Complications of Video-Assisted Thoracoscopic Lung Biopsy in Patients with Interstitial Lung Disease

Complications of Video-Assisted Thoracoscopic Lung Biopsy in Patients with Interstitial Lung Disease Mary Elizabeth Kreider, MD, MS, John Hansen-Flaschen, MD, Nadia N. Ahmad, MD, Milton D. Rossman, MD, Larry R. Kaiser, MD, John C. Kucharczuk, MD, and Joseph B. Shrager, MD Pulmonary, Allergy and Critical Care Division, and Thoracic Surgery Division, University of Pennsylvania School of Medicine; and the Department of Medicine, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania Background. Current guidelines recommend surgical lung biopsy for diagnosis of interstitial lung diseases (ILDs) in selected patients. To shed light on the risk benefit ratio for this recommendation, we examined the morbidity and mortality associated with video-assisted thoracoscopic surgical (VATS) lung biopsy in a group of outpatients. Methods. A retrospective cohort study was conducted of 68 consecutive ambulatory patients with radiographically apparent interstitial lung disease (ILD) referred for VATS biopsy during a 6-year period. Incidence of postoperative mortality, prolonged air leaks, pneumonias, and re-admissions were calculated. Risk factors for complications of surgery were examined. Results. Three deaths occurred within 60 days after biopsy for a mortality rate of 4.4% (95% confidence interval [CI], 1% to 12%), and 19.1% (95% CI, 11% to 31%) experienced one or more complications of surgery. Risk factors for morbidity included preoperative dependence on oxygen therapy and pulmonary hypertension. The three patients who died had usual interstitial pneumonia on their biopsy specimen and were reintubated postoperatively for acute lung injury. Aggregation of articles published over the past 10 years reporting on surgical lung biopsy for the diagnosis of ILD yielded a postoperative mortality rate of 2% to 4.5%. Conclusions. VATS lung biopsy for diagnosis of ILD, even in ambulatory patients, is not an entirely benign procedure. Biopsy rarely may trigger an acute exacerbation of usual interstitial pneumonitis. The risk of postoperative complications appears to be greatest in those dependent on oxygen and those who have pulmonary hypertension. This information may be used in weighing the risk benefit ratio of biopsy in individual patients. (Ann Thorac Surg 2007;83:1140 5) 2007 by The Society of Thoracic Surgeons In 2002, the American Thoracic Society (ATS) and the European Respiratory Society released a joint statement on the classification and diagnosis of the idiopathic interstitial pneumonias (IIPs) [1]. The recommended scheme classification relies heavily on lung pathology: Because the histological patterns seen by pathologists usually allow for better separation of these entities than the imaging patterns seen by radiologists, these histological patterns provide the primary basis for the various categories of the IIPs and serve as the foundation of the classification... In the absence of contraindications, surgical lung biopsy is advised in patients with suspected IIP who do not show the classic clinical and HRCT (high resolution chest CAT scan) picture of IPF/UIP. Accepted for publication Oct 2, 2006. Address correspondence to Dr Kreider, 826 W. Gates, Hospital of the University of Pennsylvania, 3400 Spruce St, Philadelphia, PA 19104; e-mail: kreiderm@mail.med.upenn.edu. Although the benefits of surgical lung biopsy are well defined in the statement, the risks and contraindications are not. In the current study, we performed a retrospective cohort analysis of all outpatients who underwent videoassisted thoracoscopic surgical (VATS) lung biopsy to establish a specific diagnosis after presentation with clinically and radiographically apparent interstitial lung disease (ILD). Our objectives were (1) to determine the incidence of major postoperative complications for this diagnostic procedure, including death, pneumonia, prolonged air leaks, prolonged hospital stay, and hospital readmission; and (2) to compare the clinical characteristics of patients in whom these complications develop with the characteristics of those who tolerate the procedure to help determine if there is a subgroup of patients in whom the risks outweigh the potential benefits. Patients and Methods Study Design We performed a retrospective cohort study of consecutive outpatients who underwent VATS lung biopsy for diagnosis of an ILD between January of 1998 and July of 2004 at The Hospital of the University of Pennsylvania in Philadelphia, Pennsylvania. During this study period, no open lung biopsy procedures were performed on outpatients referred for diagnosis of an ILD. The study protocol 2007 by The Society of Thoracic Surgeons 0003-4975/07/$32.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2006.10.002

Ann Thorac Surg KREIDER ET AL 2007;83:1140 5 COMPLICATIONS OF VATS LUNG BIOPSY FOR ILD 1141 Table 1. Study Population (n 68) Characteristic % (n) or (range) Age (mean, years) 58 (38 84) Gender Male 44% (30) Female 56% (38) Smoking status Never 43% (27) Former 50% (34) Current 7% (5) Pre-op PFTs (mean % predicted) FVC 66.9% (18 110) FEV 1 70.1% (37 108) TLC 72.7% (32 106) Dlco 50.9% (19 113) Pre-op O 2 dependence 33% (22) Pre-op pulmonary HTN (18 tested) 56% (10) Subsequent pathologic diagnoses Usual IP 34% (23) Sarcoidosis 9% (6) Honeycomb lung 9% (6) Chronic hypersensitivity 7.5% (5) Nonspecific IP 6% (4) Desquamative IP 4.5% (3) Respiratory bronchiolitis-ild Normal lung 3% (2) Acute IP 1.5% (1) Bronchiolitis 1.5% (1) Nonclassifiable 23.5% (16) Dlco diffusing capacity of the lung for carbon monoxide; FEV 1 forced expiratory volume in 1 second; FVC forced vital capacity; HTN hypertension; ILD interstitial lung disease; IP interstitial pneumonia; PFTs pulmonary function testing; TLC total lung capacity. was reviewed and approved by the Institutional Review Board at the University of Pennsylvania in December 2004 and 2005, and an exemption for the requirement for informed consent was granted. In addition, we performed a literature review and meta-analysis separately from our retrospective patient cohort. We searched Medline for English language articles published from 1995 through 2005 using the MESH search terms biopsy or video-assisted thoracic surgery and either lung diseases, or interstitial or pulmonary fibrosis. From 869 articles retrieved in this initial screen, two authors (MEK and JHF) identified original research articles on cohorts of more than 10 adults who underwent surgical lung biopsy for diagnosis of a subacute or chronic, diffuse ILD. The references cited in these articles were reviewed to find additional studies. Twenty-two articles were identified and abstracted to calculate a composite mortality estimate with a 95% confidence interval (CI). Outcome Definitions The primary outcome for the cohort study was 60-day postoperative mortality from any cause. Secondary outcomes included prolonged length of stay (defined as 5 days), readmission within 30 days of surgery, pneumonia, prolonged air leak (chest tube in place for 5 days), and any mechanical ventilation postoperatively. Finally, a composite outcome was created, termed morbidity, which was defined as any one or more of the primary or secondary outcome measures. Subjects were only counted once if they met more than one outcome. This composite outcome was used for risk factor analysis to increase the power to detect associations. Candidate Risk Factors Information on candidate risk factors abstracted from all subjects charts included preoperative pulmonary function test results, a preoperative prescription for longterm oxygen therapy, pathologic diagnosis, presence or absence of pulmonary hypertension, which was defined as pulmonary artery systolic pressure (PASP) of 40 mm Hg or more as estimated by echocardiogram or measured on right heart catheterization; race, gender, age at the time of biopsy, current pharmacologic therapy for ILD, and smoking status. Analysis The incidence of all outcomes was calculated along with their 95% CIs. To identify potential risk factors for morbidity, we performed univariate analysis with contingency tables for discrete variables and logistic regression for continuous data. All risk factors with p 0.2 in univariate analysis were then included in a multivariate logistic regression. Colinearity was assessed for, and if found, the less predictive of the two risk factors was removed from the final model. Results Study Population The cohort of 68 patients included in this report is described in Table 1. The subjects had an average age of 58 years (range, 38 to 84 years), 56% were women, and 73% were white. Fifty percent were former smokers. Most had mild-to-moderate pulmonary functional impairment as determined by pulmonary function testing. Thirteen patients (19%) were taking oral corticosteroids, another 13 (19%) were taking inhaled corticosteroids at the time of biopsy, and 1 patient was taking methotrexate preop- Table 2. Incidence of Outcomes Outcome Incidence (n) 95% CI 1. Mortality ( 60 days) 4.4% (3) 1 12% 2. Prolonged air leak ( 5 days) 4.4% (3) 1 12% 3. Post-op pneumonia 2.9% (2) 0.4 10% 4. Hospital readmission ( 1 month) 10.7% (7) 4 21% 5. Prolonged hospital stay ( 5 days) 4.4% (3) 1 12% 6. Mechanical ventilation post-op 5.9% (4) 2 14% Composite-any of the outcomes 1 6 19.1% (13) 11 31% CI confidence interval.

1142 KREIDER ET AL Ann Thorac Surg COMPLICATIONS OF VATS LUNG BIOPSY FOR ILD 2007;83:1140 5 Table 3. Clinical Characteristics of the Subjects Who Died Patient FVC % DLCO % Supplemental O2? PASP Path dx Intubated on POD # Died on POD # 58 y.o. WM 53 30 Yes at 4 LPM N/A UIP 4 43 54 y.o. WF 43 19 Yes at 3 LPM 45 UIP 8 20 71 y.o. WM 67 27 Yes at 2 LPM 46 UIP 2 4 Dlco diffusing capacity of the lung for carbon monoxide; FVC forced vital capacity; LPM liters per minute; PASP pulmonary artery systolic pressure; Path dx pathological diagnosis; POD post operative day; UIP usal interstitial pneumonitis; WF white female; WM white male. eratively. These subjects were considered to have failed empiric therapy and required a diagnosis for further therapeutic intervention. All subjects had undergone preoperative computed tomography (CT) scans that were not considered diagnostic of a specific ILD. No patients in the cohort had human immunodeficiency virus infection or active cancer. Adverse Outcomes Surgical morbidity and mortality are presented in Table 2. Three deaths occurred within 60 days after biopsy for an incidence of 4.4% (95% CI, 1% to 12%). The most frequent major complications were a need for hospital readmission in 10.7% (95% CI, 4% to 21%) and mechanical ventilation postoperatively in 5.9% (95% CI, 2% to 14%). The reasons for readmission included pneumothorax in 4 patients, hematoma at incision site in 1, and respiratory failure in 2. Altogether, 19.1% of patients experienced one or more of the recorded complications (95% CI, 11% to 31%). Description of Mortality The clinical characteristics of the 3 patients who died after VATS lung biopsy are summarized in Table 3. All three were extubated immediately after surgery, but then progressive hypoxemia developed that required reintubation 1 to 8 days after biopsy. Histologic examination of the VATS lung biopsy specimens from these 3 patients revealed usual interstitial pneumonitis by standard criteria [2]. Shortly after biopsy, new ground-glass opacities were observed on chest CT in all 3 patients. Bronchoscopy revealed no evidence of acute infection. Congestive heart failure and pulmonary embolism were excluded by echocardiogram and CT angiogram. All 3 patients were treated with intravenous antibiotics and glucocorticoids. Table 4. Multivariate Analysis of Risk Factors for Major Morbidity Risk factor OR (95% CI) p Value O2 therapy pre-op 20.9 (1.67 261.6) 0.02 TLC % pred 0.82 (0.68 1.00) 0.05 DLCO % pred 1.00 (0.93 1.08) 0.92 UIP on biopsy 1.90 (0.21 17.45) 0.51 Age biopsy 1.01 (0.90 1.12) 0.91 CI confidence interval; Dlco diffusing capacity of the lung for carbon monoxide; OR odds ratio; TLC total lung capacity; UIP usual interstitial pneumonitis. No clinical improvement occurred, and the patients died at 4, 20, and 43 days after biopsy. Postmortem examinations were performed on two patients. Both had findings of diffuse alveolar damage on a background of usual interstitial pneumonitis, consistent with an acute exacerbation of usual interstitial pneumonitis [3]. There were too few deaths to compare characteristics of survivors and nonsurvivors. Risk Factor Analysis The results of multivariate risk factor analysis are presented in Table 4. Patients who required oxygen therapy preoperatively had an elevated odds ratio (OR) of developing major morbidity of 20.9 compared with those without oxygen (95% CI, 1.67 to 261.6, p 0.02). There was also a suggestion that people with more severely impaired percentage of total lung capacity (TLC%) predicted had higher odds of poor outcome (OR, 0.82; 95% CI, 0.68 to 1.00; p 0.05). In our cohort, only 18 patients underwent testing for pulmonary hypertension, either by echocardiogram in 17 or right heart catheterization in 1. Of these, 8 had a PASP of greater than 40, and 4 patients within this subgroup had a complication for a 50% morbidity incidence (95% CI, 15% to 85%) compared with no morbidity in those without pulmonary hypertension (p 0.01 comparing proportion). Two of the three patients who died had documented pulmonary hypertension; the third was not studied. Comment Our study demonstrates a 60-day mortality rate of 4.4% and a morbidity rate of 19% in 68 consecutive outpatients undergoing VATS lung biopsy for diagnosis of an ILD at one university hospital. We find further that preoperative oxygen use, lower TLC%, and a PASP greater than 40 were statistically associated with the development of morbidity. Elevated pulmonary artery pressure was present in two thirds of those who died, but the lack of this having been measured preoperatively in all patients prevents us from drawing a firm conclusion about the relationship between pulmonary artery pressures and postoperative mortality. One recently published study is comparable to ours in research design. Lettieri and colleagues [4] retrospectively examined a cohort of 83 patients who underwent VATS lung biopsy at Walter Reed Army Medical Center for diagnosis of ILD. In contrast to our cohort, theirs

Ann Thorac Surg KREIDER ET AL 2007;83:1140 5 COMPLICATIONS OF VATS LUNG BIOPSY FOR ILD 1143 included 8 patients (9.6%) who required mechanical ventilation for respiratory failure immediately before biopsy. They reported mortality of 4.8% at 30 days and 6.0% at 90 days for their study patients. Overall, 12 patients (14%) experienced a major complication of surgery or died within 90 days. Notably, 3 of the 8 patients who required mechanical ventilation before biopsy died within 90 days after biopsy. Excluding these patients, the postoperative death rate for the remaining 75 patients was 2.7%. It may be possible to limit the risk of complications associated with surgical lung biopsy for the diagnosis of ILD by identification and careful consideration of preoperative risk factors for adverse surgical outcomes. In our cohort of 68 patients, we found that preoperative dependence on long-term oxygen therapy, lower TLC, and pulmonary hypertension as measured by echocardiogram predicted an adverse outcome. Carrillo and colleagues [9] identified multiple comorbidities, a greater score on the ATS dyspnea scale, a higher respiratory rate, a lower Pao 2, higher Paco 2, and a Paco 2 /Pao 2 index of 0.72 or higher as risk factors for mortality in their case control analysis. Preoperative ventilator dependence and an immunocompromised status were the only predictors of mortality in the Lettieri and colleagues analysis [4]. In the review of the experience at the Mayo Clinic with lung biopsy for usual interstitial pneumonitis, Utz and colleagues [5] found that a diffusing capacity of the lung for carbon monoxide of less than 35% predicted and idiopathic usual interstitial pneumonitis (versus usual interstitial pneumonitis associated with collagen vascular disease) were predictors of mortality. All in all, ILD patients with more severe lung disease appear to be at greater risk for complications of surgical lung biopsy. Our findings emphasize of the significance of preoperative pulmonary hypertension. Efforts to reduce the mortality associated with the diagnosis of ILD by surgical lung biopsy may also benefit from a more complete understanding of the causes of death in these patients. Within 8 days after biopsy, the 3 patients who died postoperatively in our series required reintubation for respiratory distress and new widespread pulmonary infiltrates that could not be attributed to known causes of acute lung injury such as aspiration or sepsis. All were found to have usual interstitial pneumonitis on the surgical lung biopsy specimen. A postmortem examination was performed on 2 of the 3 patients and found extensive diffuse alveolar damage that was not apparent at the time of the surgical lung biopsy. These findings suggest that VATS lung biopsy or some unrecognized perioperative event that may occur in association with VATS lung biopsy triggered a fatal episode of acute lung injury in 3 of our patients who had underlying usual interstitial pneumonitis. Several published studies have drawn attention recently to unexpected, apparently irreversible episodes of accelerated decline in lung function experienced by patients with idiopathic pulmonary fibrosis (IPF) [3, 6, 7]. These episodes are characterized by development of widespread ground-glass opacities on chest CT images and by the postmortem finding of diffuse alveolar damage superimposed on usual interstitial pneumonitis [6]. Because many such episodes have occurred without an identifiable precipitating cause, the phenomenon has been attributed to an acute exacerbation of IPF. The 3 patients who died in our study cohort, the 4 who died in the study by Tiitto and colleagues [8], and 2 patients described by Kim and colleagues [6] all experienced an acute, fatal deterioration in lung function that was similar clinically, radiographically, and pathologically to that reported for acute exacerbations of IPF. Whether surgical lung biopsy sometimes triggers an acute exacerbation of underlying IPF or sometimes precipitates an etiologically distinct form of acute lung injury superimposed on preexisting IPF is unknown and remains to be determined. There are several limitations to our cohort study. Because our analysis was performed retrospectively, every subject did not receive the same evaluation before biopsy, and thus, some risk factor data are missing. This was especially true for pulmonary hypertension, where only a fraction of subjects were screened. In addition, the screening was largely done with echocardiography, which may be a relatively inaccurate measure of pulmonary hypertension. Nonetheless, the observation that those with a suggestion of pulmonary hypertension did far more poorly than those documented not to have pulmonary hypertension is of significant concern and should be strongly considered in those being evaluated for lung biopsy. Another potential problem of a retrospective cohort study is loss to follow-up. Only 1 subject in our cohort was lost for mortality information, and this was completed with the Social Security death index; therefore, our data on short-term outcomes were fairly complete. However, other longer-term outcomes such as persistent worsening lung function postoperatively could not be examined because they were not routinely monitored. In addition, the study was performed on patients who underwent surgery performed by a four-physician academic thoracic surgery practice and thus may not be representative of the patient populations and clinical practices of other centers. The number of subjects examined limited our ability to examine risk factors associated with poor outcomes. To place our results in the context of previously reported studies of complications of lung biopsy for ILD, we performed a literature search and meta-analysis. Using the search strategy described in the Methods section, we identified 22 studies published in English during the past 10 years in addition to ours that reported postoperative mortality after surgical lung biopsy for the diagnosis of an ILD [4, 5, 8 29]. A composite postoperative mortality of rate 4.5% (95% CI, 3.7% to 5.5%) was calculated for the 2223 patients included in all 22 previously published studies plus our own. Studies published from 2001 through 2005 reported rates of mortality that were not significantly different from those published between 1995 and 2000. A significantly higher mortality

1144 KREIDER ET AL Ann Thorac Surg COMPLICATIONS OF VATS LUNG BIOPSY FOR ILD 2007;83:1140 5 rate of 47.3% was seen in patients requiring preoperative ventilation compared with 2.2% in those free from ventilation (p 0.001). Despite the clear limitations of even systematic literature reviews like this, several observations from this study warrant consideration. The review confirms our finding that morbidity and mortality associated with surgical lung biopsy for diagnosis of ILD disease is certainly not nil. This may be due to progression of significant disease in an already ill population, or alternatively, it is possible that the surgery itself, or associated procedures such as general anesthesia and positive pressure ventilation, can result in an accelerated decline. Importantly, surgical lung biopsy clearly may trigger an exacerbation of IPF or a closely related phenomenon of unexplained acute lung injury in some patients who have this disease. In conclusion, the potential benefits of diagnostic surgical lung biopsy must be considered against the risks of the procedure as detailed in this report, especially for patients with more severe cardiovascular-pulmonary functional impairment (specifically, preoperative oxygen use and lower TLC). Consideration should also be given to screening potential surgical subjects for the presence of pulmonary hypertension, as patients with significantly elevated PASP appear also to be at considerably higher risk. The risks of biopsy do not appear to be reduced by a VATS approach. Rapid pulmonary deterioration after biopsy in most cases is due to acute lung injury and not infection. Awareness of the complications of surgical lung biopsy should stimulate interest in developing alternatives to surgical lung biopsy for differentiating among the ILDs. References 1. American Thoracic Society; European Respiratory Society. American Thoracic Society/European Respiratory Society International Multidisciplinary Consensus Classification of the Idiopathic Interstitial Pneumonias. Am J Respir Crit Care Med 2002;165:277 304. 2. Katzenstein AL, Myers JL. Idiopathic pulmonary fibrosis: clinical relevance of pathologic classification. Am J Respir Crit Care Med 1998;157:1301 15. 3. Ambrosini V, Cancellieri A, Chilosi M, et al. Acute exacerbation of idiopathic pulmonary fibrosis: report of a series. Eur Respir J 2003;22:821 6. 4. Lettieri CJ, Veerappan GR, Helman DL, Mulligan CR, Shorr AF. Outcomes and safety of surgical lung biopsy for interstitial lung disease. Chest 2005;127:1600 5. 5. Utz JP, Ryu JH, Douglas WW, et al. High short-term mortality following lung biopsy for usual interstitial pneumonia. Eur Respir J 2001;17:175 9. 6. Kim DS, Park JH, Park BK, Lee JS, Nicholson AG, Colby T. Acute exacerbation of idiopathic pulmonary fibrosis: frequency and clinical features. Eur Respir J 2006;27:143 50. 7. Martinez FJ, Safrin S, Weycker D, et al. The clinical course of patients with idiopathic pulmonary fibrosis. Ann Intern Med 2005;142:963 7. 8. Tiitto L, Heiskanen U, Bloigu R, et al. Thoracoscopic lung biopsy is a safe procedure in diagnosing usual interstitial pneumonia. Chest 2005;128:2375 80. 9. Carrillo G, Estrada A, Pedroza J, et al. Preoperative risk factors associated with mortality in lung biopsy patients with interstitial lung disease. J Invest Surg 2005;18:39 45. 10. Ooi A, Iyenger S, Ferguson J, Ritchie AJ. VATS lung biopsy in suspected, diffuse interstitial lung disease provides diagnosis, and alters management strategies. Heart Lung Circ 2005;14:90 2. 11. Lee YC, Wu CT, Hsu HH, Huang PM, Chang YL. Surgical lung biopsy for diffuse pulmonary disease: experience of 196 patients. J Thorac Cardiovasc Surg 2005;129:984 90. 12. Yamaguchi M, Yoshino I, Suemitsu R, et al. Elective videoassisted thoracoscopic lung biopsy for interstitial lung disease. Asian Cardiovasc Thorac Ann 2004;12:65 8. 13. Ayed AK. Video-assisted thoracoscopic lung biopsy in the diagnosis of diffuse interstitial lung disease. A prospective study. J Cardiovasc Surg (Torino) 2003;44:115 8. 14. Ayed AK, Raghunathan R. Thoracoscopy versus open lung biopsy in the diagnosis of interstitial lung disease: a randomised controlled trial. J R Coll Surg Edinb 2000;45:159 63. 15. Chang AC, Yee J, Orringer MB, Iannettoni MD. Diagnostic thoracoscopic lung biopsy: an outpatient experience. Ann Thorac Surg 2002;74:1942 6; discussion 1946 7. 16. Riha RL, Duhig EE, Clarke BE, Steele RH, Slaughter RE, Zimmerman PV. Survival of patients with biopsy-proven usual interstitial pneumonia and nonspecific interstitial pneumonia. Eur Respir J 2002;19:1114 8. 17. Nicholson AG, Fulford LG, Colby TV, du Bois RM, Hansell DM, Wells AU. The relationship between individual histologic features and disease progression in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 2002;166:173 7. 18. Qureshi RA, Soorae AA. Efficacy of thoracoscopic lung biopsy in interstitial lung diseases: comparison with open lung biopsy. J Coll Physicians Surg Pak 2003;13:600 3. 19. Hunninghake GW, Zimmerman MB, Schwartz DA, et al. Utility of a lung biopsy for the diagnosis of idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 2001;164: 193 6. 20. Blewett CJ, Bennett WF, Miller JD, Urschel JD. Open lung biopsy as an outpatient procedure. Ann Thorac Surg 2001; 71:1113 5. 21. Miller JD, Urschel JD, Cox G, et al. A randomized, controlled trial comparing thoracoscopy and limited thoracotomy for lung biopsy in interstitial lung disease. Ann Thorac Surg 2000;70:1647 50. 22. Temes RT, Joste NE, Qualls CR, et al. Lung biopsy: is it necessary? J Thorac Cardiovasc Surg 1999;118:1097 100. 23. Rena O, Casadio C, Leo F, et al. Videothoracoscopic lung biopsy in the diagnosis of interstitial lung disease. Eur J Cardiothorac Surg 1999;16:624 7. 24. Neuhaus SJ, Matar KS. The efficacy of open lung biopsy. Aust N Z J Surg 1997;67:181 4. 25. Mouroux J, Clary-Meinesz C, Padovani B, et al. Efficacy and safety of videothoracoscopic lung biopsy in the diagnosis of interstitial lung disease. Eur J Cardiothorac Surg 1997;11: 22 4, 25 6. 26. Zegdi R, Azorin J, Tremblay B, Destable MD, Lajos PS, Valeyre D. Videothoracoscopic lung biopsy in diffuse infiltrative lung diseases: a 5-year surgical experience. Ann Thorac Surg 1998;66:1170 3. 27. Ravini M, Ferraro G, Barbieri B, Colombo P, Rizzato G. Changing strategies of lung biopsies in diffuse lung diseases: the impact of video-assisted thoracoscopy. Eur Respir J 1998;11:99 103. 28. Deshmukh SP, Krasna MJ, McLaughlin JS. Video assisted thoracoscopic biopsy for interstitial lung disease. Int Surg 1996;81:330 2. 29. Krasna MJ, White CS, Aisner SC, Templeton PA, McLaughlin JS. The role of thoracoscopy in the diagnosis of interstitial lung disease. Ann Thorac Surg 1995;59:348 51.