Video-Assisted Thoracic Surgery: Primary Therapy for Spontaneous Pneumothorax?

Transcription

1 Video-Assisted Thoracic Surgery: Primary Therapy for Spontaneous Pneumothorax? F. Hammond Cole, Jr, MD, Francis H. Cole, MD, Alim Khandekar, MD, J. Matthew Maxwell, MD, James W. Pate, MD, and William A. Walker, MD Section of Thoracic Surgery, University of Tennessee Center for the Health Sciences, Memphis, Tennessee, and Department of Thoracic SurgeD', Naval Medical Center San Diego, San Diego, California Background. This study assessed the role of videoassisted thoracic surgery (VATS) in current therapy for spontaneous pneumothorax. Methods. We compared a retrospective series of 89 patients treated conventionally with a consecutive group of 30 patients undergoing VATS pleural abrasion. The 89 earlier patients were predominantly male (819~). Treatment groups included observation/aspiration (7 or 17~:~), tube thoracostomy (32 or 36~/~ ), multiple tubes (7 or 99t), and thoracotomy (43 or 48'~). Of the 30 patients treated with VATS, 18 (66%) were male. Primary indications for operation were recurrent pneumothorax (17) and persistent air leak (9). Results. Hospital lengths of stay (LOS) for the earlier group were 5 days for simple tube and 7 days for primary thoracotomy; LOS for initial intervention followed by thoracotomy exceeded 15 days in all subgroups. The average LOS in the VATS group was 13 days; 6 patients treated with primary VATS (no chest tube) had a mean LOS of 6.5 days. Complications included 3 (10%) prolonged air leaks (more than 7 days) and 2 (7%) early recurrences. Conclusions. We do not recommend VATS as primary therapy for spontaneous pneumothorax; tube thoracostomy remains the treatment of choice. However, we strongly support surgical intervention early (3 days) in patients with a persistent air leak, and as primary therapy in a nonurgent situation if standard indications exist. This study shows no advantage of VATS over conventional thoracotomy in hospital stay or complication rate. (Ann Thorac Surg 1995;60:931-5) he advent of video-assisted thoracic surge D, (VATS) T has renewed interest in the therapy of spontaneous pneumothorax. We retrospectively reviewed a series of 89 patients treated by a single thoracic surgical group (F.H.C., F.H.C. Jr, and A.K.) at Methodist Hospitals of Memphis. Results in these patients were compared with a prospective, nonrandomized series of 32 pneumothoraces in 30 patients treated with VATS (all authors). We were particularly interested in hospital length of stay (LOS) and recurrence rates in these patients, and in whether we could justify VATS as therapy for the initial occurrence of spontaneous pneumothorax. Material and Methods The 89 patients in the retrospective studv were divided into two groups based on their response to initial therapy. Group I consisted of 41 patients (46%) who responded to initial therapy and had uncomplicated hospital courses. Group [1 consisted of 48 patients (54%) who required repeat therapeutic maneuvers to achieve a satisfactory result. Group 111 consisted of the 30 VATS patients treated at Methodist Hospitals of Memphis, University of Tennessee Center for the Health Sciences, and Naval Medical Center San Diego by all authors, who Presented at the Fortx-First Annual \ t'eling ot the Southern lhoracic Surgical Association, Marco Island, I:L N~x 10 12, Address reprint requems to Dr t t lanlnlond (.de, 1325 lastmoreland, Suite 310, Memphis, TN had similar training and VATS experience and were rnembers of the University of Tennessee Section of Thoracic Surgery. Management policies in groups I and II were consistent throughout the study period. Patients (n = 6) with small pneumothoraces were treated by observation only. Most patients (n 73) had initial tube thoracostomies under local anesthesia; operative intervention was reserved for prolonged air leaks (>5 to 7 days), recurrent pneumothorax, or the second ipsilateral or first contralateral pneumothorax. Axillary thoracotomy was done for primary pneumothorax with roentgenographic evidence of significant collapse and absence of adhesions; posterolateral thoracotomy was done for secondary pneumothorax with adhesions. Surgical therapy included closure of air leaks, bleb or bullae resection, and dry gauze pleural abrasion. Pleurectomy and chemical pleurodesis were not employed. Follow-up visits were at 1 week and 3 months after operation and included chest roentgenogram and clinical evaluation. We did not carry out questionnaire follow-up or formal long-term checkups on all patients in the retrospective group; complications were noted as they presented but were not formally sought and may be underreported. Since 1991, VATS techniques have been used in 30 patients. Indications for VATS were essentially as described for the earlier patients, although the definition of "prolonged air leak" was closer to 3 days. A prospective regist~, was maintained recording demographic data, 1995 by The Societ\ of lhoracic Sur~eon~ /95]$9.50 SSD[ (95)00598-F

2 932 COLE ET AL Ann Thorac Surg VATS PNEUMOTHORAX 1995;60:931-5 Table 1. Demographics Variable Group 1 Group II Group lll Total Male 32 (78%) 40 {83",,) 19 {63%) Female 9 (22%) 8 (17%) 11 (37",,) Mean age (y) " Side 22 R, 19 L 29 R, 19 L 18 R, 10 L, 2 bilateral p < 0.01 versus groups I and I1. L = left; R right. preoperative and postoperative course, duration of operation, and total and postoperative LOS. Follow-up was more thorough than in groups I and 1I in that inquiries about complications were concurrently recorded. Video-assisted thoracic surgical procedures were accomplished with general endobronchial anesthesia. A 12-ram trochar, through which the rigid thoracoscope was placed, was introduced through the sixth or seventh intercostal space. Secondary incisions, usually two, were placed under direct intrathoracic visualization. Instruments were frequently inserted directly through the incisions without further ports. Blebs were resected with the Endo-GIA stapling device (US Surgical, Norwalk, CT); pleural abrasion was mechanical and carried out with dry gauze passed into the chest. Most patients returned directly to the ward without an intensive care stay; analgesia was intravenous morphine sulfate and scheduled ketorolac. Data in each group were tabulated and are presented as mean. Means were compared between each group by the paired t test using the Bonferonni approach. Values were considered statistically significant when the p value was less than Comment The surprisingly long LOS of the "uncomplicated" pneumothorax patients prompted an evaluation of a more aggressive surgical approach to spontaneous pneumothorax. Video-assisted thoracic surgery allows inspection of the entire lung, identification and resection of bullous disease, and gauze pleurodesis, essentially duplicating an open procedure [1]. The issues involving a recommendation of VATS as the initial procedure for primary spontaneous pneumothorax involve recurrence rates after treatment, hospital LOS, postoperative pain, and the discomfort of a chest tube under local anesthesia versus a more extensive procedure under general anesthesia. Our study did not compare pain in the groups, but we are impressed with the frequency and severity of pain with VATS. Published literature indicated that recurrence rates after tube tboracostomy alone are as high as 60% [2]. In a prospective, randomized trial of tetracycline pleurodesis versus tube alone, recurrence rates were 40% for tube alone and 25% for the tetracycline group [3]. Almind and associates [4] compared simple drainage, talc pleurodesis, and tetracycline instillation and found recurrence rates of 36% for drainage, 13% for tetracycline, and 8% for talc. Before the availability of VATS, simple thoracoscopy with direct visualization of the pleural space and lung and chemical pleurodesis was frequently used in Europe. Vanderschueren [5] recommended thoracoscopy as the initial intervention, using local anesthesia and talc poudrage as the chemical agent in all patients with limited disease. He reported recurrence rates as low as 6'% although there are theoretical problems with placing pleural talc in young patients. Other sclerosing agents have been placed through the scope, including tetracycline, doxycycline, and fibrin glue [6]. Results Demographic and anatomic data are detailed in Table 1. There is a significant difference (p.: 0.01) in mean age distribution between the earlier patients (groups I and II) and VATS patients (group Ill), reflecting differences in the three institutions involved in the latter series. The increased age of the earlier patients might indicate a greater incidence of seconda D' pneumothoraces. Table 2 delineates the LOS for each group and subset, along with the details of treatment modalities. In patients in whom tube thoracostomy preceded surgical therapy, the LOS was 15 days; patients successfully treated with simple tube thoracostomy had a 5-day LOS. The 5 patients in group 11I who required open thoracotomy and the 6 patients who had only primary VATS had identical postoperative LOSs of 6.5 days. Complication rates were not comparable for reasons discussed above. In the VATS series, 2 patients (7%) had prolonged air leak, 2 (7%) had severe chest wall pain postoperatively, and 3 (10%) had early recurrence. Five patients (17%) were converted to open thoracotomy. There was one early recurrence (3%) in the 43 thoracotomy patients treated in groups I and II. Table 2. Treatment Modalities" Procedure n LOS (days) Group I (n :: 41) Obs 6... Aspiration 1... Tube 32 5 Thoracotomv 2 7 Mean 8 Group II (n 48) Obs-Thoracotomy 2 30 Obs-Tube 5 5 Tube-Thoracotomy Mean 19 Group lli {n 30) Tube-VATS Converted to open days postop VATS only Mean 11.4 ' Fhere is a significant (p ) difference in LOS between all three groups. los length,7 sta~; Obs observation; VATS video-assisted fl3oracoscopic surgery.

3 Ann "[horac Surg COLE ET AL ;60:931-5 VATS PNEUMOTHORAX The timing of definitive surgical repair in the management of spontaneous pneumothorax with prolonged air leak has been steadily moving toward a shorter observation period. Granke and associates [7] presented, at the meeting of the Southern Thoracic Surgical Association, a series of 119 patients with spontaneous pneumothorax, of whom 78 underwent thoracotomy. Postoperative LOS was 8 days for the operative group; Granke and associates recommended operation in patients with an air leak persisting more than 72 hours. Schoenenberger and colleagues [8] noted that closure of air leakage was maximal at 48 hours, and recommended proceeding to thoracotomy thereafter. Murray and associates [9] used an axillary thoracotomy as the primary therapy for the initial episode of pneumothorax because of low morbidity, short hospital stay, and low recurrence rate. Their mean postoperative LOS was 4.2 days. There were no recurrences in this group of 14 patients, Donahue and colleagues [10] reported 83 operative pleurodeses with 5 (5.6%) early and 3 (3.6%) late recurrences. Maggi and associates [11] reported a series of 94 open pleurodeses with no recurrence during the follow-up period. The key questions regarding VATS with pleural abrasion are (1) whether these procedures are equivalent to open procedures with regard to a low recurrence rate, and (2) whether morbidity will be lower and hospital LOS be shorter to warrant use of VATS in lieu of tube thoracostomy as primary therapy for the initial episode of spontaneous pneumothorax. Hazelrigg's group [1] compared 26 patients having VATS with 20 separately selected patients managed with traditional thoracotomy. Mean hospital stay was less, parenteral narcotic use after 48 hours was less, and there were no recurrences in 26 patients followed up for 8 months. Other series, however, have shown higher recurrence rates with VATS [12]. Interbitzi and colleagues [13] reported 12 cases using VATS with apical pleurectomy, and reported no recurrence in patients followed up for a mean of I year. The same authors reported a larger series [141 using initial diagnostic thoracoscopy under local anesthesia, followed by formal VATS if the pathology warranted it. They eliminated the initial step because the vast majority, of patients went on to VATS, and currently recommend VATS as the initial therapy for spontaneous pneumothorax. They reported a recurrence rate of 6.2% in this larger series with a longer follow-up period. In another recent report, Waller and colleagues [15] reported a randomized series of surgical therapy versus VATS. The only significant postoperative difference was in forced vital capacity and forced expiratory volume in 1 second measured 3 days after operation; morphine requirement seemed lower in the VATS group but did not reach statistical significance. Mean LOS was only 4 days, but 3 patients (10%) in the VATS group required thoracotomy. Long-term recurrence of pneumothorax was significant: 2 of 29 patients (6.9%) with VATS had recurrence and 1 patient died. The complications in both groups occurred only in patients with secondary pneumothoraces. To summarize the literature and our own data: (1) Tube thoracostomy can be accomplished with local anesthesia and has a slightly decreased hospital stay, but carries a recurrence rate up to 60%. (2) Conventional thoracotomy remains the "gold standard," with a recurrence rate of less than 1% and hospital stays of approximately 6.2 days. (3) Video-assisted thoracic surgery (in our hands) had an unacceptably high early recurrence rate, although results do appear to be improving with more aggressive pleural abrasion. Most of the theoretical advantages of VATS, such as decreased morbidity and earlier return to work, were not addressed in this report; the 7% incidence of chronic incisional pain in our VATS group is disturbing. We believe that preliminary tube thoracostomy for the initial spontaneous pneumothorax is appropriate, that "prolonged air leak" should be defined as 48 hours, that operative intervention with another pneumothorax should be undertaken without a preliminary tube, and that recommendation of VATS over conventional thoracotomy (on the basis of our recurrence and complication rates) is not yet justified. References 1. Hazelrigg SR, Landreneau RJ, Mack M, et al. Thorascopic stapled resection for spontaneous pneumothora. Ann Thorac Surg 1993;105: Voge VM, Anthracite R. Spontaneous pneumothorax in the USAF aircrew population: a retrospective study. Aviat Space Environ Med 1986;57: Light RW, O'Hara VS, Moritz TE, et al. Intrapleural tetracycline for the prevention of recurrent spontaneous pneumothorax. JAMA 1990;264: , 4. Almind M, Lange P, Viskum K. Spontaneous pneumothorax: comparison of simple drainage, talc pleurodesis, and tetracvcline pleurodesis. Thorax 1989;44: \;anderschueren RG. The role of thoracoscopy in the evaluation and management of pneumothorax. Lung 1990;168 (Suppl): Hansen MK, Kruse-Anderson S, Watt-Boolsen, Andersen K. Spontaneous pneumothorax and fibrin glue sealant during thorascopy. Eur J Cardiothorac Surg 1989;3: Granke K, Fischer CR, Gago O, Morris JD, Prager RL. The efficacy and timing of operative intervention for spontaneous pneumothorax. Ann Thorac Surg 1986;42: Schoenenberger RA, Haefeli WE, Weiss P, Ritz RF. Timing of invasive procedures in therapy for primary and secondary spontaneous pneumothorax. Arch Surg 1991;126: Murray KD, Matheny RG, Howanitz EP, Myerowitz PD. A limited axillary thoracotomy as primary treatment for recurrent spontaneous pneumothorax. Chest 1993;103: Donahue DM, Wright CD, Viale G, Mathisen DJ. Resection of pulmonary blebs and pleurodesis for spontaneous pneumothorax. Chest 1993;104: Maggi G, Ardissone F, Oliaro A, Ruffini E, Cianci R. Pleural abrasion in the treatment of recurrent or persistent spontaneous pneumothorax. Results of 94 consecutive cases. nt Surg 1992;77: Melvin WS, Krasna MJ, McLaughlin JS. Thoracoscopic management of spontaneous pneumothorax. Chest 1992;102: lnterbitzi RGC, Furrer M, Striffeler H, Althaus U. Thoracoscopic pleurectomy for treatment of complicated spontaneous pneumothorax. J Thorac Cardiovasc Surg 1993;105: lnterbitzi RGC, Leiser A, Furrer M, AIthaus U. Three years' experience in video-assisted thoracic surgery (VATS) for spontaneous pneumothorax. J Thorac Cardiovasc Surg 1994; 107: Waller DA, Forty J, Morritt GN. Video-assisted thoracoscopic surgery versus thoracotomy for spontaneous pneumothorax. Ann Thorac Surg 1994;58:372-7.

4 934 COI.E ET AL Ann Thorac Surg \/A1S PN t!l MO IHORAX 1995;60:931-5 DISCUSSION DR JOSEPH 1. MILLER, JR (Atlanta, GA): Doctor Cole, 1 appreciate your asking me to discuss your paper and the opportunity of receiving the manuscript before presentation. 1 rise mainly to second vour conclusions. Basically when VATS became available at a national level in 1991, enthusiasm vxas quite high in the utilization of this technique, potentially m the role of spontaneous pneumothorax, and 1, like a number of other people around the country, employed this initially in our practice, tlowever, it was in the calendar year received my first, not my own but a recurrence, and then during the calendar \ear 1 reoperated on 7 patients from outside the Emory system with the conventional techniques who had failed VATS done in the state of Georgia. That led to my mvn conviction that VATS ~as probably not better than the usual axillarv thoracotomv for this situation. During 1992 and 1993 l had the opportunity to give the talk at the Dc'orld Congress on Thoracoscopy on complications of VATS, and Dr Steven Hazelrigg provided me access to data from the VATS study bank. Time national failure rate of VATS is around 7",,, with up to a 10% failure rate reported. In this particular application of \ATS, it vxas hard for me to see where three l'/4-cm incisions were am different than, sa~, a 7~M - to 9-cm incision through the standard axillar} approach. In a personal operative experience at Emnrv over a 20-year period 1 had 176 patients either through axillary thnracotomy or limited thoracotomv with no recurrence rate and no operative mortality. It is hard for me to think that we should take an operation that has essentially a 100"<, success rate, that is, when done by conventional methods by either of the other two approaches, and convert it into an operation with a 7", national failure rate done with many individuals. 1 think the standard of care in most places would be an axillarv thoracotomv in the young patient, and in the more complicated, older patient, you might need to go to a limited posterolateral thoracotomv where you have better access, and then combine that ;xith a pleurodesis or pleurectomy or ~xhatever seems appropriate. DR KEITH S. NAUNHEIM (St. I,ouis, MO): I alwaxs hesitate to state aim opinion opposite to that of Dr Miller; 1 do admire his judgement and know he has a phenomenal clinical experience. However, I think I have to take a stance as one of the young turks who thinks that VATS is actually aim advance in thoracic surgery. 1 would say that if vou looked at the early results for atrial septal defect closure on pump bypass, mitral commissurotomy, open mitral commissurotomy, or any of those procedures we would think of as significant advances in the evolution ot cardiothoracic surgery, the morbidity and time mortality of the earliest experience was excessive and, when compared ~xith the experience from prior surgeons who were not using that technique, suggested prohibitive morbidity and mortality. I think that Dr Cole has made an excellent point. His 30 patients do represent the learning curve for three different institutions, an average oi 10 per institution. We recently' presented at the meeting of the Western Thoracic Surgical Association the combined series from Rodnev Landreneau, Steven Hazelrigg, myself, Mark Ferguson, and Michael Mack. We found a 4", rectu'rence rate in 1211 procedures; that was the early learning curve, if you will, for even those surgeons who have the greatest experience. I beliexe that VATS can be perforlned with equal efficacy once the experience is gained. I believe that the incidence of pain immediateh' postoperatively is less, even than with aim axillarv thoracotomv. 1 think the routine patient can be discharged within 48 hours of operation, and I think VATS well serves the vast majority of patients. One important point is that VATS allows visualization of the entire pleural space and the entire lung including the diaphragmatic surfaces. This is impossible through an axillary thoracotomv without the use of a thoracoscope. As I say, 1 hesitate ever to contradict Dr Miller, but hopefully he will not hold it against me after this session. DR COLE: Thank vou for those remarks. 1 certainly do not disagree with them. 1 also like the VATS technique, butwe were somewhat struck by our results. 1 must say that our length of stay is something l need to go home and improve. As one who has been married for 29 vears and has raised three teenagers, I did not need more humility, but when I learned yesterday that Dr Treasure's posttuberculosis pneumonectomy patients go home quicker than some ot my chest tube patients, l was very humbled. DR SAFUH ATTAR (Baltimore, MD): 1 felt compelled to stand and oppose the views propagated by Dr Cole and Dr Miller in presenting the data from my colleague, Dr Mark Krasna. He is quite experienced in video thoracoscopy, and 1 think this bears a hit of importance on the results. Doctor Krasna has had experience with 400 thoracic videoscopies so far, including 29 cases of video-assisted treatment of spontaneous pneumothorax between February 1991 and October Fhere are t~xo points in which we disagree with Cole and associates. First, there is no indication to operate on a patient with spontaneous pneumothorax at the initial presentation with direct operation or VATS. I think a tube thoracotolny should be enough. Therefore all our procedures were done for recurrent or persistent air leaks and none were done for initial spontaneous pneumothorax. We had only 2 patients who had recurrence postoperatively, and this was due to a learning curve. One of them had metastatic carcinoma to the lung that presented as a spontaneous pneumothorax. The second patient had the same thing, and had a cautery pleurodesis rather than pleurectomy. I hese two were the only failures. Our pleurectomy extended from the first to seventh rib, and behlw this level, pleurodesis was performed. The mean chest tube duration sxas 1.7 days. There were no recurrences in all these cases. The causes of failure we believe are due to missing the source of the leak, and there are certain tricks that can be used to detect the leak: by submerging the lung between air and fluid, and by injection of methylene blue. I wonder whether these tricks were used in localizing the site of the air leak. I enjoyed tl3e presentation very much. I)R LYNN H. HARRISON (Marrero, LA}: [ think that there are two learning curves that are involved in the implementation of this technique. One of them is the technical learning curve, but the other one is learning to which patients to apply the technique. In a sonlewhat smaller series of 23 patients with pneumothorax in w'hom we have applied the technique, only 7 had what I ~xould refer to as primary spontaneous pneumothorax occurring in an asthenic, young individual; in that group the eldest patient in our series was 28 vears. I notice that 30% in your series were older than 50 years and at least I had bullous disease ot the hmg. I think if you are going to start taking on that kind oi patient, then vnur success rate is indeed going to plummet. 1 think a more useful segregation of your patients is not simple

5 Ann Thorac Surg COLE ET AL ;60:931-5 VA]S PNEUMOTHORAX and complex but those with and without acquired parenchymal disease of the lung. If you look just at the young patient who has congenital apical blebs either in the apical or in the superior segments, then your success rate will be very good. We have had no ipsilateral recurrences and one contralatera] recurrence in those 7 patients in whom we have applied the technique in that group. I second your cautionary~ note regarding the use of talc in young patients. Some of those patients will come back with a primary lung neoplasm at some point in their lives, and 1 only wish that the surgeon who applied the talc could be the one who has to deal with that operative procedure. DR JOHN R. BENFIELD (Sacramento, CA): We started with VATS among the earliest in the counto', and we have used it enthusiastically for pneumothorax treatment. Nonetheless, we have found a certain degree of frustration in not being able to approach the pleura as directly with VATS as through traditional axillar), thoracotomy. We also learned that when one applies the argon beam laser to the pleura, one may cause a good deal of pain, and we have stopped using the argon beam to cause pleural scarring for pleurodesis. We have reached a compromise position, somewhere between what Dr Naunheim has said and what Dr Miller has said. We now use the axillary thoracotomy, usually without rib spreading, as a small access incision. This gives direct access, and then we use the telescope to see those areas of the pleura that we were not able to see with the traditional axillary thoracotomy. So 1 come as someone from the West hoping to serve as arbitrator and to suggest a compromise here in the South. DR COLE: 1 thank the discussants for their remarks. Obviously I think we have two separate issues. The first issue is whether it is reasonable to begin invasive therapy for primary pneumothorax. In favor of that are the 40% to even 60% recurrence rate after the initial primary pneumothorax and the substitution of a general anesthetic for a local anesthetic procedure. I think that certainly Murray's group recommends primary axillary thoracotomy in a fairly small series, and also Interbitzi has actually recommended primary VATS. The second issue is, whatever your indications for invasion are, is one going to do an axillary thoracotomv or a video-assisted procedure? 1 think in our own hands we simply need to work on our patient selection and length of stav a little bit more, and this is probably an unsettled issue. Bound volumes available to subscribers Bound volumes of the 1994 issues of The Annals of Tlwraci~ Surgery are available only to subscribers from the Publisher. The cost is $99.00 (outside US add $25.00 for postage) for volumes 57 and 58. Each bound volume contains a subject and author index, and all advertising is removed. The binding is durable buckram with the name of the journal, volume number, and year stamped on the spine. Payment must accompany all orders. Contact Elsevier Science Inc, 655 Avenue of the Americas, New York, NY 10010; or telephone (212) (facsimile: (212) ).