DOI: 10.1148/radiol.2353040433
(Radiology 2005;235:1045-1049.)
© RSNA, 2005
New Suction Guide Needle Designed to Reduce the Incidence of Biopsy-related Pneumothorax: Experimental Evaluation in Canine Model1
Frank A. Morello, Jr, MD,
Kenneth C. Wright, PhD and
Thomas M. Lembo, DVM
1 From the Departments of Interventional Radiology (F.A.M, K.C.W.) and Veterinary Medicine and Surgery (T.M.L.), University of Texas M. D. Anderson Cancer Center, Houston, Tex. From the 2003 RSNA Annual Meeting. Received March 5, 2004; revision requested May 19; revision received July 29; accepted August 20. Supported in part by a grant from the John S. Dunn Research Foundation and by grant NIH-NCI CA-16672 from the National Cancer Institute. Address correspondence to F.A.M., 3003 Taylorcrest Dr, Pearland, TX 77584 (e-mail: FMorello@tmh.tmc.edu).
 |
ABSTRACT
|
|---|
In an attempt to remove air that enters the pleural space during computed tomography (CT)-guided coaxial transthoracic needle biopsy, the authors fashioned an 18-gauge experimental suction guide needle and evaluated the incidence of pneumothorax with this needle in comparison to the incidence of pneumothorax with a standard 18-gauge guide needle in a canine model. This experiment had animal care and use committee approval. Ten dogs underwent a biopsy of each lung, for a total of 20 lung biopsies. Half of the biopsies were performed by using the experimental needle (five right lungs, five left lungs), and half were performed by using a standard guide needle. CT revealed pneumothorax during the procedure and was performed to reveal pneumothorax 1 and 3 hours after the procedure. A significant reduction (P < .016) in intraprocedural lung biopsy–associated pneumothorax was found when the experimental guide needle was used.
© RSNA, 2005
|
INTRODUCTION
|
|---|
Transthoracic needle biopsy (TNB) became popular after Nordenstrom introduced the fine-needle aspiration technique in 1965 (1). Wallace et al (2) have recently reviewed the literature and found that this percutaneous technique has reported diagnostic accuracy rates in excess of 93% and sensitivity rates in excess of 95%. Because of its high diagnostic accuracy and less-invasive nature compared with surgery, TNB has become the procedure of choice for the diagnosis of pulmonary lesions. In addition, some thoracic surgeons advocate the use of TNB prior to mediastinoscopy for tissue sampling of mediastinal lesions (3).
Pneumothorax remains the most common complication of TNB and has been reported to occur in 0%–61% of procedures, with 1.6%–17.0% of such pneumothoraces requiring chest tube treatment (4). Pneumothoraces are rarely life threatening, but their treatment, with either close observation or chest tube drainage, exposes patients to increased morbidity and expense. Several factors thought to increase the risk of a biopsy-related pneumothorax include large needle size, increased number of pleural punctures, increased needle dwell time, obstructive lung disease, small lesion size, and increased skin-to-lesion depth (2,5–7). Although some of these factors are controllable, there has been conflicting evidence regarding the importance of many of these factors in clinical trials. However, several reports have consistently shown that obstructive lung disease significantly increases the risk that a symptomatic pneumothorax that requires chest tube treatment will occur (8–10). Unfortunately, this factor is independent of operator or patient control during TNB.
Because many complicated mechanisms affect the risk that a biopsy-related pneumothorax will occur, the incidence of this complication has not dramatically changed over the years. Two strategies for decreasing the chances of pneumothorax and chest tube placement are dependent biopsy site positioning and injection of a blood patch or fibrin product in the needle path to help seal the pleural puncture site (11,12). Although some authors have reported success with these techniques, they are not routinely used owing to the inconsistency of their effectiveness. The exact mechanisms and numerous factors that affect the incidence of pneumothorax may never be completely understood. It is known that the space between the parietal and the visceral pleura maintains a negative pressure that helps the lung stay fully inflated. We hypothesized that the incidence of biopsy-related pneumothorax should be lower if positive-pressure air does not enter the pleural space and disrupt this natural vacuum. Thus, the purpose of our study was to evaluate, in a canine model, the incidence of pneumothorax when an experimental suction guide needle designed to remove air that enters the pleural space is used during TNB.
|
Materials and Methods
|
|---|
Animals and Sedation Protocol
Our institution’s Animal Care and Use Committee approved this canine experiment after ensuring it would be conducted according to the guidelines and requirements set forth by the National Institutes of Health Public Health Service Policy on Humane Care and Use of Laboratory Animals (13) and the U.S. Department of Health and Human Services Guide for the Care and Use of Laboratory Animals (14). Ten large dogs (The University of Texas-Houston Medical School, Houston, Tex) with a mean body weight of 30 kg were used in this protocol because results of our preliminary experiments with computed tomography (CT)-guided lung biopsies (conducted by using a similar protocol) indicated that the lungs of large dogs better tolerated an indwelling 18-gauge needle than did the lungs of pigs or smaller animals (eg, rabbits). These preliminary feasibility experiments in rabbit and pig lungs consistently resulted in severe pneumothoraces that could not be relieved with suction.
Veterinary personnel performed all animal care and sedation procedures under the direct supervision of a veterinarian (T.M.L). After a peripheral leg intravenous catheter was placed, sedation was induced with a 0.75 mg/m2 intravenous bolus of medetomidine (Domitor; Pfizer Animal Health, Eaton, Pa) and an intramuscular injection of 0.04 mg of atropine (Atropine Sulfate; American Pharmaceutical Partners, Schaumburg, Ill) per kilogram of body weight. To duplicate the clinical scenario of a human lung biopsy and to avoid pulmonary pressure changes that might affect the natural occurrence of a biopsy-related pneumothorax, we used neither endotracheal intubation nor mechanical ventilation.
Sedation was maintained during the procedure with mask delivery of 1.5% isoflurane (IsoFlo; Abbott Laboratories, Chicago, Ill) in oxygen that was delivered at a rate of 5 mL per pound per minute. The effect of the medetomidine was reversed at the end of the procedure with a 3.0-mg intravenous bolus of atipamezole (Antisedan; Pfizer Animal Health). In accordance with the instructions on the atipamezole label, the volume of atipamezole that was introduced by intramuscular injection was identical to the previously administered volume of intravenous medetomidine. After the biopsy, all dogs were closely observed for sedation recovery and potential biopsy complications.
Experimental Suction Guide Needle and Biopsy Technique
Each of the 10 dogs underwent one biopsy procedure in each lung. In the first half of the experiment, one lung was sampled for biopsy in each dog, and, after a 2-week recovery, the contralateral lung was sampled for biopsy in each dog in the second half of the experiment. Throughout the experiment, no more than two dogs underwent biopsy in 1 day. For half of the total number of procedures (n = 10 [five right lungs and five left lungs]), we used the same size coaxial system that we use for CT-guided human lung biopsies at our institution. This system included an 18-gauge, 5-cm-long Chiba needle (Cook; Bloomington, Ind) as a guide for a coaxial 22-gauge Chiba aspiration needle (Cook) and a coaxial 20-gauge Quick-Core biopsy needle (Cook).
For the other half of the total number of procedures (n = 10 [five right lungs and five left lungs]), the new suction guide needle was used. This needle was created by first cutting several side holes that communicated with the inner needle lumen along the shaft of a standard 5-cm-long 18-gauge guide needle (Fig 1). The side holes were cut in a spiral fashion along the needle shaft so as not to weaken the needle and cause inadvertent fracture. A Tuohy-Borst type adaptor with a side port was attached to the guide needle hub by means of the Luer lock end of the adaptor (Boston Scientific, Natick, Mass) (Fig 2), and its side port was attached to standard wall suction set at 100 mm Hg of suction. A needle stylet for a 10-cm-long 19-gauge Chiba needle (Cook) was then cut to fit the length of the 5-cm-long 18-gauge needle plus the length of the attached valve. A 19-gauge stylet was used because it is smaller in diameter than an 18-gauge stylet, and air entering the needle could travel through the needle lumen around the stylet and out the valve side port to the wall suction container. This system maintained a vacuum once the guide needle side holes were in the thorax. The valve was opened for placement of the coaxial biopsy needles and then closed around them to maintain the suction during tissue sampling.
Biopsy Procedure and Imaging
All CT images were reviewed by the same radiologist (F.A.M.), who had 9 years of experience reviewing chest CT images. For this protocol, a pneumothorax was defined as the presence of air between the lung and chest wall as seen at transverse CT. The degree of pneumothorax was not evaluated for this introductory experiment, as our purpose was to observe the presence or absence rather than the severity of a pneumothorax. However, CT images were evaluated to determine if any pneumothorax was greater than 50% of the estimated total hemithoracic volume.
All fabrication and assembly of experimental needle systems, as well as the use of these systems during TNB, was performed by the same operator (F.A.M.), who had 8 years of experience in performing TNB. During each lung biopsy, two pleural punctures were performed and two fine-needle aspirate samples and two core samples were obtained. For the first set of biopsies (n = 10 [one lung in each dog]), the use of the experimental suction guide needle or the standard guide needle was alternated, as was the choice of right or left lung (five right lungs, five left lungs).
After sedation was achieved, each dog was placed in a decubitus position with the biopsy lung side up on the table of a HiSpeed Advantage CT scanner (GE Medical Systems, Milwaukee, Wis). A preliminary nonenhanced scan through the thorax was obtained by using 80 kVp, 100 mA, and 5-mm section increments. A lower lobe lung access site was marked, and the skin was shaved and prepared in a sterile fashion. After the guide needle was placed in the lower lung lobe, an aspirate sample was obtained with a coaxial 22-gauge needle, and a core tissue sample was obtained with a coaxial 20-gauge core needle; interval CT scans were obtained to document pneumothorax. The guide needle was then removed and replaced in the lower lobe, creating a second pleural puncture. The fine-needle aspiration and core sampling were repeated, and all needles were removed. CT was performed immediately after the procedure, and each dog was sedated again and rescanned 1 and 3 hours after each biopsy for detection of delayed pneumothorax.
Two weeks later, the contralateral lungs were sampled for biopsy with this same protocol. For the contralateral lung biopsies, the guide needle selected was the one that had not been previously used in that particular dog. Thus, a total of 20 lung biopsy experiments were performed. Both the experimental and the standard guide needles were used in each dog, and each needle was used in an equal number of right and left lungs.
Statistical Analysis
The McNemar test was used to compare the proportion of intraprocedural pneumothoraces between the set of lungs sampled for biopsy with the suction guide needle and the set of lungs sampled for biopsy with the standard guide needle. A similar comparison was made for the proportion of overall pneumothoraces between the suction and standard guide needle groups. All reported P values are two-sided at a significance level of 5%. Analyses were performed by using SAS software version 8.2 (SAS Institute, Cary, NC).
Assuming a two-sided significance level of 5%, a sample of 10 dogs (10 lung pairs) yields 98% power to detect a 70% difference in proportions of intraprocedural pneumothorax when the proportion of discordant pairs is approximately the same and the method of analysis is a McNemar test of equality of paired proportions.
To predict how many more biopsies would be necessary to indicate statistical significance for overall pneumothorax reduction, we performed the McNemar test of equality of paired proportions with a two-sided significance level of 5%. Results of this test indicated that a sample size of 19 pairs would have 80% power to reveal a difference in proportions of 0.500 when the proportion of discordant pairs was expected to be 0.700. The small cohort in this protocol precluded analysis of significant differences between right and left lungs or the significance of the timing of an intraprocedural pneumothorax as related to number of pleural punctures performed or number of aspirate or core samples obtained.
|
Results
|
|---|
Complications
No mechanical complications were encountered. There were no animal complications other than pneumothorax. None of the pneumothoraces were more than 50% of the estimated total hemithoracic volume as seen at transverse CT. No dog with a pneumothorax showed obvious respiratory signs that necessitated additional therapy. Two weeks elapsed between right and left lung biopsies in each dog, and no prior pneumothoraces were detected at the time of the contralateral lung biopsy.
Pneumothorax Incidence
For the 10 biopsies performed by using the experimental suction guide needle (Fig 3), four (40%) pneumothoraces occurred. Two (20%) of these occurred during the procedure, and two (20%) were detected at postprocedure CT imaging. Nine (90%) pneumothoraces occurred in the 10 biopsies performed by using the standard guide needle (Fig 4); all of these occurred during the biopsy procedure. The Table shows the distribution of biopsies performed by using the two guide needles and the incidence of pneumothoraces.
View larger version (111K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3. Nonenhanced transverse CT image obtained during lung biopsy in dog 1 with the dog in the right decubitus position. The experimental suction guide needle is attached to a wall suction device. No pneumothorax is present.
|
|
View larger version (85K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4. Nonenhanced transverse CT image obtained during lung biopsy performed by using the standard 18-gauge guide needle in dog 2 with the dog in the right decubitus position. Note the presence of a pneumothorax (*).
|
|
Results of McNemar testing indicated that there was no significant difference in the overall rate of pneumothorax between the experimental suction needle and the standard guide needle (P = .125). However, when only intraprocedural pneumothoraces were considered, a significant difference was found between the two types of needles (P < .016), with the suction guide needle producing significantly fewer pneumothoraces during the biopsy procedure.
|
Discussion
|
|---|
Biopsy-related pneumothorax remains a complication that generally cannot be predicted or prevented despite efforts to alleviate it through procedural adjustments and postbiopsy maneuvers. Treatment is usually observation, but sometimes insertion of a chest tube is required to relieve the pulmonary-related symptoms. Because many factors affect the incidence of pneumothorax, including ones that are uncontrollable, it seems logical to concentrate more on removing the positive-pressure air and maintaining the pleural vacuum. It has previously been shown that manual aspiration of a pneumothorax that occurs during a lung biopsy might prevent progression and the subsequent need for chest tube placement (15). Because the suction guide needle produces constant aspiration during a biopsy procedure, the significant reduction in the incidence of pneumothorax that we observed with this technique supports the results of that earlier report. This canine experiment was the first step in a process intended to reproduce similar results in human percutaneous lung biopsies.
Careful attention was devoted to reproducing a clinical situation similar to that of a human lung biopsy performed with conscious sedation. With our canine sedation protocol, we eliminated pulmonary pressure changes associated with endotracheal intubation. Despite this, our pneumothorax rate of 90% with the standard guide needle is believed to be too high compared with the previously cited rate of human lung biopsy–associated pneumothorax (0%–61% [4]). We believe that this difference underscores inherent differences between canines and humans. One such difference is that the hemithoraces in the dog communicate, whereas those in humans do not. This communication increases the risk of a bilateral pneumothorax during a unilateral lung biopsy. In fact, Wittich et al (16) reported that such a communication occurred in a patient who underwent cardiothoracic surgery that created a common pleural cavity.
Although we did not observe bilateral pneumothoraces in a single lung biopsy in our present study, during preliminary studies to determine the most appropriate species to use as an experimental model, a severe unilateral pneumothorax in a dog produced a contralateral pneumothorax, and neither pneumothorax responded to suction. This communication creates an area of uncertainty regarding canine pulmonary physiology and affected the adjustment of several factors in our protocol. For example, normal negative pleural pressure in the canine thorax is difficult to predict. For this reason, we chose to set the wall suction at a maximum pressure of 100 mm Hg. We observed no clinical or imaging evidence that this degree of suction was excessive or damaging to the lung parenchyma. If a lesser degree of suction had been chosen, this choice would have been entirely arbitrary, and we could not have determined whether the occurrence of a pneumothorax during a lung biopsy was due to insufficient suction. However, it is uncertain whether the use of a greater degree of suction than 100 mm Hg would reduce the pneumothorax incidence even more. This will need further investigation, and our better understanding of human pulmonary physiology may lead to improved adjustments and better results in human clinical trials.
Several types of needles have previously been studied in lung biopsies (17). Although many needle types are used to obtain various tissues for diagnosis, no specific needle type has been statistically proved to significantly reduce the incidence of pneumothorax. The results of the present experiment are promising because we significantly reduced the incidence of intraprocedural pneumothoraces by using the suction guide needle. In addition to this, there are potential benefits of the suction guide needle that have not yet been investigated.
In a clinical setting, reducing intraprocedural pneumothoraces could lead to better sampling accuracy for difficult lesions. If a pneumothorax occurs during a biopsy procedure and the lesion falls away from the needle as a result, diagnostic accuracy may be compromised. Keeping the lung fully inflated during the procedure by using a suction needle may remedy this sampling difficulty.
Pneumothoraces were not graded in this protocol, nor was an attempt made to evaluate the development of a pneumothorax related to the number of pleural punctures, aspiration versus core sampling, or the timing of a delayed occurrence at 1 versus 3 hours. The significance of both the number of pleural punctures and the acquisition of an aspirate versus a core tissue sample has been studied in human lung biopsies (2,5–7). Such study results have not been consistent with the importance of these factors, but this also does not mean that uncontrolled differences in these parameters do not affect the results. However, we found that pneumothoraces occurred even with the use of the suction guide needle. This underscores the multifactorial and complicated etiology of a pneumothorax. We speculate that such factors do have an important role in the development of a lung biopsy–associated pneumothorax. With the limiting factors of a small cohort, a fixed number of pleural punctures, and a fixed number of biopsy passes in each experiment, our analysis could not reveal the statistical significance of these limiting factors. Because our protocol did not allow for the placement of chest tubes, grading the severity of a pneumothorax was not necessary. Because in humans the number of pleural punctures varies and the severity of a pneumothorax affects the placement of a chest tube, these issues would be better evaluated in human clinical trials.
A number of limitations affected our data analysis. The operator was not blinded to the type of needle used. Although we do not believe this attributed to a major difference in technique, it does introduce bias. A further selection bias involved the nonrandom selection of right and left lungs, as well as the needle choice for each procedure. Finally, the degree of pneumothorax was not quantified in this experiment. For the purpose of this initial study, we chose to concentrate on the presence or absence of pneumothorax. Of course, in human lung biopsies, the size of the pneumothorax and the presence of tension, not simply the presence of a small pneumothorax, are main causes of morbidity. On the other hand, operators who perform lung biopsies with real-time fluoroscopic guidance may observe small pneumothoraces that prove to be inconsequential. Certainly, the degree of pneumothorax is important to evaluate. This requires further study in subsequent experiments.
Our initial positive results with this experimental guide needle have prompted an application for a provisional patent and permission to undertake the steps required to create a prototype for use in human clinical trials.
|
ACKNOWLEDGMENTS
|
|---|
We thank Marcella M. Johnson, MS, for her contribution to the statistical analysis of the experiment results.
|
FOOTNOTES
|
|---|
Abbreviation: TNB = transthoracic needle biopsy
Authors stated no financial relationship to disclose.
Deceased.
Author contributions: Guarantor of integrity of entire study, F.A.M.; study concepts and design, F.A.M., K.C.W., T.M.L.; literature research, F.A.M.; clinical and experimental studies, F.A.M., K.C.W., T.M.L.; data acquisition and analysis/interpretation, F.A.M., K.C.W., T.M.L.; statistical analysis, F.A.M.; manuscript preparation, definition of intellectual content, revision/review, and final version approval, F.A.M., K.C.W., T.M.L.; manuscript editing, F.A.M., K.C.W.
|
REFERENCES
|
|---|
- Nordenstrom B. A new technique for transthoracic biopsy of lung changes. Br J Radiol 1965; 38:550-553.[Abstract/Free Full Text]
- Wallace MJ, Krishnamurthy S, Broemeling LD, et al. CT-guided percutaneous fine-needle aspiration biopsy of small (
1-cm) pulmonary lesions. Radiology 2002; 225:823-828.[Abstract/Free Full Text]
- Zwischenberger JB, Savage C, Alpard SK, Anderson CM, Marroquin S, Goodacre BW. Mediastinal transthoracic needle and core lymph node biopsy: should it replace mediastinoscopy? Chest 2002; 121:1165-1170.[Abstract/Free Full Text]
- Laurent F, Montaudon M, Latrabe V, Begueret H. Percutaneous biopsy in lung cancer. Eur J Radiol 2003; 45:60-68.[CrossRef][Medline]
- Swischuk JL, Casteneda F, Patel JC, et al. Percutaneous transthoracic needle biopsy of the lung: review of 612 lesions. J Vasc Interv Radiol 1998; 9:347-352.[Medline]
- Ko JP, Shepard JO, Drucker EA, et al. Factors influencing pneumothorax rate at lung biopsy: are dwell time and angle of pleural puncture contributing factors? Radiology 2001; 218:491-496.[Abstract/Free Full Text]
- Yeow KM, See LC, Lui KW, et al. Risk factors for pneumothorax and bleeding after CT-guided percutaneous coaxial cutting needle biopsy of lung lesions. J Vasc Interv Radiol 2001; 12:1305-1312.[Medline]
- Quon D, Fong TC, Mellor J, Brandschwei FH, Desautels JE. Pulmonary function testing in predicting complications from percutaneous lung biopsy. Can Assoc Radiol J 1988; 39:267-269.[Medline]
- Anderson CL, Crespo JC, Lie TH. Risk of pneumothorax not increased by obstructive lung disease in percutaneous needle biopsy. Chest 1994; 105:1705-1708.[Abstract/Free Full Text]
- Kazerooni EA, Lim FT, Mikhail A, Martinez FJ. Risk of pneumothorax in CT-guided transthoracic needle aspiration biopsy of the lung. Radiology 1996; 198:371-375.[Abstract/Free Full Text]
- Rozenblit AM, Tuvia J, Rozenblit GN, Klink A. CT-guided transthoracic needle biopsy using an ipsilateral dependent position. AJR Am J Roentgenol 2000; 174:1759-1764.[Abstract/Free Full Text]
- Petsas T, Siamblis D, Giannakenas C, et al. Fibrin glue for sealing the needle track in fine-needle percutaneous lung biopsy using a coaxial system. II. Clinical study. Cardiovasc Intervent Radiol 1995; 18:378-382.
- Public health service policy on humane care and use of laboratory animals. National Institutes of Health Office of Extramural Research Web site. http://grants.nih.gov/grants/olaw/references/phspol.htm. Accessed April 5 2005.
- Guide for the care and use of laboratory animals chapter 2: animal environment, housing, and management. http://oacu.od.nih.gov/ac_cbt/guide3.htm. Accessed April 5 2005.
- Yamagami T, Nakamura T, Kato T, Nishimura T. Management of pneumothorax after percutaneous CT-guided lung biopsy. Chest 2002; 121:1159-1164.[Abstract/Free Full Text]
- Wittich GR, Kusnick CA, Starnes VA, Lucas DE. Communication between the two pleural cavities after major cardiothoracic surgery: relevance to percutaneous intervention. Radiology 1992; 184:461-462.[Abstract/Free Full Text]
- Grode G, Faurschou P, Milman N. Percutaneous transthoracic fine-needle lung biopsy with 3 different needles: a retrospective study of results and complications in 224 patients. Respiration 1993; 60:284- 288.[Medline]
TOP
ABSTRACT
INTRODUCTION
