Percutaneous computed tomography-guided lung biopsy and pleural dissemination
An assessment by intraoperative pleural lavage cytology
Abstract
BACKGROUND:
Percutaneous computed tomography (CT)-guided needle biopsy remains one of the most important diagnostic tools in the management of lung nodules; however, it carries a risk of intrapleural dissemination of cancer cells.
METHODS:
CT-guided lung biopsy was performed before surgery in 171 (34.8%) of 491 patients. A coaxial biopsy system was used that comprised a 19-gauge introducer needle and a 20-gauge core biopsy needle. A total of 412 (83.9%) of the 491 patients underwent intraoperative pleural lavage cytology just after thoracotomy. Intraoperative pleural lavage cytology was performed immediately after opening the thorax, after the pleural cavity was gently washed with 50 mL of saline.
RESULTS:
No patients had implantation of cancer cells in the chest wall after a median follow-up of 20.2 months. Intraoperative pleural lavage cytology results were positive for 5 (2.9%) of the 171 patients who underwent CT-guided biopsy before surgery, in contrast to 13 (5.4%) of the 241 patients who did not undergo biopsy before surgery. The difference between the biopsy and nonbiopsy groups was not statistically significant. When the analysis was limited to patients with stage IA disease, intraoperative pleural lavage cytology results were positive for 1 (0.8%) of the 128 patients who underwent CT-guided biopsy, in contrast to 3 (2.7%) of the 110 patients who did not undergo biopsy. This difference was also not statistically significant.
CONCLUSIONS:
No significant association was observed between percutaneous CT-guided lung biopsy and intraoperative pleural lavage cytology results, even in patients with stage IA disease. Percutaneous CT-guided lung biopsy with a coaxial needle does not seem to cause pleural dissemination. Cancer 2009. © 2009 American Cancer Society.
Computed tomography (CT) is routinely used for the detection of various kinds of nodules in the lungs, such as carcinomas, benign tumors, infectious diseases, inflammatory diseases, vascular diseases, and intrapulmonary lymph nodes. As a result of technological refinements, current imaging techniques allow the detection of small solitary nodules in the pulmonary parenchyma, but the characterization of such nodules may be difficult even with high-resolution multidetector-row CT. Thus, both transbronchial lung biopsy and percutaneous CT-guided needle biopsy are used for definite diagnosis of lung cancer or to exclude malignancies. In most institutions, percutaneous CT-guided needle biopsy or aspiration cytology is indicated when a diagnosis by transbronchial lung biopsy is not possible because of the risk of potential complications with needle biopsy. In particular, many pulmonologists and thoracic surgeons believe that percutaneous CT-guided needle biopsy carries the risk of needle tract seeding or a risk of intrapleural dissemination of cancer cells. Even if percutaneous CT-guided needle biopsy may cause needle tract seeding or dissemination, this risk must be balanced against the risk of unnecessary surgical resection in patients who do not have malignancy. Thus, it would be important to know the true magnitude of the risk associated with this procedure.
Intraoperative pleural lavage cytology is the most simple and direct method of evaluating tumor cells disseminated into the pleural cavity by needle biopsy. We found only 1 article published on the relationship between CT-guided transthoracic needle biopsy and pleural recurrence1; however, no studies have been published on the relationship between needle biopsy and the results of intraoperative pleural lavage cytology. Although some reports have been published on the risk of dissemination by needle lung biopsy through investigation of cancer recurrence pattern or prognosis, the true incidence of tumor seeding along the needle may be underestimated because not all cases can be diagnosed. It is, therefore, necessary to wait several years to clinically identify dissemination after the procedure, and many patients die before these metastases become clinically apparent.
Thus, the purpose of the present study was to use intraoperative pleural lavage cytology to evaluate whether percutaneous CT-guided transthoracic needle biopsy is associated with an increased risk of pleural dissemination. We compared the results of intraoperative pleural lavage cytology between a group of patients with surgically resected lung cancer who underwent percutaneous CT-guided transthoracic needle biopsy and a concurrent cohort of patients who did not undergo this procedure.
MATERIALS AND METHODS
Patients
Between February 2004 and January 2008, 491 patients with primary lung cancer underwent radical surgical resection of the lungs at Okayama University Hospital. Only patients with a definite pathological diagnosis of primary lung cancer were included in the present study. The starting point was selected because it marked the beginning of intraoperative pleural lavage cytology at our institution. Patients who had macroscopic pleural effusion, dissemination, or severe adhesion were excluded from the analysis. This study was approved by our Institutional Review Board, and written informed consent was obtained from all patients before the procedure.
CT-Guided Lung Biopsy
The biopsy technique has been described previously.2 Briefly, it was performed percutaneously under CT-fluoroscopic guidance by an experienced radiologist. The instrument used was a coaxial biopsy system that comprised a 19-gauge introducer needle (Co-Axial Introducer Needle; Medical Device Technologies; Gainesville, Fla) and a 20-gauge core biopsy needle (Super-Core II Biopsy Instrument; Medical Device Technologies). Patients were placed in the supine or prone position. Routine techniques used were as follows: 1) after the administration of local anesthesia, the introducer needle was advanced along the determined path until its tip was in front of the lesion; 2) the internal stylet was removed, and the biopsy needle was immediately replaced; 3) the stylet of the biopsy needle was advanced into the lesion; 4) a specimen was obtained by pressing the plunger, and the biopsy needle was removed followed by immediate replacement by the stylet of the introducer needle; 5) processes 2 to 4 were repeated until the specimens obtained were considered to be adequate for histological evaluation; 6) the introducer needle was removed to finish the procedure; and 7) chest CT scan images were obtained to evaluate procedural complications.
Intraoperative Pleural Lavage Cytology
Intraoperative pleural lavage cytology was performed following the method described in a previous study3 with some modifications. Briefly, immediately after opening the thorax, the pleural cavity was gently washed with 50 mL of saline. To obtain only those cancer cells that were present in the intrapleural space, we avoided contact with any of the pleural surfaces on the primary tumor. Next, 20 mL of the lavaged fluid was collected. The sediment obtained was stained by the Giemsa, Papanicolaou, and Periodic acid Schiff methods to determine the presence or absence of cancer cells. Cytologists classified the specimen into 2 categories: Papanicolaou classes I, II, and III as negative and classes IV and V as positive.
Statistics
Data are presented as mean ± SD. Student t test was used for continuous variables and Fisher exact tests for categorical variables. All tests were 2-sided and a P value of <0.05 was considered significant. Statistical analysis was conducted using StatView 5.0 (SAS Institute, Cary, NC).
RESULTS
A total of 171 (34.8%) of the 491 patients underwent CT-guided lung biopsy for diagnostic purposes before surgery. The results are shown in Table 1. The sensitivity of the CT-guided needle biopsy was found to be 95.9%, and its positive predictive value was 100%.
| Malignancy/Suspicious of Malignancy | No. (%) |
|---|---|
| Adenocarcinoma | 119 |
| Squamous cell carcinoma | 11 |
| Nonsmall cell carcinoma | 15 |
| Bronchioloalveolar carcinoma | 5 |
| Small cell carcinoma | 2 |
| Carcinoid | 1 |
| Suspicious of adenocarcinoma | 7 |
| Suspicious of squamous cell carcinoma | 2 |
| Suspicious of bronchioloalveolar carcinoma | 1 |
| Malignant tumor | 1 |
| Total | 164 (95.9) |
| Others | |
| Atypical cells | 2 |
| No malignancy | 5 |
| Total | 7 (4.1) |
A total of 412 (83.9%) of the 491 patients underwent intraoperative pleural lavage cytology to detect malignant cells immediately after thoracotomy. A median follow-up at 20.2 months revealed that none of the patients had implantation of cancer cells in the chest wall. The rates of recurrence of lung cancer including distant metastases, lymph node metastases, local recurrence, dissemination, or malignant pleural effusion were 17.4% (42 of 241) in the biopsy group and 13.5% (23 of 171) in the nonbiopsy group. This difference between the groups was not statistically significant (Fisher exact test: P = .415). The rates of pleural dissemination or malignant pleural effusion were 3.3% (8 of 241) in the biopsy group, and 2.3% (4 of 171) in the nonbiopsy group. This difference was also not statistically significant (Fisher exact test: P = .768). Among all patients who underwent intraoperative pleural lavage cytology, there were 18 positive cases (4.4%) and 394 negative cases (95.6%). The intraoperative pleural lavage cytology result was positive in case of 5 of the 171 patients (2.9%) who underwent CT-guided biopsy before surgery, whereas it was positive for 13 of 241 patients (5.4%) who did not undergo biopsy before surgery. This difference between the groups was not significant (Fisher exact test: P = .217).
The patients' clinicopathological characteristics according to whether they underwent CT-guided needle biopsy are shown in Table 2. The extent of pleural invasion did not differ significantly between the biopsy and nonbiopsy groups (Fisher exact test; P = .35). However, there were significant differences in gender, tumor size, pathological T-factor, and pathological stage between the groups (Table 2). These phenomena can be explained because the biopsy group patients had smaller nodules as compared with the nonbiopsy group patients (Student t tests; P<.0001). Most patients who have large tumors can be diagnosed by transbronchial lung biopsy and need not to undergo CT-guided needle biopsy.
| Factors | Intraoperative Pleural Lavage Cytology Performed, N=412 | P | |
|---|---|---|---|
| Biopsy n=171 | Nonbiopsy n=241 | ||
| Intraoperative pleural lavage cytology+ | 5 (2.9%) | 13 (5.4%) | .217 |
| Intraoperative pleural lavage cytology− | 166 (97.1%) | 228 (94.6%) | |
| Age | 66.6 | 65.3 | .235 |
| Sex | |||
| Men | 94 | 158 | |
| Women | 77 | 83 | .031 |
| CEA | 4.50 | 7.78 | .055 |
| Pathological stage | |||
| I | 151 | 172 | |
| II | 7 | 32 | <.0001 |
| III | 13 | 35 | Stage I vs others |
| IV | 0 | 2 | |
| Histology | |||
| Adenocarcinoma | 139 | 170 | |
| Squamous cell carcinoma | 21 | 55 | .660 |
| Large cell carcinoma | 2 | 4 | Adenocarcinoma vs others |
| Small cell carcinoma | 4 | 3 | |
| Adenosquamous carcinoma | 3 | 4 | |
| Others | 2 | 3 | |
| Tumor size | 21.7 | 28.6 | <.0001 |
| Pathological pleural involvement | |||
| p0 | 134 | 179 | |
| p1 | 25 | 40 | 0.35 |
| p2 | 4 | 9 | (p0 vs others) |
| p3 | 8 | 13 | |
| Pathological T classification | |||
| T1 | 135 | 128 | |
| T2 | 29 | 95 | <.0001 |
| T3 | 5 | 11 | T1 vs others |
| T3 | 2 | 7 | |
- CEA indicates carcinoembryonic antigen; +, positive; −, negative.
To avoid any differences arising from tumor staging, we next analyzed the intraoperative pleural lavage cytology results of only patients with only stage IA disease (Table 3). Of 238 patients with stage IA disease, the intraoperative pleural lavage cytology results were positive for 4 patients (1.7%) and negative for 234 patients (98.3%). In contrast to 1 of the 128 patients who had undergone CT-guided needle biopsy (0.8%), 3 (2.7%) of the 110 patients who had not undergone biopsy were intraoperative pleural lavage cytology positive. Although there were a few more positive cases in the biopsy group than in the nonbiopsy group, this difference was not significant (Fisher exact test; P = .338).
| Factors | Intraoperative Pleural Lavage Cytology, IA: n=238 | P | |
|---|---|---|---|
| Biopsy n=128 | No Biopsy n=110 | ||
| Intraoperative pleural lavage cytology+ | 1 (0.8%) | 3 (2.7%) | .338 |
| Intraoperative pleural lavage cytology− | 127 (98.7%) | 107 (95.3%) | |
| Age | 66.2 | 64.1 | .147 |
| Sex | |||
| Men | 67 | 60 | .795 |
| Women | 61 | 50 | |
| CEA | 3.16 | 3.41 | .437 |
| Tumor size | 17.80 | 17.35 | .622 |
| Histology | |||
| Adenocarcinoma | 106 | 90 | .866 |
| Others | 22 | 20 | |
| Pathological pleural involvement | |||
| p0 | 115 | 96 | |
| p1 | 13 | 14 | .546 |
- CEA indicates carcinoembryonic antigen; +, positive; −, negative.
DISCUSSION
Histopathological findings are often crucial for diagnosing pulmonary nodules and for developing an appropriate therapeutic strategy. CT-guided needle biopsy remains one of the most important diagnostic tools in the management of such nodules; however, it is very important to understand the comorbidities related to the procedure.
It is well known that the sensitivity of bronchoscopy is high for endobronchial diseases, but poor for peripheral lesions, ranging from 36 to 86%, especially in those with small diameters. Recently, newer technologies, such as electromagnetic navigation bronchoscopy, have been introduced and the reported sensitivity ranges from 62 to 82%.4 The sensitivity of percutaneous CT-guided needle biopsy or aspiration cytology is much better for peripheral malignant nodules.5 Fine-needle aspiration with cytological examination has been used for a long time because samples can be obtained with smaller needles; however, the accuracy of aspiration cytology is lower than that of needle biopsy. The specificity and positive predictive value of percutaneous CT-guided needle biopsy are very high6, 7; thus, there are virtually no false-positive findings. In contrast, the negative predictive value of this technique remains low.8 With the use of percutaneous CT-guided biopsy of pulmonary lesions with a coaxial automated device, the sensitivity, specificity, positive predictive value, and negative predictive value for the diagnosis of malignancy were 92.2% to 93%, 99.1% to 100%, 99.8% to 100%, and 73.3% to 88%, respectively.9, 10 In the present study, the sensitivity of percutaneous CT-guided needle biopsy was 95.9%, and the positive predictive value was 100%.
Besides characterizing nodules as either malignant or nonmalignant, the results of a biopsy may help to address the important issue of tumor genetic status, such as mutations in the epidermal growth factor receptor (EGFR) gene.11, 12 Mutations in the EGFR gene have been reported to be related to the responsiveness of tumors to EGFR tyrosine kinase inhibitors.13, 14
Percutaneous CT-guided transthoracic needle biopsy is generally regarded as a safe procedure with limited associated morbidity, but it can be associated with some minor complications such as pneumothorax and pulmonary hemorrhage, most of which improve without any treatment. It is also well known that rare but potentially fatal complications such as air embolism and tumor seeding can occur. According to a survey that included 9783 biopsy specimens collected from 124 centers in Japan, pneumothorax was the most common complication, occurring in 35% of the cases. Overall, severe complications were observed in 0.75% of the cases. Of these, 0.061% had air embolism, tumor seeding at the site of the biopsy route, and severe pulmonary hemorrhage or hemoptysis, and 0.10% had tension pneumothorax and hemothorax. Of 62 cases with severe complications, 54 cases (0.55%) were recovered without sequela, and 1 case (0.01%) recovered, albeit with hemiplegia due to cerebral infarction. Unfortunately, 4 (0.04%) of the remaining 7 cases died immediately after the CT-guided biopsy procedure.15
Although it has been suggested that both needle biopsy and surgical resection in patients with lung malignancies are associated with a spreading of malignant cells within the pleural cavity, it has not been clearly shown that coaxial needle biopsy increases the risk of recurrence or poor prognosis. In the liver field, a meta-analysis was published that included 8 studies identified by systematic review on biopsy of hepatocellular carcinoma. These authors showed that the incidence of needle tract tumor seeding after biopsy of a hepatocellular carcinoma is 2.7% overall, or 0.9% per year.16 Also in the liver field, a study suggested that the risk of seeding could be reduced by the use of a coaxial cutting needle technique.17
In the field of intrathoracic biopsy, several case reports have suggested that tumor seeding into the needle tract seems to be a rare possibility.18-25 However, the true incidence of tumor seeding along the needle may be underestimated because not all cases can be diagnosed, and many patients die before these metastases become clinically apparent.15 Pleural dissemination distant from the puncture site has rarely been reported. On the other hand, it has been reported that the possibility of chest wall implantation caused by CT-guided lung biopsy was 0.04-1% or less.15, 26
Ayar et al27 reported the results of a survey of the membership of the Society of Thoracic Radiology and a review of the English-language literature to assess the incidence of metastasis along the needle track after a transthoracic needle biopsy and its predisposing factors as well as natural history. Approximately 68,346 procedures were included in the analysis. Five departments reported 8 needle track metastases, resulting in an incidence of 0.012%, and there were no predictable risk factors. In another study, Sinner et al28 reported that after 5300 percutaneous transthoracic needle aspiration biopsy procedures, needle track implantation occurred in 0.02% of the patients.
Matsuguma et al1 retrospectively investigated the relationship between the diagnostic methods used and cancer recurrence patterns. Pleural recurrence or needle track implantation was observed for 8.6% of the patients who underwent a needle biopsy, whereas it was observed for 0.9% of the patients who were examined by means of other diagnostic modalities such as bronchoscopy and wedge resection. These authors concluded that needle biopsy can cause a pleural recurrence in addition to needle track implantation.
In a survey conducted in Japan,15 tumor seeding at the site of the biopsy route was seen in 0.061% of the cases. In addition, it occurred in 2 cases after biopsy was performed by the coaxial method, although it has been believed that CT-guided biopsy performed with the coaxial method has a lower frequency of tumor seeding because the outer cannula minimizes direct contact of the tumor cells with the biopsy route. In 1 of those 2 cases, however, the tip of the outer cannula was placed within the chest wall, such that seeding obviously occurred by direct contact of the inner needle with the biopsy route.
The detection of free cancer cells within the peritoneal cavity at the time of surgery has been shown to influence prognosis in abdominal malignancies, especially in gastric,29-31 colorectal,32 and gynecological cancers.33 Recently, intraoperative pleural lavage cytology has also been recognized as an independent prognostic factor in lung cancer.
In our series, the proportion of positive intraoperative pleural lavage cytology results was relatively low (all patients, 4.4%; patients with stage IA disease, 1.7%). However, the percentage of positive pleural lavage cytology results reported in the literature ranges from 3.2% to 22.6%.34-39 It is notable that in our series, 1.7% of patients with pathological stage IA disease were also included in the intraoperative pleural lavage cytology positive group. In the literature, the percentage of positive pleural lavage cytology results ranges from 2.2% to 7.3%34, 36, 38 in patients with stage I disease. The proportion of positive results may be affected by the method used to acquire the sample, especially the volume of lavage fluid used for acquiring samples. In addition, we hypothesized that some preoperative intervention such as CT-guided needle biopsy may vary these results.
Many studies have identified positive pleural lavage results as an independent predictor of prognosis.34-39 In addition, some studies that reported a poorer survival were able to confirm even in stage I.37-39 Interestingly, regarding the recurrence pattern in patients with positive pleural lavage cytology results, distant metastases were observed more frequently than local recurrences.36, 38 However, where and how the result of pleural lavage cytology could be incorporated into international lung cancer staging is still controversial.
Only a few papers have been published on the correlation between needle biopsy of the lung and the results of intraoperative pleural lavage cytology. Kjellberg et al40 reported on 78 patients who underwent curative resection for lung cancer in whom lavage cytology testing was performed before lung manipulation and after resection. Positive lavage cytology results before resection were identified in 14% of patients. Of the 78 patients, 30 had undergone preoperative diagnostic transthoracic needle biopsy, 17% of whom had positive lavage cytology results. No statistically significant association was found between lavage cytology results before resection and transthoracic needle biopsy.
CONCLUSIONS
When invasive diagnostic or therapeutic procedures are available, it is necessary to consider both the benefits and the risks of these options. We found no significant association between percutaneous CT-guided lung biopsy and intraoperative pleural lavage cytology results, even in patients with stage IA disease. Percutaneous CT-guided lung biopsy with a coaxial needle does not seem to cause pleural dissemination.
Conflict of Interest Disclosures
The authors made no disclosures.
