CT-guided needle biopsy of lung lesions is a. Management of Pneumothorax After Percutaneous CT-Guided Lung Biopsy*

Transcription

1 Management of Pneumothorax After Percutaneous CT-Guided Lung Biopsy* Takuji Yamagami, MD, PhD; Toshiyuki Nakamura, MD; Shigeharu Iida, MD; Takeharu Kato, MD; and Tsunehiko Nishimura, MD, PhD Objectives: To evaluate the efficacy of simple aspiration of air from the pleural space to prevent increased pneumothorax and to avoid chest tube placement in cases of pneumothorax following CT-guided lung biopsy. Design: Observational. Materials and methods: One hundred thirty-four consecutive percutaneous needle lung biopsies using real-time CT fluoroscopy guidance formed the basis of our study. All patients that demonstrated moderate or severe pneumothorax on postbiopsy chest CT images underwent percutaneous manual aspiration regardless of symptoms while on the CT scanner table. Correlation between the incidence of pneumothorax after biopsy and many factors (ie, gender, age, number of pleural passes, presence of emphysema, lesion size, and lesion depth) were determined, and management of each case of biopsy-induced pneumothorax was reviewed. Results: Postbiopsy pneumothorax occurred in 46 of 134 procedures (34.3%). Twenty of the 46 patients were treated by manual aspiration, while 26 patients were simply observed. In 43 of the 46 pneumothoraces (93.5%), the pneumothorax resolved completely on follow-up chest radiographs without requiring tube placement. Only three patients (2.2% of the entire series; 6.5% of those who had pneumothorax develop) required chest tube placement. The risk of pneumothorax significantly increased with lesion size and depth. Conclusion: Results of our nonprospective, nonrandomized study suggest that percutaneous manual aspiration of biopsy-induced pneumothorax performed immediately after biopsy may prevent progressive pneumothorax and subsequent chest tube placement. (CHEST 2002; 121: ) Key words: biopsies; complications; lung biopsy; pneumothorax CT-guided needle biopsy of lung lesions is a well-established and safe technique for diagnosing lung nodules. 1 Most CT-guided lung biopsies in earlier reports were performed by fine-needle aspiration and were useful in differentiating malignant from benign lesions. 2 4 More recently, the use of tissue core biopsy utilizing an automated cutting needle has been implemented in an attempt to further enhance diagnostic accuracy. 4,5 Modalities commonly employed for imaging-guided percutaneous needle biopsy include fluoroscopy, 6 conventional CT, 1,7 and helical CT, 7 which has become more widely used. CT fluoroscopy, which was developed most recently, has simplified the process and decreased the time requirements of CT-guided needle *From the Department of Radiology, Kyoto Prefectural University of Medicine, Kyoto, Japan. Manuscript received May 31, 2001; revision accepted October 2, Correspondence to: Takuji Yamagami, MD, PhD, Department of Radiology, Kyoto Prefectural University of Medicine, 465 Kajii, Kawaramachi-Hirokoji, Kamigyo, Kyoto, , Japan; yamagami@koto.kpu-m.ac.jp biopsies. 8 In accordance with the increased diagnostic accuracy and the progress in developing better imaging modalities, attempts have been made at For editorial comment see page 1013 performing needle lung biopsies in outpatient clinics. 9,10 To further the acceptance of this procedure, solutions to the complications inherent in its performance, the commonest being pneumothorax, are necessary. 9,10 The problem most responsible for complicating outpatient management is not the presence of the pneumothorax per se, but an increase in the size of the pneumothorax that requires chest tube placement. 9,10 In 1996, Yankelevitz et al 11 reported success with manual aspiration at the time of biopsy for large pneumothoraces, thus avoiding chest tube placement in 70% of such cases. This led us to wonder whether more widespread use of this simple procedure might obviate the need for chest tube placement in the great majority of these patients. CHEST / 121 / 4/ APRIL,

2 Hence, in our institution, manual percutaneous aspiration of air from the pleural space has been performed while the patient remains on the CT scanner table. This is done in all biopsy-induced pneumothoraces, with the exception of small pneumothoraces, in an effort to prevent their progression. The goal of the present study was to determine whether immediate, simple aspiration can reduce the incidence of progressive, biopsy-induced pneumothorax and the subsequent need for chest tube placement. Subjects Materials and Methods One hundred thirty-four consecutive percutaneous needle biopsies of the lung under real-time CT fluoroscopy guidance were performed in our institution between October 1997 and March 2001 in 127 patients (53 women and 74 men; mean age, 67.9 years; age range, 16 to 92 years; 7 patients underwent repeat biopsies). Of the 134 lung nodules sampled, 122 were solid nodular lesions, 7 were cavitary nodules, and 5 were focal consolidations. The mean lesion diameter was 23.5 mm (range, 3 to 100 mm). Both fine-needle aspiration and core biopsy were performed in 111 patients, 21 core biopsies were performed alone, and 2 patients underwent fine-needle aspiration alone. A 21-gauge Sonopsy needle (Hakko; Nagano, Japan) was used for all 113 fine-needle aspirations, and a 20-gauge Auto Surecut needle (Create Medics; Yokohama; Kanagawa, Japan) was used in 115 of the 132 core biopsies. For the other 17 core biopsies, either an 18- or 20-gauge Monopty needle (Bard; Covington, GA), an 18-gauge ASAP needle (Boston Scientific Japan; Tokyo, Japan), or a 20-gauge Temno needle (Bauer Medical; Clearwater, FL) was used. Biopsy Procedures All patients had undergone diagnostic CT of the chest with 10-mm-thick contiguous axial tomographic sections before the time of biopsy. At the time of biopsy, preliminary helical CT images were obtained in 5-mm-thick sections through the lesion. From a review of these preliminary images, patient position, level of the entry needle site, and direction of approach for biopsy were planned to provide the most direct route for biopsy, to traverse the least amount of aerated lung, and to avoid bullae and fissures. The positions of patients during biopsy were supine (n 58), prone (n 58), or lateral (n 18). The CT unit used in this study was the X Vigor Laudator (Toshiba Medical Systems; Tokyo, Japan). The procedure was performed by one of three interventional radiologists experienced in CT-guided biopsy after obtaining informed consent from the patient. A CT fluoroscopic imaging system was used for all CT-guided biopsy procedures. Details of CT fluoroscopy were described elsewhere. 12 Imaging parameters during CT fluoroscopy included a CT beam width collimated to 3 mm, tube voltage of 120 kilovolt peak, current of 30 to 50 ma, and a scanning speed of 0.75 s per rotation (360 ). An operator wearing a protective lead apron in the CT room was responsible for control of CT fluoroscopic exposure via a foot pedal and assisted in table movement, gantry tilt, and directed the laser light beam via the control panel. The control panel was covered with a sterile transparent drape when a single operator performed the procedure. Alternatively, an assistant would help by adjusting the controls. Intermittent real-time CT fluoroscopic technique 8 was preferred while advancing the biopsy needle, of which the center was held by surgical forceps, with the distal edge on the operator s side held by the operator s hand as necessary. This technique was performed in a stepwise manner with quick application of CT fluoroscopy to confirm the path of the needle while meticulous care was taken to minimize direct radiation to the operator s hands. Patients who could not cooperate with breath holding underwent the procedure during usual respiration. After confirming that the needle tip had reached the lesion, a specimen was obtained, and the needle was withdrawn. When the obtained specimen was insufficient, a repeat biopsy was performed. Treatment for Pneumothorax Approximately 5 min after the biopsy procedure, or immediately thereafter in patients with rapidly developing pneumothoraces that could already be seen on CT fluoroscopy images during the biopsy procedure, axial images were obtained during a single breath hold through the level of the biopsy or, if necessary, through the whole chest using helical CT scanning to evaluate for the presence of a pneumothorax. The beam width was 10 mm, table-feeding speed was 10 mm/s, and image reconstruction was performed with a width of 10 mm during postbiopsy CT. Patients with a moderate or severe pneumothorax (defined in principle as involving more than seven slices on postbiopsy CT), irrespective of the presence of symptoms, underwent immediate manual aspiration while on the CT scanner table. An 18-gauge IV catheter held by surgical forceps was inserted near the initial biopsy site or at the site of the largest aspect of the pneumothorax on contiguous transverse tomographic sections and advanced into the pleural space. After the tip was confirmed by real-time CT fluoroscopy to have precisely entered the pleural space, the stylet was removed. Using an IV extension tubing and three-way stopcock, the IV catheter was connected to a 50-mL syringe. As the valve of the three-way stopcock was alternately opened and closed, air from the pneumothorax was drawn into the syringe and expelled. Whenever resistance was encountered, CT fluoroscopy was briefly performed to confirm whether a pneumothorax remained. If so, the IV catheter was withdrawn slightly and aspiration was resumed. Finally, helical CT images were obtained through the area to confirm re-expansion of the lung. All biopsy and aspiration procedures were performed under local anesthesia. All biopsy patients who had pneumothorax develop were transferred to the recovery room where oxygen was administered via nasal cannula (100% at 3 L/min). Follow-up A follow-up chest radiograph was obtained 3 to 4 h after the biopsy procedures. Patients without a pneumothorax on this radiograph were discharged; patients with small, asymptomatic pneumothoraces were discharged with scheduled clinic visits the next day. The day after the biopsy, a chest radiograph was obtained in all patients with a pneumothorax or with symptoms and was then repeated every 1 to 2 days until the pneumothorax disappeared. A rapidly enlarging or symptomatic pneumothorax was treated with placement of a chest tube. Investigated Parameters The relationship between the incidence of a pneumothorax and various variables were investigated. Gender, age, number of pleural passes, presence of emphysema as revealed on CT 1160 Clinical Investigations

3 according to the criteria of Sanders et al, 13 size of the lesion, and lesion depth measured from the surface of the pleura to the edge of the lesion were each analyzed. Moreover, for procedures complicated by pneumothorax, any additional intervention used in each patient was evaluated. Statistical analysis For statistical analysis, commercial software (Statview; Abacus Concepts; Berkeley, CA) was used. Quantitative variables were compared using an unpaired t test. Qualitative variables were compared using the 2 test. Results Risk Factors for Pneumothorax Among 134 CT-guided biopsy procedures, pneumothorax was detected in 45 patients using CT performed immediately after the biopsies and was revealed in another patient on chest radiography performed the next day. Of the 46 patients with pneumothorax (age range, 49 to 89 years; mean SD age, years), 19 patients were women (41.3%) and 18 patients had emphysema (39.1%). There were pleural passes. The size and depth of the lesion ranged from 3 to 45 mm ( mm) and from 0 to 79 mm ( mm), respectively. However, of the 88 patients without pneumothorax (age range, 16 to 92 years; years) 36 patients (40.9%) were women and 22 patients had emphysema (25.0%). In this group, there were pleural passes. The size and depth of the lesion ranged from 6 to 100 mm ( mm) and from 0 to 63 mm ( mm), respectively. Size and depth of the lesion correlated significantly with the frequency of pneumothorax (size, p , unpaired t test; depth, p , unpaired t test), while other variables had no statistical association with frequency. Specifically, values for gender (p , 2 test), age (p , unpaired t test), number of pleural passes (p , unpaired t test), and presence of emphysema (p , 2 test) did not indicate significance. Management for Each Patient With Pneumothorax Data on all 46 procedures complicated by pneumothorax are shown in Tables 1, 2. Pneumothorax was detected on more than seven slices by CT Table 1 Characteristics of 26 Biopsies Complicated by Pneumothorax Managed by Observation Alone* Procedure No./ Gender/Age, yr No. of Pleural Passes Emphysema Size of Lesion, mm Depth from Pleura, mm No. of Slices With Pneumothorax on Chest CT Duration Until Pneumothorax Disappearance Confirmed 1/f/77 2 Absent d 2/m/71 3 Absent d 3/f/65 4 Absent d 4/m/72 2 Absent d 5/m/57 2 Present d 6/m/70 2 Present d 7/f/86 3 Absent d 8/f/66 2 Absent d 9/m/86 2 Present d 10/f/67 4 Present d 11/m/89 1 Present d 12/m/89 2 Present d 13/f/53 4 Present d 14/f/73 2 Present d 15/f/71 4 Absent d 16/m/79 3 Present d 17/m/59 1 Present d 18/m/75 2 Absent h 19/f/49 3 Present d 20/m/59 2 Absent d 21/m/68 2 Absent d 22/f/69 2 Absent d 23/f/77 2 Absent d 24/m/63 2 Absent d 25/f/54 2 Absent h 26/f/73 3 Absent d *f female; m male. Tube placement was performed for 6 days starting the 15th day after biopsy. Delayed pneumothorax occurring for 4 days starting the day after biopsy. CHEST / 121 / 4/ APRIL,

4 Table 2 Management of 20 Biopsies Complicated by Pneumothorax With Immediate Aspiration* Procedure No./ Gender/Age, yr No. of Pleural Passes Emphysema Size of Lesion, mm Depth From Pleura, mm No. of Slices With Pneumothorax on Chest CT Volume of Aspirated Air, ml Duration Until Pneumothorax Disappearance Confirmed 27/m/79 2 Absent d 28/m/52 2 Absent d 29/m/64 2 Present d 30/f/77 3 Present d 31/m/78 3 Absent d 32/m/65 2 Absent d 33/m/74 3 Absent d 34/m/75 1 Absent d 35/m/69 3 Absent d 36/m/68 2 Present d 37/f/57 2 Absent d 38/m/87 2 Absent d 39/m/68 2 Present d 40/m/70 2 Absent d 41/f/66 2 Absent d 42/m/66 3 Absent d 43/f/56 2 Present h 44/m/73 4 Present d 45/f/58 3 Absent d 46/f/66 3 Present h *See Table 1 for expansion of abbreviations. Tube placement was performed for 4 days starting the day after biopsy. Tube placement was performed for 6 days starting the third day after biopsy. obtained just after biopsy in 22 cases. In two of these cases (procedures 1 and 2), an inconsequential area was occupied by the pneumothorax on each slice and, although involving many slices, aspiration was not performed. The remaining 20 of these 22 patients underwent manual aspiration of the pneumothorax immediately after the biopsy (Fig 1). The amount of air aspirated ranged from 20 to 720 ml (mean, 280 ml). Forty-three of the 46 pneumothoraces (93.5%) resolved without chest tube insertion during the course of observation. In four cases of pneumothorax demonstrated on CT done immediately after biopsy, there was no pneumothorax on chest radiography obtained 3 h after the procedure (procedures 18, 25, 43, and 46). However, tube placement was required after three procedures, two aspirated patients (procedures 30 and 36), and one patient in which no aspiration was performed at the time of biopsy but in which tube placement was required 15 days after biopsy for an enlarging, symptomatic pneumothorax noted on follow-up chest radiograph (procedure 16). The average interval from biopsy until confirmed disappearance of the pneumothorax was days (range, 0 to 9 days; median, 3 days) in cases with aspiration and days (range, 0 to 22 days; median, 2 days) in cases without aspiration. There was no statistically significant difference between these average intervals (unpaired t test, p ). Discussion Pneumothorax is the most common complication of CT-guided lung biopsy, with a frequency of 17.9% 7 to 54.3% 4 according to reports published in the past 10 years that have included 100 procedures. 4,7,14 18 Many researchers have investigated risk factors for this problem ,19,20 Which factors statistically correlate with the frequency of pneumothorax remain controversial, but most reports have suggested that lesion size 14,15 and depth 14,16,20 are the main factors influencing the incidence of pneumothorax after CT-guided needle biopsy. On the contrary, gender, age, and the number of pleural passes have not been shown to correlate with the incidence of pneumothorax The frequency and risk factors associated with pneumothorax in the present study were similar to previous reports, suggesting that the subjects and the operator skill in our CT-guided lung biopsies did not differ appreciably from those of earlier studies. Some researchers 15,19 have reported a significant relationship between emphysema and pneumothorax after biopsy. In the present study, emphysema was detected on CT in 18 of 46 procedures (39.1%) complicated by pneumothorax, while 22 of 88 procedures (25.0%) were not complicated by pneumothorax. Therefore, although not statistically significant by 2 test (p ), this may suggest a trend toward significance, and the 1162 Clinical Investigations

5 Figure 1. Top: Real-time CT fluoroscopic scan with patient in the left-side-down position shows the tip of a 20-gauge needle penetrate the right lower lobe nodule. Note that the pneumothorax developed during this procedure. Center: Real-time CT operator should be cautioned about this tendency for complications by pneumothorax after needle biopsy of the lung. The majority of cases of pneumothorax developing after needle biopsy resolved spontaneously, but a few intractable cases required chest tube placement. In most cases, patients treated with chest tubes typically have been admitted to the hospital. 9,10 Recently, however, CT-guided lung biopsy has been performed in the outpatient setting as a way of minimizing costs and inconvenience to the patient. 9,10 Thus, efforts are required to decrease the necessity of chest tube placement for pneumothoraces that result from complications incurred from the biopsy. 9,10 A review of all studies published in the last 10 years that involve 100 procedures reveals a frequency of chest tube placement ranging from 2.0% 16 to 15.0% 4 of all biopsy procedures, and 6.7% 17 to 33.3% 14 of cases complicated by pneumothorax. 4,7,14 18 In comparison, our frequency was at the low end, with only 2.2% (3 of 134) and 6.5% (3 of 46) of biopsies and pneumothoraces requiring chest tubes, respectively. Although this was a nonprospective, nonrandomized study, it appears, at least from the results of our study, that the immediate aspiration of a pneumothorax, irrespective of the presence of symptoms, caused by CT-guided needle lung biopsy is effective at preventing progression of the pneumothorax, thus avoiding chest tube placement. The rapid aspiration of a pneumothorax was initially reported by Yankelevitz et al 11 as a method of preventing an increased pneumothorax that would require tube placement. However, there is a paucity of literature on the use of this simple technique since their first report. 11 In the current study, we used this technique for biopsy-induced pneumothorax more widely than Yankelevitz et al. 11 Hence, we believe that the information gained from the present study is important for physicians performing CT-guided lung biopsy. The mechanism that allows for re-expansion of the pneumothorax lung after manual aspiration is the re-approximation of the visceral and parietal pleural surfaces creating a physical barrier to further leakage of air. In the present study, CT images were used to grade the severity of pneumothoraces because this modality is considered superior to chest radiographs for this purpose. 21 A pneumothorax seen on more than seven CT slices was defined as modfluoroscopic scan shows an 18-gauge IV catheter entering the pleural space for aspiration. Note that a small amount of pulmonary bleeding occurred during the biopsy procedure. Bottom: Final CT scan obtained after 620 ml of air aspiration shows that only a small pneumothorax remained. The small pneumothorax completely disappeared on radiograph 7 days later. CHEST / 121 / 4/ APRIL,

6 erate or severe. As a result, 20 patients were judged as having moderate or severe pneumothorax and underwent manual aspiration regardless of symptoms. Only 2 of these 20 patients, in whom 600 ml and 720 ml of air was aspirated, respectively, required tube placement while the other 18 patients did not require chest tubes. The mean interval until resolution of the pneumothorax was approximately 3 to 4 days for patients with moderate or severe pneumothorax who underwent manual aspiration and for those with small pneumothorax that did not necessitate aspiration. Statistical analysis revealed no significant difference between these two groups (unpaired t test, p ). The median follow-up period for both groups was also similar. This result of our study demonstrates the value of immediate aspiration as a means of shortening the typical period required for resorption of a pneumothorax. However, there was one case involving a small pneumothorax on CT immediately after biopsy that was initially considered not to require aspiration, but later did necessitate tube placement. This suggests that manual aspiration might be further adapted for small pneumothoraces to further diminish the frequency of increased pneumothorax after biopsy. The limitation of our study is management of delayed pneumothorax. Rapid manual aspiration while the patient remains on the CT table limits the application of this immediate intervention to those cases detected at the time of biopsy. Therefore, cases in which a delayed pneumothorax is detected would not be amenable to this form of management. Although the one case in our study in which a delayed pneumothorax was detected the day after biopsy did not progress, previous reports have shown that increasing pneumothoraces often require chest tube placement. 22 In conclusion, when pneumothorax is revealed after CT-guided biopsy, performing immediate percutaneous aspiration even for an asymptomatic and not large pneumothorax increases the possibility of avoiding treatment using a chest tube. Consequently, the frequency of inpatient management or close outpatient monitoring of these pneumothoraces should decrease. In addition, real-time CT fluoroscopy as an adjunct to a series of procedures (ie, lung biopsy, detection of pneumothorax, and performance of percutaneous manual aspiration) allows safe and rapid management without excessive time demands. References 1 Westcott JL. Percutaneous transthoracic needle biopsy. Radiology 1988; 169: Austin JH, Cohen MB. Value of having a cytopathologist present during percutaneous fine-needle aspiration biopsy of lung: report of 55 cancer patients and meta-analysis of the literature. AJR Am J Roentgenol 1993; 160: Li H, Boiselle PM, Shepard JO, et al. Diagnostic accuracy and safety of CT-guided percutaneous needle aspiration biopsy of the lung: comparison of small and large pulmonary nodules. AJR Am J Roentgenol 1996; 167: Klein JS, Salomon G, Stewart EA. Transthoracic needle biopsy with a coaxially placed 20-gauge automated cutting needle: results in 122 patients. Radiology 1996; 198: Boiselle PM, Shepard JO, Mark EJ, et al. Routine addition of an automated biopsy device to fine needle biopsy of the lung: a prospective assessment. AJR Am J Roentgenol 1997; 169: Westcott JL. Direct percutaneous needle aspiration of localized pulmonary lesions: results in 422 patients. Radiology 1980; 137: Laurent F, Latrabe V, Vergier B, et al. Percutaneous CTguided biopsy of the lung: comparison between aspiration and automated cutting needle using a coaxial technique. Cardiovasc Intervent Radiol 2000; 23: Daly B, Templeton PA. Real-time CT. fluoroscopy: evolution of an interventional tool. Radiology 1999; 211: Brown KT, Brody LA, Getrajdman GI, et al. Outpatient treatment of iatrogenic pneumothorax after needle biopsy. Radiology 1997; 205: Gurley MB, Richli WR, Waugh KA. Outpatient management of pneumothorax after fine-needle aspiration: economic advantages for the hospital and patient. Radiology 1998; 209: Yankelevitz DF, Davis SD, Henschke CI. Aspiration of a large pneumothorax resulting from transthoracic needle biopsy. Radiology 1996; 200: Katada K, Kato R, Anno H, et al. Guidance with real-time CT fluoroscopy: early clinical experience. Radiology 1996; 200: Sanders C, Nath PH, Bailey WC. Detection of emphysema with computed tomography correlation with pulmonary function tests and chest radiography. Invest Radiol 1988; 23: Kazerooni EA, Lim FT, Mikhail A, et al. Risk of pneumothorax in CT-guided transthoracic needle aspiration biopsy of the lung. Radiology 1996; 198: Cox JE, Chiles C, McManus CM, et al. Transthoracic needle aspiration biopsy: variables that affect risk of pneumothorax. Radiology 1999; 212: Laurent F, Michel P, Latrabe V, et al. Pneumothorax and chest tube placement after CT-guided transthoracic lung biopsy using a coaxial technique: incidence and risk factors. AJR Am J Roentgenol 1999; 172: Cattelani L, Campodonico F, Rusca M, et al. CT-guided transthoracic needle biopsy in the diagnosis of chest tumours. J Cardiovasc Surg 1997; 38: Collings CL, Westcott JL, Banson NL, et al. Pneumothorax and dependent versus nondependent patient position after needle biopsy of the lung. Radiology 1999; 210: Kazerooni EA, Hartker FW, Whyte RI, et al. Transthoracic needle aspiration in patients with severe emphysema: a study of lung transplant candidates. Chest 1996; 109: Haramati LB, Austin JHM. Complications after CT-guided needle biopsy through aerated versus non-aerated lung. Radiology 1991; 181: Bungay HK, Berger J, Traill ZC, et al. Pneumothorax post CT-guided lung biopsy: a comparison between detection on chest radiographs and CT. Br J Radiol 1999; 72: Traill ZC, Gleeson FV. Delayed pneumothorax after CTguided percutaneous fine needle aspiration lung biopsy. Thorax 1997; 52: Clinical Investigations