| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 All authors: Department of Diagnostic Imaging, The Ottawa Hospital, University of Ottawa, 501 Smyth Rd., Ottawa, ON, Canada K1H 8L6.
Received April 22, 2008;
accepted after revision June 30, 2008.
Address correspondence to C. A. Souza
(csouza{at}ottawahospital.on.ca
or
carolina_althoff{at}yahoo.ca).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Retrospective review of the medical records of patients who underwent fluoroscopically guided transthoracic biopsy of pulmonary nodules with a 25-gauge needle yielded the cases of 107 consecutively registered patients (59 men, 48 women; mean age, 62 years). In 43 of the biopsies, the major fissure was crossed, and in 64 biopsies, the control procedures, the fissure was avoided. CT scans were assessed for lesion size and location, biopsy approach, length of needle path, number of needle punctures, and presence of emphysema.
RESULTS. Pneumothorax occurred in 11 patients (25%) in the transfissural biopsy group and in 19 patients (30%) in the group in which the fissure was avoided (p = 0.64). Pneumothorax necessitated chest tube placement in two patients (5%) in the transfissural biopsy group and seven patients (11%) in the control group (p = 0.25). In both groups, emphysema in the needle path was associated with increased risk of pneumothorax (p < 0.01).
CONCLUSION. Transthoracic needle biopsy with an ultrathin needle that crosses a lung fissure can be safely performed without increasing the rate of pneumothorax or the need for chest tube insertion.
Keywords: lung • lung biopsy • percutaneous biopsy • pneumothorax
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Various studies have been conducted to assess the risk factors for pneumothorax in patients undergoing transthoracic fine-needle aspiration. In usual practice, the needle path selected should be the shortest distance between the lesion and the chest wall and should avoid fissures, areas affected by emphysema, and bullae in the needle track. In some cases, however, a transfissural approach is the only way to reach lesions in the posterior segment of the upperlobes; to avoid bullae, emphysematous changes, and large vessels; and to minimize the length of the needle path to the target lesion. Crossing a lung fissure traditionally has been thought to be associated with increased risk of pneumothorax, and many authors consider the need to cross a fissure a relative contraindication to transthoracic fine-needle aspiration [8, 9]. To our knowledge, no study has been conducted to assess the incidence and clinical significance of pneumothorax associated with use of a transfissural approach to aspiration biopsy in comparison with biopsies in which a lung fissure is not crossed.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The biopsies were performed under fluoroscopic guidance (Axiom Artis Biplane angiography system, Siemens Medical Solutions) by either a staff thoracic radiologist or a fellow in thoracic radiology under staff supervision. In all cases, an on-site cytotechnologist immediately stained the smears and evaluated them for adequacy. After biopsy, patients were monitored by the nursing staff in the diagnostic imaging department. Dependent positioning of the puncture site was not advocated. According to the routine of our department, an upright posteroanterior chest radiograph was obtained 30 minutes after biopsy for patients without symptoms and immediately after the procedure for patients with symptoms such as chest pain or discomfort, dyspnea, and decreasing oxygen saturation. The chest radiographs were interpreted by the radiologist who performed the biopsy. Patients with no symptoms or pneumothorax were discharged. When a small pneumothorax was present, chest radiography was repeated 30 minutes after the first examination. If there was no increase in the size of the pneumothorax, patients without symptoms were discharged with postprocedure instructions. This approach to postbiopsy care and early discharge had been validated by our local experience [10].
A thoracostomy tube for pneumothorax drainage was placed in patients with symptomatic (dyspnea and decreasing oxygen saturation) small pneumothorax, those with asymptomatic increasing pneumothorax, and all patients with large-volume pneumothorax (> 25% of the pleural cavity) identified on the initial chest radiograph. All pneumothorax drainage catheters were placed by the radiologist who performed the biopsy. A nonlocking 8-French pigtail catheter was inserted in the midclavicular line of the second intercostal space and attached to a Heimlich valve (Heimlich Chest Drain Valve, BD Bard-Parker). After tube insertion, patients were discharged with chest tube instructions and were instructed to return for removal of the tube. None of the patients needed admission to the hospital because of persistent symptoms.
Two chest radiologists reviewed the chest CT scans obtained before biopsy and the biopsy reports for assessment of lesion size (largest transverse diameter on axial images) and location (right or left upper lobe), biopsy approach (anterior or posterior), length of the needle path, and number of needle punctures. The CT images were reviewed randomly, and the reviewers were blinded to the biopsy technique used and occurrence of pneumothorax. The target lesions were classified according to size as 2 cm or smaller, 2–4 cm, and 4 cm or larger. The length of aerated lung traversed by the needle was measured from the pleural surface to the edge of the lesion and classified as smaller than 2.5 cm, 2.5–5 cm, 5–7.5 cm, and larger than 7.5 cm. The presence of emphysematous changes and bullae in the needle path was recorded. The presence of emphysema in the lungs away from the needle path also was recorded, and a visual assessment was used for grading the severity of emphysema when present. The grade was considered mild when parenchymal involvement was 25% or less, moderate when it was 25–60%, and severe when it was 60% or greater.
The postbiopsy chest radiographs and biopsy reports were reviewed for the presence of complications such as hemoptysis and pneumothorax and for the need for chest tube placement. Information regarding the presence of delayed complications, defined as complications occurring after discharge from the diagnostic imaging department, was obtained from the computerized medical record.
Data analysis was performed with Stata software (version 10 for Microsoft Windows, Stata). The analysis consisted of logistic regression to test the causal relation between variables and the occurrence of pneumothorax. Variables were first tested individually and then in conjunction with other variables. Presence of emphysema and presence of emphysema in the needle path were categoric in nature, indicating the severity of the condition, and regression analysis was performed for each level of severity. Joint significance tests of one or more levels of severity were performed as needed. Noncategoric variables were kept intact. Pearson's chi-square tests were performed, and p < 0.05 was considered statistically significant.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
No significant correlation was found between the biopsy approach used (anterior or posterior) and the occurrence of pneumothorax. All patients in the transfissural biopsy group (43 of 107) had upper lobe lesions biopsied through a posterior transfissural approach (Figs. 1, 2, 3). In the control group (64 of 107 patients), lesions were situated in the upper lobes and were biopsied with a posterior (n = 19) or anterior (n = 45) approach in which the major fissure was not crossed.
|
|
|
|
The odds of pneumothorax occurrence were estimated in relation to the size of the lesion, number of punctures, length of lung traversed, and presence and severity of emphysema and of emphysema in the needle path. The presence of emphysematous changes in the needle path was an isolated variable associated with a significant increase in the risk of pneumothorax (p < 0.01). As the overall severity of emphysema elsewhere in the lungs increased, the odds of occurrence of pneumothorax also increased (p < 0.05). No significant correlation was found between presence of pneumothorax and lesion size, number of pleural punctures, or length of lung parenchyma traversed.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Crossing a lung fissure during transthoracic biopsy has been avoided in clinical practice, and the need to cross a fissure is considered a relative contraindication to biopsy when such an approach is the only option [1, 8, 9]. This is based on the principle that a greater number of pleural punctures may increase the risk of pneumothorax. Lung fissures consist of double layers of invaginations of visceral pleura. When a fissure is crossed, two additional pleural punctures are made, theoretically increasing the risk of pneumothorax. When a transfissural approach is used, each needle pass represents three pleural punctures. In our study, the frequency of pneumothorax with a transfissural approach was similar to that in the control group, in which a fissure was not crossed (25% and 30%, respectively; p = 0.64). Most of the patients in both groups had more than one needle pass, and the number of passes (maximum of four) was not associated with an increased rate of pneumothorax. Although it may be considered surprising, this finding is corroborated by the results of studies [1, 4, 7, 13] in which no correlation was found between number of pleural punctures and frequency of pneumothorax.
All biopsies in our study were performed with an ultrathin (25-gauge) needle. In a study involving 123 patients, our group [12] found ultrathin needle biopsy accurate in the diagnosis of malignant lung lesions, having a sensitivity, specificity, and diagnostic accuracy of 93.6%, 100%, and 94.4%. In all patients in the transfissural biopsy group in the current study, the biopsy was performed with a posterior approach, which has been favored by some authors [8, 9] because it may be associated with a decreased rate of pneumothorax. The reason for a lower rate of pneumothorax may be reduced motion of the posterior chest wall compared with the anterior chest wall during respiration. The reduction in motion consequently decreases the possibility of shearing the pleura when the needle is placed. Having the patient assume the supine or seated position after biopsy places the posterior side down, compressing the puncture site and possibly reducing the risk of pneumothorax. Moreover, the patient is unable to see the needle and view the biopsy procedure when in the prone position and this may decrease the level of anxiety. According to Moore [8], a posterior approach should be used whenever possible if no fissures will be violated. It is uncertain whether choice of approach influenced the rate of pneumothorax in our study or whether crossing the fissure in an anterior approach should be avoided.
Another consideration is whether the use of larger-gauge needles increases the incidence of pneumothorax. A 25-gauge needle makes a small pinhole in the pleura, and the flexibility of the needle may prevent shearing of the pleural hole during respiration. Within the range of fine needles generally used for lung biopsy (less than 20-gauge), most studies [1, 7, 8, 11] have shown no significant correlation between needle gauge and occurrence of pneumothorax. However, many centers in North America that have limited cytologic experience have adopted the use of larger needles and often perform core needle biopsy. Further studies comparing pneumothorax rates with larger-needle core biopsy and biopsy by the transfissural approach are warranted.
Laurent et al. [7], in a study of 307 transthoracic needle biopsies, found that lesion depth was the sole variable associated with a significant increase in the risk of pneumothorax, the risk increasing with the depth of the lesion. The presence of emphysema in our study was not associated with an increased risk of pneumothorax, in agreement with findings in other series [6, 7, 13]. In the study by Kazerooni et al. [6], the risk increased by a factor of 2 for lesions farther than 2 cm from the pleural surface. Topal and Berkman [11] found that not only the depth of the lesion but also the severity of emphysema was associated with increased risk of pneumothorax. In contrast, in our study, the overall rate of pneumothorax and the rates of pneumothorax for each of the groups were not influenced by the depth of the lesion. This finding may be explained by the short length of lung traversed in our patients; in many cases the transfissural approach indeed was used to minimize the distance between the lesion and the chest wall.
There is debate in the literature with regard to the relation between the size of a lesion and the occurrence of pneumothorax. Whereas some authors [1] found an increased pneumothorax rate associated with biopsy of small lesions, other authors, as did we, found the size of the lesion was not a single independent variable in the occurrence of pneumothorax [6, 11].
An isolated variable associated with a significant increase in the risk of pneumothorax in our study was the presence of emphysema, including paraseptal emphysema and bullae, in the needle path (p < 0.01). Similar results were described in other studies [1, 8], in which the presence of emphysema in the vicinity of the lesion was associated with a significant increase in the frequency of pneumothorax. It is well known that bullae and emphysema in the needle path should be avoided because broaching them usually causes substantial air leak. In addition, the odds of pneumothorax increase as the severity of emphysema elsewhere in the lungs increases (p < 0.05). Previous reports described a high rate of pneumothorax in patients with obstructive lung disease [12–14], and the presence of emphysema and hyperinflation has been associated with an increased frequency of chest tube placement [6–8]. Several authors [1, 6, 7, 12] have assessed the role of pulmonary function test results as a predictor of pneumothorax. The study findings are controversial because the patient populations differed and different variables were analyzed. It seems, however, that no single respiratory function variable is a good predictor of increased risk of pneumothorax.
The risk of pneumothorax at CT-guided transthoracic fine-needle aspiration may be higher than that at fluoroscopically guided transthoracic fine-needle aspiration, probably because CT-guided biopsy requires more time and the average size of the lesions is smaller [1, 8]. The superior sensitivity of CT in the detection of pneumothorax also may contribute to the higher rate [15]. All biopsies in our study were performed under fluoroscopic guidance, and it is uncertain whether our data can be extrapolated to CT-guided biopsy.
Most pneumothoraces that occur after transthoracic biopsy are small and asymptomatic and resolve spontaneously. The incidence of clinically significant pneumothorax necessitating chest tube placement varies from 2% to 21% [1, 3, 4, 7, 11]. In our study, 8% of the patients with pneumothorax needed chest tube drainage, and there was no significant difference between the two groups (5% in the transfissural biopsy group, 11% in the control group; p = 0.25). It has been found [1, 8] that evidence of emphysema on CT scans is associated with a significant increase in the number of procedures necessitating chest tube placement. In the study by Cox et al. [1], for example, patients with pneumothorax and CT evidence of emphysema needed chest tubes at three times the rate of patients with pneumothorax and no CT evidence of emphysema. We found no significant relation between the severity of emphysema on CT and the need for chest tube placement (p = 0.38). There was, however, a statistically significant correlation between the severity of emphysematous changes along the needle path and pneumothorax necessitating chest tube drainage (p = 0.01).
The limitations of our study were the retrospective design and the relatively small number of patients. In addition, all biopsies were performed under fluoroscopic guidance with an ultrathin needle, which might have been the reason for the similar rates of pneumothorax in the two groups. The safety of biopsy that crosses a fissure with a larger-gauge needle and of biopsy performed with CT guidance cannot be determined with our data, although previous studies have shown no correlation between the size of the needle and an increased rate of pneumothorax. This study was the first, to our knowledge, specifically addressing the incidence of pneumothorax with use of a transfissural biopsy approach in comparison with findings in a control group in which a lung fissure was not crossed. Transthoracic fine-needle aspiration with a posterior transfissural approach was found safe, and the need for this approach should not be a contraindication to percutaneous biopsy. If a transfissural approach is being viewed as a contraindication to percutaneous biopsy, our findings may be helpful in reducing the number of thoracotomies. Our results are particularly important in the era of lung cancer screening and with the increasing detection of nodules requiring diagnosis.
Transthoracic fine-needle aspiration with an ultrathin (25-gauge) needle and crossing a lung fissure can be safely performed without an increase in the risk of pneumothorax or chest tube insertion. A transfissural approach can be used to biopsy lesions in the posterior segment of the upper lobes; to minimize the length of the needle path; and to avoid bullae, emphysematous changes, and large vessels. These pos sibilities are particularly important when crossing the fissure is the only alternative to percutaneous biopsy and to avoid more invasive diagnostic procedures. A needle path that traverses areas substantially affected by emphysema should be avoided whenever possible because traversing such areas can increase the risk of pneumothorax and the need for chest tube placement.
Acknowledgments
We thank Eduardo Magalhães for his outstanding assistance with the
statistical analysis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. De Filippo, R. Averna, and M. Zompatori Ability to Reduce the Incidence of Pneumothorax in Transthoracic CT-Guided Biopsy Am. J. Roentgenol., November 1, 2009; 193(5): W466 - W467. [Full Text] [PDF]
|
|
|
|
|
|
C. A. Souza Reply Am. J. Roentgenol., November 1, 2009; 193(5): W468 - W468. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
