60% after the second pneumothorax (5). Optimal management of this benign disease, especially after repeat attack, has been a matter of debate. Advances in video-assisted thoracoscopic surgery (VATS) that combine bullectomy with pleural abrasion provide a preferred intervention for treating primary spontaneous pneumothorax (6–8). However, the literature reports widely varied recurrence rates of pneumothorax after VATS. Several articles show significantly lower rates (9–11), but there are also articles demonstrating that the recurrence rates of pneumothorax after VATS range between 5 and 12%, which are higher than the rates reported after open thoracotomy (12–17). It has been suggested that a higher chance of missed leaking blebs and a less intense pleural inflammatory reaction are induced by the VATS procedure than by thoracotomy (18, 19). As a result, the efficacy of VATS has been questioned and more aggressive procedures, such as limited thoracotomy with pleurectomy, are sometimes performed to enhance the effects of pleural symphysis (15, 19). Intrapleural instillation of a chemical irritant (chemical pleurodesis) is an effective way to shorten the duration of air leaks and reduce the rates of recurrent spontaneous pneumothorax in surgical and nonsurgical patients (20, 21). However, guidelines recommend that chemical pleurodesis should be attempted only if the patient is either unwilling or unable to undergo surgery because the rates of recurrence of pneumothoraces after surgical intervention either by thoracotomy or VATS, with or without surgical pleurodesis, are far less than after chemical pleurodesis (6, 21–24). As a result, chemical pleurodesis has rarely been used in primary spontaneous pneumothorax, and the effect of additional chemical pleurodesis after surgical pleurodesis in preventing recurrence of pneumothoraces has never been evaluated. 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 (25). In a previous retrospective study, we found that additional minocycline pleurodesis is a safe and convenient procedure that may be associated with lower rates of prolonged air leaks and ipsilateral recurrence after thoracoscopic treatment of primary spontaneous pneumothorax (26). In the present study, additional minocycline pleurodesis was randomly administered in patients with primary spontaneous pneumothorax after VATS to test this hypothesis.
Study Design This study was a randomized trial of additional minocycline pleurodesis after VATS for primary spontaneous pneumothorax. It was begun in June 2001 in the Division of Thoracic Surgery of National Taiwan University Hospital (Taipei, Taiwan). After VATS for primary spontaneous pneumothorax, eligible patients were randomly assigned to additional minocycline pleurodesis or observation. A sample size of 200 patients (100 in each group) with at least 12 mo of follow-up was originally assumed. The primary endpoint was rate of ipsilateral recurrence after the operation. Secondary endpoints included safety, early postoperative results, and long-term effects of minocycline pleurodesis. Informed consent was obtained from patients after thorough explanation. The protocol was approved by the Institutional Review Board of National Taiwan University Hospital.
Eligibility Criteria
Operative Technique of VATS Conventional VATS was performed in a standard fashion under general anesthesia, using intubation with a double-lumen endotracheal tube. The patients were placed in a lateral decubitus position, and the ipsilateral lung was deflated. A 10-mm 30° telescope (Karl Storz, Tuttlingen, Germany) was first inserted through the previous chest tube wound to examine the pleural cavity. If a chest tube wound was not available, a 12-mm port was made at the sixth or seventh intercostal space. Two 15-mm skin incisions were made at the third or fourth intercostal space, anterior and posterior axillary line. Light pleural adhesions were freed by electrocautery. When blebs were identified, they were grasped with ring forceps and excised with a 45-mm endoscopic stapler. Blind apical stapling was done at the most suspicious area if no bleb could be identified. The entire parietal surface was abraded by inserting a dissector with a strip of diathermy scratch pad through the port sites. After postoperative lung reinflation, normal saline solution was instilled to check for air leaks. A chest tube (28 F) was placed in the apex through one of the insertion wounds. Surgical specimens were routinely sent for pathologic examination. The operation technique of needlescopic VATS has been described previously (27). The anesthesia, preparation, and operative procedures of needlescopic VATS are almost identical to those of conventional VATS. However, two sets of independent videothoracoscopic equipment and monitors, one for needlescopic videothoracoscopy and the other for 10-mm videothoracoscopy, were used simultaneously and placed near the patient's head. Basically, we used the 10-mm videothoracoscopy for most of the surgical steps. A needlescope was indicated only when we needed the chest tube wound to insert the endoscopic stapler and ring forceps, to extract the specimen, or to perform pleural abrasion.
Postoperative Care and Minocycline Pleurodesis
Randomization and Treatment Regimen To optimize the effects of pleural symphysis, minocycline was administered after full expansion of the lung, which was usually accomplished on the first postoperative day. However, in those patients who had persistent air leaks and could not fully inflate their lung, minocycline was still used after 3 to 5 d of waiting. In the observation group, nothing was instilled. A blinded study conducted by saline instillation was not suitable in our experiment because the pain associated with minocycline injection would unmask the blinding; and use of saline injection would place the patient at unnecessary risk of contamination of the pleural space. When prolonged air leaks developed in the observation group, patients could still undergo minocycline instillation although the analysis was based on an intent-to-treat method. The tube was removed in both groups when the lung was fully expanded and no air leaks were noted in a 24-h period. All complications after the operation were recorded. Prolonged air leaks were defined as air leaks lasting longer than 5 d. Pleural detachment was defined as a pneumothorax developing right after removal of the chest tube.
Follow-Up
Postoperative Pulmonary Function Analysis
Data Collection and Statistical Analysis
Characteristics of Patients Between June 2001 and February 2004, 235 consecutive patients with spontaneous pneumothorax were referred to the Division of Thoracic Surgery of National Taiwan University Hospital for VATS. Figure 1 shows the enrollment, assignment, and follow-up of patients in the study. Thirty-three (14.0%) patients were excluded from the study. Patients were excluded because of age greater than 50 yr (n = 16), underlying pulmonary disease (n = 10), previous ipsilateral VATS operation (n = 6), and unwillingness to undergo randomization (n = 1). The remaining 202 patients were randomized to receive minocycline pleurodesis (103 patients) or observation (99 patients) after VATS. Needlescopic VATS was performed in 142 patients (70%) and conventional VATS was performed in 60 patients (30%). Their average age was 25 yr (ranging from 15 to 50 yr) and 57 (28%) were smokers. Blebs or bullae were identified in 190 patients (94%). Among them, 75 patients (37%) had multiple blebs (3 or more blebs). In 12 patients, the air leaks could not be identified and blind apical stapling was performed. The mean operation duration was 79 min (ranging from 40 to 180 min). The two groups were evenly distributed with respect to age, sex, weight, height, smoking status, side involvement, surgical indications, surgical approaches, operative findings, number of endoscopic stapler cartridges used, and duration of operation (Table 1).
Minocycline Administration and Short-Term Results Minocycline was administered to all 103 patients in the minocycline group. Among them, two patients underwent instillation twice because of prolonged air leaks. In the observation group, four patients also underwent minocycline pleurodesis because of prolonged air leaks. There was no hypersensitivity or major adverse reactions to minocycline instillation. Chest pain was a common complaint after minocycline instillation. Meperidine was requested more commonly by patients in the minocycline group than by those in the observation group (83 vs. 66%, p = 0.006). The mean accumulated dose of meperidine injection was also higher in the minocycline group (115 mg in the minocycline group and 69 mg in the observation group; p = 0.002). The VAS rating showed that the main difference in chest pain between the two groups was on the first postoperative day, when minocycline was administered. The VAS rating decreased and became comparable on the second and third postoperative days. Patients treated with minocycline had comparable postoperative chest drainage duration, postoperative hospital stay, and complication rates as the observation group. No deaths occurred and no thoracotomy was needed during and after treatment. Eight patients (4.0%) had air leaks lasting longer than 5 d. Patients in the minocycline group demonstrated trends toward a decreased rate of prolonged air leaks (1.9 vs. 6.1%; p = 0.100), although the difference was not statistically significant. Pleural detachment occurred in three patients after removal of the chest tube. Empyema occurred in two patients. Wound infection occurred in four patients (Table 2). Patients with empyema were managed by tube thoracostomy and treatment with antibiotics. Other patients were managed conservatively. After discharge from the hospital, patients had similar duration for return to work or school in each group.
Residual Chest Pain and Recurrence Follow-up ranged from 12 to 47 mo (median, 30 mo) in the minocycline group and from 15 to 47 mo (median, 28 mo) in the observation group. Most patients were pain free 6 mo after the operation. The incidence and intensity of residual chest pain were comparable in both groups (Table 2). During the postoperative follow-up, recurrent ipsilateral pneumothorax was noted in two patients (1.9%) in the minocycline group and in eight patients (8.1%) in the observation group. Kaplan-Meier estimates of rates of freedom from recurrence are shown in Figure 2. Additional minocycline pleurodesis significantly reduced the rate of recurrence as compared with the observation group (p = 0.044 by the log-rank test).
Secondary analyses using the Cox regression model were performed to identify other factors associated with recurrence of pneumothorax. We found that bleb number was also a significant predictor for pneumothorax recurrence (p = 0.027). When one or two blebs were identified during VATS, the recurrence rate was 1.7% (2 of 117). When multiple blebs (3 or more) or no air leakage was noted, the recurrence rate was 9.4% (8 of 85). Minocycline pleurodesis is a borderline significant predictor by the Cox regression model (p = 0.060). Other factors, such as age, sex, smoking status, operation indications, operation method, number of endoscopic stapler cartridges used, meperidine use, dose of requested meperidine, and the presence of complications, were not associated with pneumothorax recurrence. We then stratified the patients into a low-risk group (bleb number, 1 or 2) and a high-risk group (bleb number, 0 or 3 or more) to evaluate the effect of additional minocycline pleurodesis. We found that minocycline pleurodesis decreased the rate of recurrence from 16.3 to 2.3% in the high-risk group (p = 0.030), but had no obvious effects in the low-risk group (1.7 vs. 1.8%, p = 0.999). Figure 3 shows Kaplan-Meier estimates of freedom from recurrent pneumothorax among the high-risk patients after stratification by minocycline pleurodesis. The recurrent pneumothoraces were managed in various fashions. In six patients (minocycline group, 1; observation group, 5), repeated VATS bullectomy with pleural abrasion and minocycline instillation (three patients) or pleurectomy (three patients) was undertaken. Three patients were treated with tube thoracostomy and minocycline instillation. One with a small loculated pneumothorax was managed by observation.
Pulmonary Function Analysis To evaluate the long-term effects of minocycline pleurodesis on pulmonary function, 102 patients (53 in the minocycline group and 49 in observation group) were available for spirometric measurement. The results showed that the two groups had comparable FVC and FEV1.0 (Table 3).
This prospective, randomized study documents the benefit of additional minocycline pleurodesis for the prevention of recurrence after VATS in patients with primary spontaneous pneumothorax. Minocycline therapy seems to be more effective in patients with multiple (3 or more) blebs or for whom air leaks cannot be identified. Additional minocycline pleurodesis also provides a trend toward decreasing the rate of prolonged air leaks. The main disadvantage of minocycline is immediate chest pain after instillation, which is relieved spontaneously within 3 d and does not impair pulmonary function or increase the risk of residual chest pain 6 mo after the operation. Since the introduction of VATS in the management of spontaneous pneumothorax in 1990 (28), it has become the preferred intervention for preventing pneumothorax recurrence. However, many authors still questioned the role of VATS as compared with the "gold standard"—that is, an open thoracotomy procedure. The major concern about VATS comes from its subsequent higher rate of recurrence. In a comprehensive review, Massard and colleagues suggested that the published recurrence rates of VATS range from 5 to 10%, in spite of a shorter follow-up period, which is higher than the 1% rates reported after open procedures (15). In this study, our results also show that, without additional minocycline instillation, the recurrence rate of VATS bullectomy and pleural abrasion is 8.1% after a median follow-up of 28 mo. The higher rates of recurrence after VATS may result from a less intense pleural inflammatory reaction induced by VATS than by thoracotomy (18, 19). Methods of pleurodesis have included mechanical abrasion and instillation of chemical irritants. Pleural abrasion reduces the rate of recurrence after VATS bullectomy (29). Previous controlled trials also showed that intrapleural tetracycline reduces the recurrence rates for spontaneous pneumothorax from 41 to 25% and from 36 to 13%, respectively (21, 23). However, the combined effect of the two methods after thoracoscopic bullectomy has rarely been mentioned in the literature. In our study, additional minocycline pleurodesis resulted in a significant decrease in the rate of ipsilateral recurrence and a trend toward a decreased rate of prolonged air leaks. The beneficial effect may be due to combined chemical and mechanical pleurodesis that inflames the entire pleural surface more evenly and thoroughly. This effect hastens the healing of blebs that are occasionally missed during thoracoscopic procedures and prompts symphysis of pleura to prevent the recurrence of pneumothorax. Many chemical irritants (tetracycline, minocycline, and talc) have been used to decrease the rate of recurrence or to hasten the healing of air leaks in the nonsurgical treatment of spontaneous pneumothorax. Tetracycline, which was the most widely used agent, is no longer available. Talc insufflation of the pleural cavity either by VATS or by medical thoracoscopy is safe and effective for primary spontaneous pneumothorax (30, 31), although it has been reported that talc may cause granuloma formation and decrease pulmonary function after long-term follow-up (32). In this study, minocycline was selected for chemical pleurodesis because it is a derivative of tetracycline and shares the same merits, that is, it is easily administered, safe, inexpensive, and widely available. Minocycline also compares favorably with tetracycline for producing pleurodesis in rabbits (25). In addition, the instillation of minocycline (7 mg/kg) produces satisfactory results in controlling postoperative air leaks and malignant pleural effusions in humans (20). Chest pain was the most common complaint associated with minocycline pleurodesis. Whereas a previous study suggested that an intrapleural dose of 250 mg of lidocaine is effective to control pain during tetracycline pleurodesis (33), our results showed that meperidine was immediately required in nearly 50% of patients in the minocycline group even though a large dose of intrapleural lidocaine (400 mg) was given as the pretreatment. In addition, the mean doses of meperidine injection were significantly higher in the minocycline group. The VAS rating showed that the main difference in chest pain between the two groups occurred on the first postoperative day, indicating that administration of minocycline did induce more intense chest pain. The VAS rating became comparable on the second and third postoperative days and the incidence of postoperative residual chest pain was similar, suggesting that the chest pain induced by minocycline persisted for no longer than 3 d and was not associated with a higher incidence of residual chest pain. In general, minocycline instillation was safe and no patient had pleural empyema. Other systemic side effects of minocycline such as neutropenia, thrombocytopenia, jaundice, and abnormal renal function were not noticed in our patients; however, the possibility cannot be completely eliminated because blood tests were not routinely performed after minocycline instillation. In theory, a more intense pleurodesis may result in constrictive pulmonary dysfunction. However, there is no evidence showing that additional minocycline pleurodesis significantly impairs pulmonary function. Previous studies showed that the mean vital capacity of patients with previously documented pneumothorax who did not have surgery was 87 to 88% of the normal predicted values (34, 35). By comparison, our data showed that the vital capacity after pleural abrasion or pleural abrasion plus minocycline pleurodesis did not change significantly. In addition, the FVC and FEV1.0 values were also comparable in both groups of patients, suggesting that minocycline pleurodesis does not cause significant impairment of pulmonary function after VATS. Although minocycline pleurodesis decreases the rates of recurrence after VATS, it may not be beneficial to all patients with primary spontaneous pneumothorax requiring VATS. Through secondary analysis, we found that additional minocycline pleurodesis is more effective in patients with multiple blebs (bleb number, 3 or more) or no identifiable bleb than in patients with one or two blebs. Previous studies and our results have shown that patients with multiple blebs or no identifiable bleb during VATS are associated with higher rates of recurrence, probably caused by an increased risk of overlooked small blebs or air leaks (7, 36). Under such conditions, intense pleural symphysis induced by additional minocycline pleurodesis is more favorable. On the contrary, if only one or two blebs were identified, the blebs can be definitely resected, making the rates of recurrence low despite any kind of pleurodesis. The results suggested that the treatment of spontaneous pneumothorax should be individualized according to different risks of recurrence. More aggressive pleurodesis should be considered for patients with higher risk of recurrence after VATS bullectomy. However, because this information did not originate from a preplanned analysis, more data and prospective studies are indicated to verify this finding. Previous studies suggest that minocycline may also decrease the rate of prolonged air leaks (20, 26). In this study, patients in the minocycline group presented trends toward a decreased rate of prolonged air leaks, although the difference was not significantly different. It is possible that the number of patients with prolonged air leaks in this study was so small that statistical significance could not be reached. This trial demonstrates that additional chemical pleurodesis can enhance the effect of mechanical abrasion in primary spontaneous pneumothorax. Patients with secondary spontaneous pneumothorax were excluded from this study because the performance status and underlying pulmonary diseases of these patients are heterogeneous. Whether patients with secondary spontaneous pneumothorax or intractable pleural effusions will benefit from this adjuvant remains unknown, but we believe that further studies are worthwhile to test the hypothesis. Some may argue that limited thoracotomy and pleurectomy also achieve the same or even decreased rates of recurrence compared with our patients in the minocycline group. In practice, limited thoracotomy creates a large wound that carries a higher incidence of residual chest pain and poor scar satisfaction. Dense and permanent pleurodesis induced by pleurectomy makes further operation difficult and may impair pulmonary function. Compared with the above disadvantages, VATS with additional minocycline pleurodesis is a safe, easy, effective, and minimal invasive alternative with satisfactory results. It has been observed in an animal study that early pleural fibrosis induced by minocycline faded gradually after 6 mo (37). This explains why our patients in the minocycline group had comparable pulmonary function as in the observation group. However, the effect of minocycline is still significant in this study because many of the recurrences happened within this period (26). In summary, this study indicates that additional minocycline pleurodesis provides a safe, convenient, and effective procedure to prevent ipsilateral recurrence after VATS for patients with primary spontaneous pneumothorax. The effects could have been additive or the result of different mechanisms of pleural inflammation induced by minocycline pleurodesis and pleural abrasion. We recommend the liberal use of this adjuvant, especially for patients who are likely to have missed blebs or inadequate mechanical pleurodesis after VATS. Modifications, such as instilling minocycline during general anesthesia to reduce pain or the use of other irritant agents, may also be tested to optimize the effects of adjuvant chemical pleurodesis.
Supported by research grants from the National Science Council, Taiwan (NSC 93–2314-B-002–237, to J.-S.C.), and by the National Taiwan University Hospital (NTUH 94-S12, to J.-S.C.), Taiwan. Originally Published in Press as DOI: 10.1164/rccm.200509-1414OC on December 15, 2005 Conflict of Interest Statement: None of the authors have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form September 9, 2005; accepted in final form December 15, 2005
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
25 yr of age), sex (male vs. female), smoking status (yes vs. no), operation indication (ipsilateral recurrence vs. others), operation method (needlescopic vs. conventional VATS), number of blebs (1 or 2 vs. 0 or 3 or more), number of endoscopic stapler cartridge, meperidine use (yes vs. no), dose of requested meperidine, presence of complications (yes vs. no), and minocycline pleurodesis (yes vs. no). The number of blebs was arbitrarily used to allocate patients into the low-risk group (1or 2 blebs) versus the high-risk group (0 or 3 or more blebs) because "0 bleb" indicates that the source of the leak was missed and "3 or more blebs" indicates that missed blebs are likely owing to significant disease. A p value less than 0.05 was considered statistically significant. 
