Transbronchial Biopsy and Needle Aspiration

Transbronchial Biopsy and Needle Aspiration

----4~IIIII.....--b-ro-nc-h-os-co-py---------Transbronchial Biopsy and Needle Aspiration Deborah Shure, A/.D., F.C.C.R* Fiberoptic bronchoscopy has g...

2MB Sizes 0 Downloads 41 Views

----4~IIIII.....--b-ro-nc-h-os-co-py---------Transbronchial Biopsy and Needle Aspiration Deborah Shure, A/.D., F.C.C.R*

Fiberoptic bronchoscopy has greatly expanded the diagnostic capability of bronchoscopy over that provided by the rigid bronchoscope. Much of this expansion has been related to the ability to use transbronchial sampling with greater precision and safety. For many years no\\; transbronchial forceps biopsy, first pioneered as a technique for rigid bronchoscopyl and as a radiologic technique, has been a standard diagnostic procedure. More recently, a number of aspiration needles have become available for use through the fiberoptic bronchoscope, which have further expanded the diagnostic range of bronchoscopy, providing enhanced access to peripheral as well as to submucosal and mediastinal lesions. Although diagnosis of cancer is perhaps the most common application of these techniques, they can be applied to a wide range of pathologic processes. In many situations, bronchoscopy with transbronchial biopsy can play an important role in the diagnosis of infections. Application to noninfectious benign diseases is also possible, although it may be more limited by the experience of the cytopathologist. This review covers the instrumentation needed for transbronchial sanlpling, the techniques employed in transbronchial biopsy and transbronchial needle aspiration, and the applications of these procedures. Their limitations and possible corrections of these limitations are also explored. EQUIPrvfENT


Bronchoscopic forceps have changed only slightly since their introduction almost 20 years ago. They are hinged cups with either a cutting edge or a serrated edge (alligator forceps). Most plain cup forceps today have a window in each cup to limit crushing the specimen, although it has never been demonstrated that this modification is effective in limiting crush artifact. While these forceps have a cutting edge, in *Associate Professor of ~fedicine, University of California, San Diego, and Chief, Division of Pulmonarv and Critical Care Medicine, Veterans Administration ~1edical Center, San Diego. Reprint requests: Dr. Shure. Pu11nonary Division 111-]. VA ~\ledical Center, 3350 fA Jolla Village Drive, San Diego 92161 1130

practice tissue is often torn rather than cleanly cut in the course of transbronchial biops)': This tearing action does not appear to have any clinically deleterious effect. Alligator forceps are most useful for grasping foreign bodies, but they can also be used for transbronchial biopsy. A possible, but unproved, disadvantage of alligator forceps for transbronchial biopsy is their lack of a cutting edge. The entire action is a tearing one. In the past, the cups of alligator forceps were significantly larger than those of plain forceps, but the diagnostic yield was not higher using them despite the increase in specinlen size. In fact, \vhile both cup and alligator forceps come in t\\,O sizes to fit large- and slnall-channel bronchoscopes, the diagnostic yield is similar (unpublished data). Accuracy in salnpling the lesion and the number of sanlples taken appear to be more important than the actual sample size. As a result, plain cup forceps of either size are appropriate instruments for transbronchial biopsy. Needles

A nunlber of bronchoscopic aspirating needles are now available in sizes ranging fronl 18 to 23 gauge. The basic design of most needles is sinlilar, consisting of a metal needle attached to a catheter that is housed in an outer sheath into \vhich the needle and its catheter can be withdrawn. Some devices have obturators whose major function is to provide stiffening for the catheter. An obturator is probably not necessary for needle occlusion to prevent sampling ofintervening tissue, as it is for transthoracic aspiration, since the needle generally passes only through the relatively thin bronchial or tracheal wall. A fe\\r needles that are not retractable use a blunt-tipped obturator protruding from the needle to prevent the needle fronl danlaging the bronchoscope channel. Nevertheless, these nonretractable devices probably still pose sonle risk of channel dalnage. Needles also vary in caliber and length. For sanlpIing peripheral nlasses and hilar and subcarinal nodes, a length of 1 cnl is useful. Shorter needles (0.5 cnl) can be used to sanlple subnlucosal tissue, but the l-cm length is also appropriate for this application. Transbronchial Biopsy and Needle Aspiration (Deborah Shure)

Longer needles, at least 1.3 cm, are best for transtracheal aspiration because of the oblique angle often involved in this technique. Longer needles (greater than 1 cm) are not practical, however, for peripheral applications, because the longer rigid distal section limits maneuverability of the device through the bronchial branches. For most applications, the gauge should be at least 20 to 21. Thinner needles often provide insufficient specimens. Eighteen- or 19-9auge needles have recently become available and appear to be useful in adding cores of tissue to the aspiration specimen. 2 The ability to combine histologic with cytologic examination usually increases the diagnostic yield. Despite the larger size, these needles do not appear to increase the risk of the procedure. One unique needle made entirely of plastic appears to be particularly useful for peripheral applications. It consists of an outer sheath and an inner catheter with a sharpened, beveled tip. The inner catheter can be withdrawn into the sheath during passage through the bronchoscope. The advantage of this needle lies in its lack of the rigid section that is characteristic of both forceps and needles. As a result, these needles can often be passed along sharper bends than metal needles or forceps. This property can be particularly useful in sampling upper lobe apical segment and lower lobe superior segment lesions, although it is useful in other areas as well. An additional advantage of this needle is its 18-gauge caliber, which often allows for obtaining histologic specimens in addition to cytologic ones. Its disadvantage is that it is not optimally radiopaque in its current version. As a result, it can sometimes be difficult to see fluoroscopically. Some general precautions are necessary with all needles to prevent damage to the bronchoscope channel. The needle should always be retracted into its sheath during passage through the channel. The needle should be advanced only after the distal end of the sheath is visible through the bronchoscope. With some devices, a detent keeps the needle in either the retracted or advanced position. With others, it is important for the operator to ensure that the needle remains in the proper position by fixing the needle catheter at its proximal end during passage through the channel. It is also advisable to pass the needle device through the channel with little or no distal tip deflection of the bronchoscope. The bronchoscope can be flexed as needed once the distal end of the sheath is out of the channel. Precautions are also necessary to prevent damage to some needle devices. Needles can penetrate the outer sheath of some devices that have plastic sheaths, usually because the outer sheath becomes kinked during passage through the bronchoscope. Two precautions can be preventive. One is passing the needle

with the bronchoscope fairly straight until the distal end of the sheath can be seen. The other simultaneous precaution is to place the tip of the needle within a millimeter ofthe distal end ofthe sheath. Ifthe needle is too far back in the sheath and the channel is not straight, the tubing of the sheath can become bent and provide a focus for penetration of the needle when attempting to advance it from the sheath.

Curet An uncommonly used device for sampling peripherallesions is the curet. These have been available in both single-hinged and double-hinged models, although only double-hinged curets are now commercially available. The curet consists of a shallow; elongated cup with sharp edges. The device can be flexed at its hinges, allowing some steerability not available with current forceps or needles. The curet must be straight as it is passed through the bronchoscope channel. Care must also be taken to avoid pushing the curet against an obstruction, since the device is more fragile than standard forceps. The specimen is obtained by flexing the curet hinge and pulling the curet along the lesion to scrape off cells. The curet must be straightened again before withdrawing it from the bronchoscope. Like most aspirating needles, the curet obtains a cytologic specimen. TECHNIQUES

Transbronchial Biopsy Transbronchial biopsy is a simple procedure. It is helpful to advance the forceps slightly from the end of the bronchoscope so that they can be accurately placed in the segmental or subsegmental bronchus of choice. They are then passed until beyond view; after which fluoroscopy is used to guide the forceps into the lesion. For mass lesions, the position of the forceps in the lesion can be confirmed by biplane fluoroscopy, rotating the arm of a C-arm fluoroscope, or moving the patient. For the latter two techniques, one looks to see if the forceps move with or away from the lesion. If the forceps and the lesion move together, they are in the same plane. Once in the lesion, the forceps can be opened, advanced slightly, depending on the lesion's size, closed, and withdrawn. Sufficient tissue is pushed into the forceps cups by advancing the forceps in the lesion. A technique of timing the biopsy to the respiratory cycle has been advocated. 3 In this technique, the forceps are opened, the patient inhales and exhales deeply, and the forceps are closed at endexpiration. The concept of this technique is that tissue will invaginate into the forceps cups during exhalation, producing a larger tissue sample. A study of transbronchial biopsies done during inhalation and exhalation, however, found that timing the biopsy to exhalaCHEST I 95 I 5 I MA'(, 1989


tion made no difference in the number of alveoli obtained or in the overall specimen area. 4 If the bronchoscope tip is flexed during the biopsy, it should be straightened before removing the forceps to avoid damage to the bronchoscope. It is important, however, for the bronchoscope to remain close to the segment or subsegment from which the biopsy was taken to control potential bleeding. Wedging the bronchoscope in the bronchus will serve this purpose, but it is probably not necessary to actually wedge the bronchoscope as long as the distal tip can effectively be moved to block the bronchus, if necessary. A single report of air embolism during transbronchial biopsy suggested that embolization was related to the increased pressure resulting from wedging the bronchoscope. 5 This complication is certainly rare, but it remains a consideration. Patient comfort might be more relevant. Many patients do not easily tolerate wedging the bronchoscope, even with topical anesthesia. Again, the important principle is to maintain control of the bronchus to quickly prevent blood from spreading to other regions. The use of fluoroscopy in transbronchial biopsy is the subject of occasional controvers~ Fluoroscopic guidance clearly increases the diagnostic yield in focal lesions, 6 but transbronchial biopsy can be performed without Huoroscopic guidance in diffuse lung disease. A careful technique is necessary, though, with blind transbronchial biops~ The forceps are passed with the cups closed until resistance is met or the patient expresses pain, which indicates that the pleura has been reached. The forceps are then withdrawn 1 to 2-em, the cups opened, and the forceps advanced slightly and closed} In many studies, fluoroscopic guidance appears to be associated with a lower incidence ofpneumothorax. In one large survey; performing transbronchial biopsy without fluoroscopic guidance resulted in a 60 percent greater incidence of pneumothorax. 7 More recently; it has been suggested that fluoroscopy is not necessary in diffuse disease and that the incidence of pneumo6,8 One thorax is the same with or without Huoroscop~ such study had a particularly high incidence of pneumothorax both with or without fluoroscopy; 8 which makes the conclusions difficult to evaluate. Since the risk of blind transbronchial biopsy has been variable, it is probably safest to use fluoroscopic guidance. Iffluoroscopy is not available, the technique should be used only by operators who are comfortable with the approach and have acceptable complication rates. It should also probably be limited to biopsy of the lower lobes, since blind transbronchial biopsy of other areas appears to have a higher incidence of pneumothorax. Transbronchial biopsy does increase the risk of bronchoscop~ The major associated risks are bleeding 1132

and pneumothorax. In one study, transbronchial biopsy tripled the mortality rate associated with bronchoscopy (from 0.04 to 0.12 percent) and increased the complication rate 22-fold7 (from 0.12 to 2.7 percent). While the morbidity and mortality remain lo~ it is important to recognize the increased risk in order to practice necessary precautions. In general the risk of pneumothorax should be low (less than 3 percent); however, mechanical ventilation appears to increase this risk. In one study that examined this issue, transbronchial biopsy was associated with a 7 percent incidence of tension pneumothorax. 9 This finding suggests that transbronchial biopsy should be performed with circumspection in this group of patients and that one should be prepared to treat an emergency complication. Bullous emphysema probably also increases the risk ofpneumothorax, although this issue has not been systematically examined. Some authors also suggest that patients with acquired immunodeficiency syndrome (AIDS) and Pneumocystis carinii pneumonia have an increased incidence of pneumothorax with biopsy based on an increased incidence of spontaneous pneumothorax, 8 but the data supporting an increased incidence of pneumothorax in patients with Pneumocystis pneumonia and AIDS are not well supported yet. 10 Life-threatening bleeding is unusual with transbronchial biops~ When it occurs in a patient without an underlying coagulation disorder, it usually results from the unpredictable biopsy ofa sizable pulmonary artery or a bronchial artery. II Small to moderate amounts of bleeding (less than 100 ml) are more common3 and, although controllable, bleeding is to be avoided if possible. With this idea in mind, mechanical ventilation, coagulation disorders, and pulmonary hypertension all pose an increased risk of bleeding from transbronchial biopsy. Mechanical ventilation is associated with a 20 percent incidence of significant bleeding. 9 Thrombocytopenia with preoperative platelet transfusion is associated with a 16 percent incidence. I2 While pulmonary hypertension has long been thought to be a risk factor for bleeding, a recent study has documented this risk, finding a 15 percent incidence of significant bleeding in patients with mean pulmonary artery pressure over 16 mm Hg in contrast to no significant bleeding in a similar group (status postcardiac transplantation) without elevated mean pulmonary artery pressure. 13 Finally, many bronchoscopists think that patients with AIDS have an increased tendency to bleed with brushing or biopsy, although this risk has not been explored or documented. Despite the increased risk of transbronchial biopsy over simple bronchoscopy in patients without specific risk factors for bleeding or pneumothorax,7 it appears to be a safe procedure on an outpatient basis in many nansbronchial Biopsy and Needle Aspiration (Deborah Shure)

patients. Suitable patients should not have significant risk factors for bronchoscopy in general and for bleeding and pneumothorax in particular. Additionally, such patients should be willing and able to be followed up closely on an outpatient visit and to return to the hospital if necessary.

Transbronchial Needle Aspiration The technique of transbronchial needle aspiration depends in part on the site being aspirated. In general, the needle, but not its sheath, should be advanced from the end of the bronchoscope .14,15 The channel of the bronchoscope can be used to support the needle and prevent the tubing from kinking. For endobronchial and mediastinal applications, the needle should be secured in the site before jabbing it further into the tissue of interest. In the trachea, one must be careful to place the needle between tracheal rings at as perpendicular an angle as possible,14 although this approach can be difficult. Preventing the needle from advancing too far from the distal end of the bronchoscope can be helpful. In addition, asking the patient to cough gently can help to impale the needle in the tissue. 16 For peripheral lesions, transbronchial needle aspiration is performed similarly to transbronchial biopsy using fluoroscopic guidance .17 For any approach, the needle should be withdrawn into its outer sheath prior to withdrawing it through the bronchoscope channel. In addition, multiple aspirations (at least three) of a given site probably help to decrease sampling error. 15 Since the cumulative experience with transbronchial needle aspiration is much smaller than that with transbronchial biopsy, their complication rates are difficult to compare. Only two reports of pneumothorax with mediastinal aspiration exist,16 and no serious bleeding has been reported. In my own experience, we have not had a case of pneumothorax or significant bleeding with mediastinal, endobronchial, or peripheral transbronchial needle aspiration. This experience is similar to the community hospital experience as well. 18 Transbronchial needle aspiration so far appears to be a safe procedure, possibly safer than transbronchial biopsy, but time will further test this hypothesis. ApPLICATIONS

Cancer Transbronchial biopsy can be useful in the diagnosis of both primary and metastatic peripheral lung cancers, while transbronchial needle aspiration can, in addition, be useful in the diagnosis of submucosal, peribronchial, and mediastinal tumor. Transbronchial biopsy has been the mainstay of the bronchoscopic diagnosis of peripheral masses, although its limitations are well known. The diagnostic yield of transbronchial biopsy has been variable l 9-22 (30 to 80 percent), aver-

aging about 50 percent. The variation in yield probably depends on the incidence of the factors that have been found to affect yield. The size of a peripheral mass is perhaps the most important such factor. Most series find a low yield with lesions less than 2 cm in diameter (15 to 35 percent).17,20.22-24 Yield also falls off with very large lesions (greater than 6 cm in diameter),20 probably because of the sampling error created by necrosis and inflammation in such tumors. Distance from the hilum is a variable factor. In some studies, the diagnostic yield was lowest with inner third lesions,20 in others with outer third lesions,23 and in some, location was not a significant factor. 24 These findings may have been confounded by not separating the effect of size from that of location. Other factors influence the yield of transbronchial biopsy. The yield appears to be higher for primary bronchogenic carcinoma than for metastatic nodular carcinoma,22.25,26 possibly because of the vascular origin of metastatic lung tumors. Transbronchial biopsy appears most suited to sampling processes of a bronchial origin. The number of samples taken from a lesion also influences the yield. In one study the yield continued to increase through the sixth biopsy.27 Others have found that the maximal yield is achieved in four biopsies (unpublished data). In either case, more biopsies are needed to diagnose peripheral masses than visible endobronchial tumors. 28 The number of sections examined from each biopsy specimen may also be important. In general, it is useful to examine multiple sections. 27 Washings or brushings also increase the yield over transbronchial biopsy alone. 17 Recently, transbronchial needle aspiration has been found to further increase the diagnostic yield in peripheral cancers. 17,29 Transbronchial needle aspiration has the same limitation of lesion size as transbronchial biopsy, but for lesions 2 cm or larger in diameter, transbronchial needle aspiration provides a 50 percent greater yield than transbronchial biopsy.17 Transbronchial needle aspiration appears to be particularly useful in the subset of patients with peripheral masses that cannot be penetrated by standard forceps, possibly because the bronchus leading to the mass is compressed extrinsically by the tumor itself. The needle aspiration technique detects at least 80 percent of such lesions, which are inaccessible to transbronchial biopsy. In addition, transbronchial needle aspiration appears to be preferable in the diagnosis of metastatic nodules, possibly because of its ability to access tissue beyond the peribronchial region. Based on this experience, a case could be made for performing transbronchial needle aspiration with either brushing or washing (or lavage) without transbronchial biopsy for peripheral lung cancers. Using an I8-gauge needle, such as the plastic one described CHEST I 95 I 5 I MA'f, 1989


earlier for histologic specimens on lesions 2 cm or greater in diameter, yields as high as 85 percent can be obtained with this approach. This result is appealing as well because of the lower incidence of bleeding. In one stud~ the largest bleed from needle aspiration was 2 ml, while the largest from transbronchial biopsy was 30 ml. 17

One technique has been described using the curet for peripheral masses less than 2 cm in diameter with a striking yield of 95 percent,30 well beyond most results reported with any technique, including transthoracic needle aspiration. This procedure was adapted from a similar nonbronchoscopic technique using Metras catheters. The bronchial path to the lesion is first outlined with selective bronchograph)'Approximately two weeks later, to allow the inflammatory effects of bronchography to subside, bronchoscopy is performed, and a double-hinged curet is guided into the lesion using the bronchogram as a map. Repeated curettage is performed until a suitable specimen is obtained. Although the yield with this procedure is dramatically better than with most other techniques, one could easily argue that it is probably not worth the increased time, discomfort, and cost inevitably involved. The success of the technique, however, does point out the need for better guidance of sampling devices into peripheral lesions. Moving from the periphery to the larger airways, transbronchial needle aspiration has opened up new diagnostic possibilities in detecting lung cancer. Submucosal spread of tumor and extrinsic compression by peribronchial tumor have been difficult to detect on standard forceps biopsy, in part because the tissue can be too hard to biopsy adequately and in part because the process may be too deep to be reached by standard biopsy forceps. In these situations, transbronchial needle aspiration has a higher yield than forceps biopsy31 (71 vs 55 percent). Unlike the case with peripheral masses, however, both are clearly worth doing, since the combined yield is 90 percent, and adding either washing or brushing increases the yield still further, to 97 percent. Thus, with this difficult diagnostic situation, forceps biopsy, needle aspiration, and brushing are all necessary to maximize the yield. Perhaps the most exciting application of transbronchial needle aspiration has been in the detection of mediastinal spread of cancer, because it offers the possibility of avoiding the cost and morbidity of surgical staging in selected patients whom surgery is unlikely to benefit. Needle aspiration can access paratracheal, subcarinal, aorticopulmonary windo~ and hilar nodes. 32•33 Detection of tumor in the hilum may be most useful in the diagnosis of small cell carcinoma, which is often central and difficult to detect endobronchiall)'- The significance of tumor in other areas of the mediastinum has been a matter of 1134

recent controvers)'- Some groups think that limited tumor in the paratracheal area may still have an improved prognosis with surger~ particularly with squamous cell carcinoma, although this experience is still not wide enough to be generally applicable. If surgery is contemplated in stage III disease, needle aspiration cannot provide the necessary information on nodal structure (intranodal or perinodal disease), and surgical staging becomes necessary. In most reports, however, widespread mediastinal spread and disease in the subcarinal nodes still carry a very poor prognosis not warranting surgery. In this context, needle aspiration can be particularly helpful. If computed tomograms (CT) are available, they can be used to determine the need for transcarinal (or paratracheal) needle aspiration. 16 If CT is not available before the initial diagnostic bronchoscopy, transcarinal needle aspiration is worthwhile when the primary lesion is endobronchial (25 percent positive) or when the carina appears broad33 (40 percent positive). In our experience, transcarinal needle aspiration is rarely positive when the primary tumor is peripheral and the carina appears sharp. Using these criteria (endobronchial tumor or abnormal carina) or using CT guidance, 50 to 60 percent of malignant mediastinal nodes can be detected with an extremely low false positive rate, if necessary precautions are observed. 16.33.34 Two false positive paratracheal aspirates have been reported,16 and both serve to emphasize some of the measures necessary to ensure accuracy of the procedure. In one case, a right paratracheal aspirate, which was malignant, turned out to be a tumor in the medial aspect of the right upper lobe abutting the trachea rather than a paratracheal node. This situation highlights the importance of being certain of the anatomy of the area being aspirated. Since parenchymal lung tissue does not occur in the subcarinal area, this factor is not a concern in subcarinal aspirates. In the second case, the aspirate contained a few scattered malignant cells with bronchial epithelial cells and without lymphocytes. This aspirate was properly identified as a contaminant by the cytopathologist, although it was reported as a false positive for purposes of the stud)'In practice, improperly cellular aspirates should not be interpreted. Other precautions to avoid false positive results center around avoiding contamination by the primary tumor.14.15.33 Aspiration of the mediastinum should be done before examination of the lower airways. The mid-portion of the carina should be aspirated from the side opposite the primary lesion, and the aspirate should not be performed if the site is covered by lower airway secretions, since the risk of contamination is too high. In all cases, if the clinical situation is at odds with the result of the aspiration, the patient should be Transbronchial Biopsy and Needle Aspiration (Deborah Shure)

given the benefit of the doubt and allowed to have conventional surgical staging. No technique should be stretched beyond its reasonable limits. A negative needle aspiration, on the other hand, does not imply the absence of tumor. False negative results are not worrisome, though, since the patient can then undergo surgical staging as indicated. The addition of histologic specimens2 with larger-bore (18gauge) needles may lower this false negative rate in the future. Despite the false negative rate, performing this relatively simple procedure during the initial diagnostic bronchoscOpy can result in a considerable cost savings as well as in avoiding the morbidity of surgical procedures. In one study, transcarinal needle aspiration provided the only presurgical evidence of unresectability in 70 percent of those in whom it was positive. 33 In addition, the detection of small cell carcinoma may direct some patients to appropriate chemotherapy. Unlike mediastinal aspiration, the role of transbronchial needle aspiration in the diagnosis of frank endobronchial tumor is still not clear. One study35 found it has a higher yield than forceps biopsy (80 vs 67 percent), while another36 found the reverse (65 vs 85 percent). Since the diagnostic yield on forceps biopsy is usually so high27.28 (close to 100 percent) and offers the advantage of providing a histologic specimen, the only role of transbronchial needle aspiration in endobronchial masses may be in avoiding excessive bleeding from biopsy of friable lesions.37

Infection The role of transbronchial biopsy in infectious processes has varied greatly, apparently related to the patient population. Important factors appear to be the acuteness of the process and the patient population reflecting the incidence of different infectious processes. In general, transbronchial biopsy has a higher yield in acute infectious processes excluding bacterial pneumonia. Because of upper airway contamination, transbronchial biopsy is not a specific technique in the diagnosis of bacterial pneumonia. 38 Transbronchial needle aspiration suffers a similar drawback. A protected aspirating needle similar to the protected specimen brush has been evaluated recently and found to have similar, but not superior, sensitivity and specificity to those of the protected specimen brush. 39

In my own experience, the transbronchial needle aspiration does not obtain adequate samples in a diffuse infiltrative process and appears to be more suited to sampling dense lesions. This limitation also applies to transthoracic needle aspiration, which usually has a low yield in the diagnosis of diffuse infiltrates. 40 Transbronchial biopsy has been found to be very useful in the diagnosis of a number of nonbacterial infections. Several studies have found it to be helpful

in the diagnosis of sputum-negative tuberculosis. 41.42 Histologic examination can provide an early diagnosis in one third to two thirds of cases, and culture of bronchoscopic specimens brings the diagnostic yield to 95 percent. Similar results have been reported for coccidioidomycosis. 43 Curiously, in all of these series, coincidental, unexpected carcinomas were found in about 10 percent of the patients at bronchoscop~ Traditionally, transbronchial biopsy has not been as useful in the diagnosis of invasive aspergillosis. In these cases, the diagnostic yield has been 20 to 50 percent. 13.44 •45 This low yield probably reflects the difficulty sampling perivascular areas and is similar to the situation of metastatic cancer. Transbronchial biopsy also appears to have a low yield in the diagnosis of cytomegalovirus pneumonia,13.45 although the reason for this low yield is not known. Considerable data exist demonstrating a high diagnostic yield (90 to 95 percent) of transbronchial biopsy in Pneumocystis carinii pneumonia. 13.4.5 The unfortunately increasing experience with infections in patients with AIDS has served to highlight this yield. 46 In this area, bronchoalveolar lavage appears to be able to replace transbronchial biopsy. Lavage using 100 to 200 ml saline solution is capable of similar diagnostic yields (90 to 95 percent) with less risk of bleeding and pneumothorax. Lavage may have a similar role in other infectious processes. Recent studies have found it to have a better diagnostic yield in suspected tuberculosis47 (90 percent), invasive aspergillosis I3 .45 (75 percent), and cytomegalovirus infection 13 (64 percent). The improved yield in these diseases makes it a very good diagnostic procedure in many settings in which transbronchial biopsy poses an increased risk, such as cardiac transplantation patients 13 and patients with thrombocytopenia. 12 Since the infections that lavage is capable of detecting are common in patients with AIDS (Pneumocystis and tuberculosis), lavage is also an appropriate first procedure in these patients. Occasional debates have occurred about the relative merits of open lung biopsy and transbronchial biopsy in the diagnosis of acute infiltrative processes (local or diffuse). Some studies have found open lung biopsy has a better yield than transbronchial biops~ 48-50 Others have found that open lung biopsy does not detect significantly more treatable diagnoses after a negative transbronchial biopsy.51.52 The difference in these results probably depends on the patient populations. For example, since patients with AIDS tend to have infections easily detected on transbronchial biopsy (or lavage), open lung biopsy is unlikely to significantly improve the diagnostic yield. On the other hand, patients immunosuppressed by hematologic malignancies or chemotherapy tend to have a higher incidence of viral infections, invasive aspergillosis, leukemic infiltrates, and toxic reactions to drugs, CHEST I 95 I 5 I MA~



all of which are more easily detected by open biopsy. The decision to perform open biopsy or bronchoscopy can be based in part, then, on the patient population. The urgency of obtaining a diagnosis may also be a factor, as may be the patient's clinical status. If the patient's status is deteriorating quickly, open biopsy may be preferable, particularly in patient groups in whom a bronchoscopic diagnosis is less likel~ Open biopsy can also be considered if the risk of bleeding is significant. It should be remembered, however, that open biopsy is not without risk itself. The complication rate is about 10 percent, including such problems as pneumothorax, hemothorax, and empyema. 48 In most clinical situations of fever and pneumonia in immunocompromised patients, bronchoscopy with lavage with or without transbronchial biopsy will be a reasonable first procedure. Open biopsy can then be considered if the results of bronchoscopy are negative and the clinical situation warrants further exploration. I nterstitial Diseases

Transbronchial biopsy is also useful in several noninfectious, chronic interstitial diseases. It is capable of detecting sarcoidosis,53,54 lymphangitic carcinoma, 55 pulmonary alveolar proteinosis, Goodpasture's disease,l.55 Wegener's granulomatosis, and eosinophilic granuloma. The role of transbronchial biopsy in the diagnosis of sarcoidosis has been studied extensivel~ Transbronchial biopsy can detect most cases (97 percent) of sarcoidosis if four to six biopsies are performed. 53.54 Transbronchial needle aspiration is also capable ofdetecting sarcoidosis in mediastinal nodes. 56 Larger-bore needles may facilitate this diagnosis in the future. Eosinophilic granuloma can be diagnosed by routine histology as well as by finding characteristic X bodies on electron microscopic examination. 57 Because of these kinds of findings, it can be useful to process at least one specimen from such bronchoscopies for electron microscop~ X bodies can also be detected in bronchoalveolar lavage specimens, adding yet another role for lavage. 57 The overall role of transbronchial biopsy and its relation to open lung biopsy have been disputed in chronic interstitial diseases as in acute infiltrative processes. Again, the reported yields of transbronchial biopsy and open lung biopsy have varied. 58 •59 Much of the variation relates to the underlying disease processes in a given population and to the reporting of nonspecific interstitial inflammation as a positive diagnosis. To summarize, a great deal of controversy, transbronchial biopsy appears to be a reasonable procedure in detecting the diseases mentioned above. It is less reliable in accurately categorizing nonspecific pneumonitis or fibrotic processes (which are difficult to biopsy adequately). 1136

A clinical case can be made for watching patients with nonspecific pneumonitis on transbronchial biopsy or treating presumed interstitial pneumonia with a therapeutic drug trial without a specific diagnosis in an appropriate clinical setting. This approach is based on the finding that 75 percent of patients with diffuse disease without a specific diagnosis on transbronchial biopsy have a benign course. 59 On the other hand, open lung biopsy can detect significant processes missed by transbronchial lung biopsy. In one study, such diseases as usual interstitial pneumonitis, eosinophilic granuloma, and sarcoidosis were missed by transbronchial biopsy and detected on open lung biopsy.58 In practice, the choice of transbronchial or open biopsy probably depends on the clinical setting as well as on local results with each procedure. SUMMARY

Transbronchial biopsy and transbronchial needle aspiration greatly increase the utility of bronchoscopy in the diagnosis of a variety of disease processes. Transbronchial needle aspiration has brought into focus the importance of good cytopathologic support. 16.33 The addition ofhistologic specimens (for light and electron microscopy) with the newer large-bore needles may further increase the utility of transbronchial needle aspiration. 2 Both techniques are limited, in part, by the lack of distal tip deflection of the sampling instrument for steering accurately to peripheral masses. Tip deflection may have been partly responsible for the good yields reported for the double-hinged curet on small nodules, although the bronchographic map was also a factor. 28 A steerable brush60 was described several years ago, but it was somewhat difficult to accurately maneuver, and longterm results were never reported. As yet, no easy answer is available for this problem. In the future, new generations of ultrathin bronchoscopes may permit much more accurate placement of sampling devices in the periphery of the lung and will represent an exciting diagnostic advance. REFERENCES 1 Andersen HA, Fontana RS, Harrison EG. Transbronchoscopic lung biopsy in diffuse pulmonary disease. Dis Chest 1965; 48:187-92 2 Wang KE Flexible transbronchial needle aspiration biopsy for histologic specimens. Chest 1985; 88:860-63 3 Zavala DC. Pulmonary hemorrhage in fiberoptic transbronchial biopsy. Chest 1976; 70:584-88 4 Shure 0, Abraham JL, Konopka R. How should transbronchial biopsies be performed and processed? Am Rev Respir Dis 1982; 126:342-43 5 Erickson AD, Irwin RS, Teplitz C, Corrao WM, Tarpey JT. Cerebral air embolism complicating transbronchoscopic lung biopsy. Ann Intern Med 1979; 90:937-38 6 Anders GT, Johnson JE, Bush BA, Matthews JI. Transbronchial biopsy without fluoroscopy: a seven-year perspective. Chest 1988; 94:557-60 TransbronchiaJ Biopsy and Needle Aspiration (Deborah Shure)

7 Simpson FG, Arnold AG, Purvis A, Belfield P~ Muers MF, Cooke NJ. Postal survey ofbronchoscopic practice by physicians in the l!nited Kingdom. Thorax 1986; 41:311-17 8 Milligan SA, Luce JM, Golden J, Stulbarg M, Hopewell PC. Transbronchial biopsy without fluoroscopy in patients with diffuse roentgenographic infiltrates and the acquired immunodeficiency syndrome. Am Rev Respir Dis 1988; 137:486-88 9 Papin TA, Grum CM, Weg JG. Transbronchial biopsy during mechanical ventilation. Chest 1986; 89: 168-70 10 Goodman PC, Daley C, Minagi H. Spontaneous pneumothorax in AIDS patients with Pneumocystis carinii pneumonia. AJR 1986; 147:29-31 11 Flick MR, Wasson K, Dunn LJ, Block AJ. Fatal pulmonary hemorrhage after transbronchiallung biopsy through the fiberoptic bronchoscope. Am Rev Respir Dis 1975; III :853-56 12 Papin TA, Lynch JP III, Weg JG. Transbronchial biopsy in the thrombocytopenic patient. Chest 1985; 88:549-52 13 Schulman LL, Smith CR, Drusin R, Rose EA, Enson Y, Reemtsma K. Utility of airway endoscopy in the diagnosis of respiratory complications ofcardiac transplantation. Chest 1988; 93:960-67 14 Wang K~ Haponik EF, Gupta PK, Erozan Y. Flexible transbronchial needle aspiration: technical considerations. Ann Otol Rhinoll984; 93:233-36 15 Shure D. Fiberoptic bronchoscopy-diagnostic applications. Clin Chest Med 1987; 8:1-13 16 Schenk DA, Bower JH, Bryan CL, Currie RB, Spence TH, Duncan, CA, et al. Transbronchial needle aspiration staging of bronchogenic carcinoma. Am Rev Respir Dis 1986; 134:146-48 17 Shure D, Fedullo PF. Transbronchial needle aspiration of peripheral masses. Am Rev Respir Dis 1983; 128:1090-92 18 Harrow EM, Oldenburg A, Smith AM. Transbronchial needle aspiration in clinical practice. Thorax 1985; 40:756-59 19 Ellis JH Jr. Transbronchial lung biopsy via the fiberoptic bronchoscope: experience with 107 consecutive cases and comparison with bronchial brushing. Chest 1975; 68:524-32 20 Cortese DA, McDougall JC. Biopsy and brushing of peripheral lung cancer with fluoroscopic guidance. Chest 1979; 75:141-45 21 Hanson RR, Zavala DC, Rhodes ML, Keirn L~ Smith JD. Transbronchial biopsy via flexible fiberoptic bronchoscope: results in 164 patients. Am Rev Respir Dis 1976; 114:67-72 22 Fletcher EC, Levin DC. Flexible fiberoptic bronchoscopy and fluoroscopically guided transbronchial biopsy in the management of solitary pulmonary nodules. West J Med 1982; 136:47783 23 Stringfield JT, Markowitz DJ, Bentz RR, Welch RH, Weg JG. The effect of tumor size and location on diagnosis by fiberoptic bronchoscopy. Chest 1977; 72:474-76 24 Radke )R, Conway WA, Eyler WR, Kvale PA. Diagnostic accuracy in peripheral lung lesions: factors predicting success Chest 1979; 76: 176-79 with flexible fiberoptic bronchoscop~ 25 Cortese DA, McDougall JC. Bronchoscopic biopsy and brushing with fluoroscopic guidance in nodular metastatic lung cancer. Chest 1981; 79:610-11 26 Mohsenifar Z, Chopra SK, Simmons DH. Diagnostic value of 6beroptic bronchoscopy in metastatic pulmonary tumors. Chest 1978; 74:369-71 27 Popovich J Jr, Kvale PA, Eichenhorn MS, Radke )R, Ohorodnik JM, Fine G. Diagnostic accuracy of multiple biopsies from flexible fiberoptic bronchoscopy: a comparison of central versus peripheral carcinoma. Am Rev Respir Dis 1982; 125:521-23 28 Shure D, Astarita RW Bronchogenic carcinoma presenting as an endobronchial mass: optimal number of biopsy specimens for diagnosis. Chest 1983; 83:865-67 29 Wang K~ Haponik EF, Britt EJ, Khouri N, Erozan Y. Transbronchial needle aspiration of peripheral pulmonary nodules. Chest 1984; 86:819-23

30 Ono R, Loke J, Ikeda S. Bronchofiberscopy with curette biopsy and bronchography in the evaluation of peripheral lung lesions. Chest 1981; 79:162-66 31 Shure D, Fedullo PF. Transbronchial needle aspiration in the diagnosis of submucosal and peribronchial bronchogenic carcinoma. Chest 1985; 88:49-51 32 Wang K~ Terry PD. Transbronchial needle aspiration in the diagnosis and staging ofbronchogenic carcinoma. Am Rev Respir Dis 1983; 127:344-47 33 Shure D, Fedullo PF. The role of transcarinal needle aspiration in the staging of bronchogenic carcinoma. Chest 1984; 86:69396 34 Versteegh RM, Swierenga J. Bronchoscopic evaluation of the operability of pulmonary carcinoma. Acta Otolaryngol 1963; 56:603-11 35 Buirski G, Calverley PMA, Douglas NJ, Lamb D, McIntyre M, Sudlow MF, et al. Bronchial needle aspiration in the diagnosis of bronchial carcinoma. Thorax 1981; 36:508-11 36 Lundgren R, Bergman F, Angstrom '[ Comparison of transbronchial fine needle aspiration biopsy, aspiration of bronchial secretion, bronchial washing, brush biopsy and forceps biopsy in the diagnosis of lung cancer. Eur J Respir Dis 1983; 64:37885 37 Givens CD, Marini JJ. Transbronchial needle aspiration of a bronchial carcinoid tumor. Chest 1985; 88:152-53 38 Shure D, Moser KM, Konopka R. Transbronchial needle aspiration in the diagnosis of pneumonia in a canine model. Am Rev Respir Dis 1985; 131:290-91 39 Lorch DG, John JF Jr, Tomlinson JR, Miller KS, Sahn SA. Protected transbronchial needle aspiration and protected specimen brush in the diagnosis of pneumonia. Am Rev Respir Dis 1987; 136:565-69 40 Greenman RL, Goodall PT, King D. Lung biopsy immunocompromised hosts. Am J Med 1975; 59:488-96 41 Danek SJ, Bower JS. Diagnosis of pulmonary tuberculosis by flexible fiberoptic bronchoscopy. Am Rev Respir Dis 1979; 119:677-79 42 So SY, Lam WK, Yu DYC. Rapid diagnosis of suspected pulmonary tuberculosis by fiberoptic bronchoscop~ Tubercle 1982; 63: 195-200 43 Wallace JM, Catanzaro A, Moser KM, Harrell JH II. Flexible fiberoptic bronchoscopy for diagnosing pulmonary coccidiodomycosis. Am Rev Respir Dis 1981; 123:286-90 44 Albelda SM, Talbot GH, Gerson SL, Miller Wf, Cassileth PA. Role of fiberoptic bronchoscopy in the diagnosis of invasive pulmonary aspergillosis in patients with acute leukemia. Am J Med 1984; 76: 1027-34 45 Stover DE, Zaman MB, Hajdu SI, Lange M, Gold J, Armstrong D. Bronchoalveolar lavage in the diagnosis of diffuse pulmonary infiltrates in the immunocompromised host. Ann Intern Med 1984; 101:1-7 46 Ognibene F~ Shelhamer J, Gill ~ Macher AM, Loew D, Parker MM, et aI. The diagnosis of Pneumocystis carinii pneumonia in patients with the acquired immunodeficiency syndrome using subsegmental bronchoalveolar lavage. Am Rev Respir Dis 1984; 129:929-32 47 de Garcia J, Curull ~ Vidal R, Riba A, Orriols R, Martin N, et aI. Diagnostic value of bronchoalveolar lavage in suspected pulmonary tuberculosis. Chest 1988; 93:329-32 48 Cockerill FR III, Wilson WR, Carpenter HA, Smith TF, Rosenow EC III. Open lung biopsy in immunocompromised patients. Arch Intern Med 1985; 145:1398-1404 49 Jaffe J~ Maki DG. Lung biopsy in immunocompromised patients. Cancer 1981; 48:1144-53 50 Nishio IN, Lynch JP III. Fiberoptic bronchoscopy in the immunocompromised host: the significance of a unon-specific" transbronchial biopsy. Am Rev Respir Dis 1980; 121:307-12 CHEST I 95 I 5 I MA~



51 Feldman NI: Pennington JE, Ehrie MG. Transbronchial lung 52 53 54 55

biopsy in the compromised host. JAMA 1977; 238:1377-79 Pulcsa 0, Hutcheon MA, Hyland RH. Usefulness oftransbronchial biopsy in immunocompromised patients with pulmonary infiltrates. Thorax 1983; 38:146-50 Koerner SK, Sakowitz AJ, Appelman RI, Becker NH, Schoenbaum SW. Transbronchial lung biopsy for the diagnosis of sarcoidosis. N Engl J Med 1975; 293:268-70 Gilman MJ, Wang KE Transbronchiallung biopsy in sarcoidosis: an approach to determine the optimal number of biopsies. Am Rev Respir Dis 1980; 721-24 Joyner LR, Scheinhorn OJ. Transbronchial forceps lung biopsy through the fiberoptic bronchoscope: diagnosis of diffuse pulmonary disease. Chest 1975; 67:532-35

56 Pauli G, Pelletier A, Bohner C, Roeslin N, Warter A, Roegel E. Transbronchial needle aspiration in the diagnosis of sarcoidosis. Chest 1984; 85:482-84 57 Basset F, Soler ~ Jaurand MC, Bignon J. Ultrastructural examination of broncho-alveolar lavage for diagnosis of pulmonary histiocytosis X. Thorax 1977; 32:303-06 58 Wall C~ Gaensler EA, Carrington CB, Hayes JA. Comparison of transbronchial and open lung biopsies in chronic infiltrative lung diseases. Am Rev Respir Dis 1981; 123:280-85 59 Wilson RK, Rechner RE, Greenberg SD, Estrada R, Stevens PM. Clinical implications of a cenon-specific" transbronchial biopsy. Am J Med 1978; 65:252-56 60 Sanderson DR, Fontana RS. Use of a new controllable-tip brush with the flexible fiber bronchoscope. Chest 1974; 65:620-21

Transbronchial Biopsy and Needle Aspiration (Deborah Shure)