- Email: [email protected]
The Role of Fiberoptic Bronchoscopy for Diagnosis of Pulmonary Tuberculosis in Patients at Risk for AIDS
The Role of Fiberoptic Bronchoscopy for Diagnosis of Pulmonary Tuberculosis in Patients at Risk for AIDS· Adelaida M. Mira, M.D.;t Eugene Gibilara, M.D., F.C.C.E;+
Steven Powell, M.D., F.C.C.E; and Stephan L. Kamholz, M.D., F.C.C.E§
In patients with acquired immunode6ciency syndrome (AIDS)-associated pulmonary Mycobacterium tuberculoaiB (MTB) (group 1), we analyzed whether the addition of transbronchial biopsy (TBB) and bronchial brushings augmented the diagnostic MTB yield over nonbiopsy sampling. Positive acid-fast bacilli (AFB) smears from combined sputum, bronchoalveolar lavage (BAL), and washings were 30 percent compared with 37 percent when brushings and TBB were added (p = NS). The addition of TBB increased culture yield from 96 percent to 100 percent (p NS). Similar results were seen in patients with pulmonary MTB without human immunode6ciency virus (HIV) risk factors (group 2). Group 1 patients most commonly had a noospecmc inftammation on TBB histopathology and had a lower
=
ulmonary infection due to Mycobacterium tuberP culosis (MTB) has been noted with increasing
frequency in persons infected with human immunodeficiency virus (HIV}.I-3 Although not diagnostic of acquired immunodeficiency syndrome (AIDS), pulmonary tuberculosis has been reported in conjunction with other Centers for Disease Control (CDC)-defined opportunistic infections or more commonly, during the prodromal stages of HIV diseases.!" In these patients, the clinical and roentgenographic features of pulmonary tuberculosis are often atypical and indistinguishable from those due to other opportunistic pathogens.Y" Diagnostic 6beroptic bronchoscopy (FOB) is often performed before final results of sputum analysis are available to exclude other common infections such as Pneumocystis carinii pneumonia (PCP). Therefore, we sought to (1) evaluate the relative diagnostic contribution of sputum and various bronchoscopically obtained specimens for MTB, and (2) determine whether the addition of trans bronchial biopsy (TBB) or brushings improved the diagnostic yield over sputum, bronchial washings, and bronchoalveolar lavage (BAL).
*From the King's County Hospital Center, Woodhull Medical and Mental Health Center, and the Pulmonary Division, Department of Internal Medicine, State University of New York(SUNY) Health Science Center at Brooklyn, Brooklyn, NY. tAssistant Professor of Anesthesiology/Critical Care Medicine, Assistant Professor of Medicine, University of Pittsburgh, Pittsburgh. iAssistant Professor of Clinical Medicine. §Professor of Medicine. Manuscript received April 15; revision accepted August 20. Reprint requests: Dr. Miro, Presbyterian University Hospital, 642A Scaife HaU, Pittsburgh 15213
incidence of granuloma formation than group 2 (p
=
=
APB acid-fast bacilli; AIDS acquired immunode8ciency syndrome; BAL=bronchoalveoiar lavage; BRUSH bronchial brushinp; CDC=Centen £or Disease Con~l FOB=8be... optic broncboscopy; HIV human immUDOdellcieney virus; MTB Mf/CObtJcteriufll tuberculoeIa; pcp = PneufflOCfldia Cellinii pneumonia; TBB transbronchial biopsy; WASH pooled broDchial washings
=
=
=
=
=
METHODS
Patients with positive M tuberculosis cultures from any bronchoscopic specimen from Jan I, 1986 to Dec 31, 1986, from King's County Hospital Center and Woodhull Medical and Mental Health Center, Brooklyn, NY,were selected for study. Patients' charts were reviewed and a history of HIV risk behavior, clinical stigmata of HIV infection (oral Candida, lymphadenopathy), and associated CDC-defined opportunistic infections or tumors were noted. Patients were divided into two groups based on the above information. Group I (n = 27) consisted of patients who underwent FOB because they were suspected of having HIV-related pulmonary abnormalities based on clinical history. Patients without risk factors who underwent FOB and had documented MT8 were designated as group 2 (n= 16). Diagnostic FOB was performed by faculty and trainees of the Pulmonary Division, The nature of specimens obtained during the procedure was at the discretion of the participating physicians. BAL was performed in the lobe with greatest parenchymal infiltrates when there was localizing disease. When infiltrates were symmetrically diffuse, the middle lobe or lingula was sampled. Between 100 and 200 ml of normal saline solution at room temperature was instilled into the selected lobe with an average of 50 percent return available for analysis. The amount of intraoperative lidocaine applied to the airways varied between patients and was titrated to patient comfort and cough suppression. All specimens for tuberculosis smear and culture were processed by a single microbiology laboratory. Sputum, bronchial washings, and BAL were pretreated with 2 percent NaOH solution and concentrated by centrifugation at 5,000 rpm for 15 min, decanting supernatant and repeating centrifugation. Ground tissue and concentrated specimens were examined microscopically for acid-fast bacilli (AFB) using the Kinyoun modi6cation of Ziehl-Neelsen's staining. The specimens were plated on Lowenstein-Jensen and Middlebrook 7Hll culture media. Formalin-fixed histologic preparations were interpreted by staff pathologists at each respective institution. Statistical comparison of the diagnostic yield of combined specimens within the same group was done by t test of proportions. Comparisons between the two groups were calculated by an unpaired Student's t test. Values less than 0.05 were considered statistically significant. CHEST I 101 151 MA~
1992
1211
Table 2-HiBtopathologic Condition oj TrafllbronchitJl Biopsy SpecimeraB*
Table I-HlV-Aaociated Opportunistic InfectionB or DiagnosU iraGroup 1*
Previous
Concurrent
Esophageal Candida CNS lymphoma
Group 1 (%) n=22
Subsequent
5
pcp Esophageal Candida Oral Candida
PCP
5
2 13
*HIV = human immunodeficiency virus; CNS = central nervous system; PCP = Pneumocqstis carinii pneumonia. RESULTS
The high-risk activities in group 1 patients included intravenous drug abuse (16), homosexuality (three), blood transfusion (one), and other (seven). Within this group, 18 patients had documented signs of HIV disease (Table 1). The individual diagnostic yield of sputum and FOB specimens by direct smear and culture is summarized in Figure 1. In group 1, sputum and pooled bronchial washings had the highest proportion (23 percent and 26 percent, respectively) of positive AFB smears, while BAL and washings (95 percent each) had the highest yield by culture. In group 1, the diagnostic yield of positive AFB smears
GROUP 1
Granuloma formation Nonspecific inflammation Interstitial fibrosis No acute inflammation PCP only
~
a..
GROUP 1
E
12
70
60
~ 50
Q.
40
~ 30
4 2
....._"'&";;o
OL.......£.::.::..&.:=..~~~:..A.__a.:.:.&.:.::.a__.L.~
ffi] • Number
20 10
...._w. . . . .
0-----
of sp.cim.n submitt.d
o
(%) • Per c.nt positlv.
e · Positive AF B sm.ar
=·
•
Positive MfS culture
GROUP 2
UJ 14
en o
10
a. a::
p: HS
~
SPUTUM • •AL • WASH
IPUTUM • •AL • WAIH • TBB • BRUIH
compared with rlZJ
GROUP 2
80 ~
70
~
50
i= 60
8
e, 40
OJ
~ 30
:E 4 ::)
20 10
2
o
O~--_--..I
[ [ ] • Number of sp.cim.n submitt.d
o
CULTURE
SMEARS
SPUTUM OHLY
rz:2
SPUTUM. BAL. WASH
(%) • Per cent pasitiv.
e · Positive AFB smear
rtllJ • Positive
MfS culture
FIGURE 1. AFB smear and MTB culture yield of individual diagnostic specimens for both patient groups.
1212
•
100-----------90
(92%)
UJ 6
Z
CULTURE
SMEARS
SPUTUM ONLY
", bronchial brulhlng not cultured
16---------------------, 12
#
80 ~
10
~
•
90
CD 8 :E 6
~
o (0)
100~-------------.,
18
~
5 (63)t 2 (25) 1 (12)
from combined sputum, BAL, and bronchial washings was 30 percent compared with 37 percent when bronchial brushings and TBB were added (p = NS). The addition ofTBB increased the culture yield from 96 percent to 100 percent (p = NS). The group 2 yield from direct smears and cultures was not significantly different from group 1 (Fig 1 and 2). The histopathologic condition of the TBB submitted from each group is described in Table 2. A nonspecific in8ammatory response was the most commonly reported finding in group 1. Granuloma formation was seen in only 9 percent of biopsy specimens in the high-risk group
(95%)
t:: 16 CJ) o 14
ffi
(9) (68) (9) (5) (9)
*n = number of patients; PCP = Pneumocqstis carinii pneumonia. tp
24--------------------
22 W 20
2 15 2 1 2
Group 2 (%) n=8
•
•
SPUTUM. BAL • WASH. TB • • BRUIH
p: HI co.pared with rlZJ
", bronchial brulhlng not cultured
FIGURE 2. AFB smear and MTB culture yield of combined diagnostic specimens for both patient groups. FOB in Diagnosis of Pulmonary TB in Patients at Risk for AIDS (Mira at 81)
compared with 63 percent in the control population (p
During bronchoscopic evaluation of pulmonary pathologic processes in suspected AIDS patients, BAL is almost always performed. Other sampling modalities such as bronchial brushing, transbronchial needle aspiration, and TBB are used less consistently. Although BAL has a reported sensitivity between 94 and 99 percent for detecting PC~7,8 it is not known which specimens are necessary to adequately maximize the detection of AIDS-associated pulmonary MTB. Our study suggests that a combination of sputum, bronchial washings, and BAL analysis is sufficient to establish the microscopic and culture diagnosis of pulmonary MTB (without TBB) in the AIDS population (Group 1). These findings were similar in our group 2 patients and consistent with previous studies from an unselected population where TBB did not make additional contributions to the diagnosis of pulmonary MTB.9 In only one instance in this series was culture of the TBB the sole modality by which the diagnosis was established. Similarly, nonbiopsy bronchoscopic sampling methods have proved adequate for diagnosing AIDS-associated cryptococcal pulmonary disease as was recently reported from our institution. 10 In addition, the relative rarity of a normal cell-mediated granulomatous tissue reaction to the presence of MTB (nondiagnostic histopathologic condition in 91 percent of the TBB) further reduces the diagnostic utility of the lung biopsy The purified protein derivative (PPD) status and HIV serology of our patients were not consistently available for all patients. Furthermore, skin and serologic test results are often not routinely available prior to the clinically driven decision to proceed with diagnostic FOB. Also, the high incidence of anergy among HIVinfected patients makes a PPD testing a less reliable marker for the presence of tuberculosis infection. 11 Interestingly, in the high-risk group, the presentation of pulmonary disease due to MTB most often occurred simultaneously with other opportunistic pathogens (Table 1). Although Candida infection was most commonly detected, 19 percent (5I27) had concurrent pce These data underscore the importance and value of early diagnostic FOB with submission of appropriate specimens to detect the wide array of pulmonary infections typically occurring in these patients with complex conditions. Our data may seem at variance with previous studies where pulmonary MTB was frequently diagnosed as the earliest evidence of host immunodeficiency I-f However, patients were selected if they had documented MTB and underwent diagnostic FOB. Thus, our population is
biased toward a group that was more likely to undergo FOB because of a greater suspicion of having PCP as the primary diagnosis. The 19 percent incidence of concurrent MTB and PCP in this series is similar to that reported by others. 6 In both groups of patients, bronchoscopy did not statistically improve immediate or ultimate MTB diagnosis over sputum alone. Although the utility of bronchoscopy in patients with smear-negative sputum is often debated in the literature,t2,13 it was not the intention of this study to address this issue. We sought to determine which bronchoscopic specimens are sufficient to properly evaluate suspected AIDS patients for pulmonary MTB infections. If nonbronchoscopic specimens, such as induced-sputum obtained as part of the evaluation for PC~ are employed as the first diagnostic tool, then independent considerations must be given to other possible concurrent pulmonary infectious complications such as tuberculosis. In summary, our results suggest that more invasive diagnostic sampling with bronchial brushing and TBB does not contribute to the microscopic, bacteriologic, or histopathologic diagnosis of AIDS-associated pulmonary MTB. Furthermore, the risk-benefit ratio for all invasive procedures should be optimized in these frequently hypoxemic and seriously ill patients. The information gained by TBB must be weighed against a more prolonged procedure, risk of bleeding and pneumothorax, ftuoroscopic radiation exposure, and cost-effectiveness. The combination of sputum, washings, and BAL appropriately stained and cultured for mycobacteria appears sufficient. REFERENCES
1 Pitchenilc AE, Cole CR, Bussell B~ Fischl MA, Spira Tj, Snider DE Jr. Tuberculosis, atypical mycobacteriosis, and the acquired immunodeficiency syndrome among Haitian and nonHaitian patients in South Florida. Ann Intern Moo 1984; 101:641-45 2 Sunderam G, McDonald 8J, Maniatis 1: Oleske J, Kapila R, Reichman LB. Tuberculosis as a manifestation of the acquired immunodeficiency syndrome (AIDS). JAMA 1986; 246:362-66 3 Pitchenilc AE, Burr J, Suarez M, Fertel D, Gonzalez G, Moas C. Human T-celllymphotrophic virus-III (HTLV-III) seropositivity and related disease among 71 consecutive patients in whom tuberculosis was diagnosed. Am Rev Bespir Dis 1987; 135:875-79 4 Handwerger S, Mildvan D, Senie R, McKinley FW Tuberculosis and the acquired immunodeficiency syndrome at a New YorkCity hospital: 1978-1985. Chest 1987; 91:176-80 5 Pitchenilc AE, Rubinson HA. The radiographic appearance of tuberculosis in patients with the acquired immune deficiency syndrome (AIDS) and pre-AIDS. Am Rev Respir Dis 1985; 131:393-96 6 Modilevsky T, Sattler FR, Barnes PF. Mycobacterial disease in patients with immunodeficiency virus infection. Arch Intern Moo 1989; 149:2201-05 7 Stover DE, White DA, Romano PA, GeDene RA. Diagnosis of pulmonary disease in acquired immune deficiency syndrome (AIDS): role of bronchoscopy and bronchoalveolar lavage. Am CHEST I 101 I 5 I MA~
1992
1213
Rev Respir Dis 1984; 130:659-62 8 Orenstein M, Webber C, Cash M, Heurich AE. Value of bronchoalveolar lavage in the diagnosis of pulmonary infection in acquired immune deficiency syndrome. Tborax 1986; 41:34549 9 Stenson ~ Aranda C, Beleaqua FA. Transbroncbial biopsy culture in pulmonary tuberculosis. Chest 1983; 83:883-84 10 Malabonga VM, Basti J, Kambolz SL. Utility of broncboscopic sampling tecbniques for cryptoeoccal disease in AIDS. Cbest 1991;99:370-72
11 Cbaisson RE, Schecter GF. Theuer C~ Rutherford G~ Ecbenberg DF, Hopewell PC. Tuberculosis in patients with acquired immunodeficiency syndrome: clinical features, response to therapy, and survival. Am Rev Respir Dis 1987; 136:570-74 12 WallaceJM, Deutscb AL, Harrell JB, Moser KM. Broncboscopy and transbronchial biopsy in evaluation of patients with suspected active tuberculosis. Am J Moo 1981; 70:1189-94 13 Kvale PA, Johnson MC, Wroblewski DA. Diagnosis oftubereulosis: routine cultures of broncbial washings are not indicated. Cbest 1979; 76:140-42
Postgraduate Course: Surgery of the Airways Massachusetts General Hospital and Harvard Medical School will hold this course July 2021 at the Massachusetts General Hospital, Boston. For information, contact: Thoracic Surgical Unit, Massachusetts General Hospital, Boston 02114 (617:627-2806).
1214
FOBin Diagnosis of PulmonaryTB in Patients at Risk for AIDS (Mito at al)
Steven Powell, M.D., F.C.C.E; and Stephan L. Kamholz, M.D., F.C.C.E§
In patients with acquired immunode6ciency syndrome (AIDS)-associated pulmonary Mycobacterium tuberculoaiB (MTB) (group 1), we analyzed whether the addition of transbronchial biopsy (TBB) and bronchial brushings augmented the diagnostic MTB yield over nonbiopsy sampling. Positive acid-fast bacilli (AFB) smears from combined sputum, bronchoalveolar lavage (BAL), and washings were 30 percent compared with 37 percent when brushings and TBB were added (p = NS). The addition of TBB increased culture yield from 96 percent to 100 percent (p NS). Similar results were seen in patients with pulmonary MTB without human immunode6ciency virus (HIV) risk factors (group 2). Group 1 patients most commonly had a noospecmc inftammation on TBB histopathology and had a lower
=
ulmonary infection due to Mycobacterium tuberP culosis (MTB) has been noted with increasing
frequency in persons infected with human immunodeficiency virus (HIV}.I-3 Although not diagnostic of acquired immunodeficiency syndrome (AIDS), pulmonary tuberculosis has been reported in conjunction with other Centers for Disease Control (CDC)-defined opportunistic infections or more commonly, during the prodromal stages of HIV diseases.!" In these patients, the clinical and roentgenographic features of pulmonary tuberculosis are often atypical and indistinguishable from those due to other opportunistic pathogens.Y" Diagnostic 6beroptic bronchoscopy (FOB) is often performed before final results of sputum analysis are available to exclude other common infections such as Pneumocystis carinii pneumonia (PCP). Therefore, we sought to (1) evaluate the relative diagnostic contribution of sputum and various bronchoscopically obtained specimens for MTB, and (2) determine whether the addition of trans bronchial biopsy (TBB) or brushings improved the diagnostic yield over sputum, bronchial washings, and bronchoalveolar lavage (BAL).
*From the King's County Hospital Center, Woodhull Medical and Mental Health Center, and the Pulmonary Division, Department of Internal Medicine, State University of New York(SUNY) Health Science Center at Brooklyn, Brooklyn, NY. tAssistant Professor of Anesthesiology/Critical Care Medicine, Assistant Professor of Medicine, University of Pittsburgh, Pittsburgh. iAssistant Professor of Clinical Medicine. §Professor of Medicine. Manuscript received April 15; revision accepted August 20. Reprint requests: Dr. Miro, Presbyterian University Hospital, 642A Scaife HaU, Pittsburgh 15213
incidence of granuloma formation than group 2 (p
=
=
APB acid-fast bacilli; AIDS acquired immunode8ciency syndrome; BAL=bronchoalveoiar lavage; BRUSH bronchial brushinp; CDC=Centen £or Disease Con~l FOB=8be... optic broncboscopy; HIV human immUDOdellcieney virus; MTB Mf/CObtJcteriufll tuberculoeIa; pcp = PneufflOCfldia Cellinii pneumonia; TBB transbronchial biopsy; WASH pooled broDchial washings
=
=
=
=
=
METHODS
Patients with positive M tuberculosis cultures from any bronchoscopic specimen from Jan I, 1986 to Dec 31, 1986, from King's County Hospital Center and Woodhull Medical and Mental Health Center, Brooklyn, NY,were selected for study. Patients' charts were reviewed and a history of HIV risk behavior, clinical stigmata of HIV infection (oral Candida, lymphadenopathy), and associated CDC-defined opportunistic infections or tumors were noted. Patients were divided into two groups based on the above information. Group I (n = 27) consisted of patients who underwent FOB because they were suspected of having HIV-related pulmonary abnormalities based on clinical history. Patients without risk factors who underwent FOB and had documented MT8 were designated as group 2 (n= 16). Diagnostic FOB was performed by faculty and trainees of the Pulmonary Division, The nature of specimens obtained during the procedure was at the discretion of the participating physicians. BAL was performed in the lobe with greatest parenchymal infiltrates when there was localizing disease. When infiltrates were symmetrically diffuse, the middle lobe or lingula was sampled. Between 100 and 200 ml of normal saline solution at room temperature was instilled into the selected lobe with an average of 50 percent return available for analysis. The amount of intraoperative lidocaine applied to the airways varied between patients and was titrated to patient comfort and cough suppression. All specimens for tuberculosis smear and culture were processed by a single microbiology laboratory. Sputum, bronchial washings, and BAL were pretreated with 2 percent NaOH solution and concentrated by centrifugation at 5,000 rpm for 15 min, decanting supernatant and repeating centrifugation. Ground tissue and concentrated specimens were examined microscopically for acid-fast bacilli (AFB) using the Kinyoun modi6cation of Ziehl-Neelsen's staining. The specimens were plated on Lowenstein-Jensen and Middlebrook 7Hll culture media. Formalin-fixed histologic preparations were interpreted by staff pathologists at each respective institution. Statistical comparison of the diagnostic yield of combined specimens within the same group was done by t test of proportions. Comparisons between the two groups were calculated by an unpaired Student's t test. Values less than 0.05 were considered statistically significant. CHEST I 101 151 MA~
1992
1211
Table 2-HiBtopathologic Condition oj TrafllbronchitJl Biopsy SpecimeraB*
Table I-HlV-Aaociated Opportunistic InfectionB or DiagnosU iraGroup 1*
Previous
Concurrent
Esophageal Candida CNS lymphoma
Group 1 (%) n=22
Subsequent
5
pcp Esophageal Candida Oral Candida
PCP
5
2 13
*HIV = human immunodeficiency virus; CNS = central nervous system; PCP = Pneumocqstis carinii pneumonia. RESULTS
The high-risk activities in group 1 patients included intravenous drug abuse (16), homosexuality (three), blood transfusion (one), and other (seven). Within this group, 18 patients had documented signs of HIV disease (Table 1). The individual diagnostic yield of sputum and FOB specimens by direct smear and culture is summarized in Figure 1. In group 1, sputum and pooled bronchial washings had the highest proportion (23 percent and 26 percent, respectively) of positive AFB smears, while BAL and washings (95 percent each) had the highest yield by culture. In group 1, the diagnostic yield of positive AFB smears
GROUP 1
Granuloma formation Nonspecific inflammation Interstitial fibrosis No acute inflammation PCP only
~
a..
GROUP 1
E
12
70
60
~ 50
Q.
40
~ 30
4 2
....._"'&";;o
OL.......£.::.::..&.:=..~~~:..A.__a.:.:.&.:.::.a__.L.~
ffi] • Number
20 10
...._w. . . . .
0-----
of sp.cim.n submitt.d
o
(%) • Per c.nt positlv.
e · Positive AF B sm.ar
=·
•
Positive MfS culture
GROUP 2
UJ 14
en o
10
a. a::
p: HS
~
SPUTUM • •AL • WASH
IPUTUM • •AL • WAIH • TBB • BRUIH
compared with rlZJ
GROUP 2
80 ~
70
~
50
i= 60
8
e, 40
OJ
~ 30
:E 4 ::)
20 10
2
o
O~--_--..I
[ [ ] • Number of sp.cim.n submitt.d
o
CULTURE
SMEARS
SPUTUM OHLY
rz:2
SPUTUM. BAL. WASH
(%) • Per cent pasitiv.
e · Positive AFB smear
rtllJ • Positive
MfS culture
FIGURE 1. AFB smear and MTB culture yield of individual diagnostic specimens for both patient groups.
1212
•
100-----------90
(92%)
UJ 6
Z
CULTURE
SMEARS
SPUTUM ONLY
", bronchial brulhlng not cultured
16---------------------, 12
#
80 ~
10
~
•
90
CD 8 :E 6
~
o (0)
100~-------------.,
18
~
5 (63)t 2 (25) 1 (12)
from combined sputum, BAL, and bronchial washings was 30 percent compared with 37 percent when bronchial brushings and TBB were added (p = NS). The addition ofTBB increased the culture yield from 96 percent to 100 percent (p = NS). The group 2 yield from direct smears and cultures was not significantly different from group 1 (Fig 1 and 2). The histopathologic condition of the TBB submitted from each group is described in Table 2. A nonspecific in8ammatory response was the most commonly reported finding in group 1. Granuloma formation was seen in only 9 percent of biopsy specimens in the high-risk group
(95%)
t:: 16 CJ) o 14
ffi
(9) (68) (9) (5) (9)
*n = number of patients; PCP = Pneumocqstis carinii pneumonia. tp
24--------------------
22 W 20
2 15 2 1 2
Group 2 (%) n=8
•
•
SPUTUM. BAL • WASH. TB • • BRUIH
p: HI co.pared with rlZJ
", bronchial brulhlng not cultured
FIGURE 2. AFB smear and MTB culture yield of combined diagnostic specimens for both patient groups. FOB in Diagnosis of Pulmonary TB in Patients at Risk for AIDS (Mira at 81)
compared with 63 percent in the control population (p
During bronchoscopic evaluation of pulmonary pathologic processes in suspected AIDS patients, BAL is almost always performed. Other sampling modalities such as bronchial brushing, transbronchial needle aspiration, and TBB are used less consistently. Although BAL has a reported sensitivity between 94 and 99 percent for detecting PC~7,8 it is not known which specimens are necessary to adequately maximize the detection of AIDS-associated pulmonary MTB. Our study suggests that a combination of sputum, bronchial washings, and BAL analysis is sufficient to establish the microscopic and culture diagnosis of pulmonary MTB (without TBB) in the AIDS population (Group 1). These findings were similar in our group 2 patients and consistent with previous studies from an unselected population where TBB did not make additional contributions to the diagnosis of pulmonary MTB.9 In only one instance in this series was culture of the TBB the sole modality by which the diagnosis was established. Similarly, nonbiopsy bronchoscopic sampling methods have proved adequate for diagnosing AIDS-associated cryptococcal pulmonary disease as was recently reported from our institution. 10 In addition, the relative rarity of a normal cell-mediated granulomatous tissue reaction to the presence of MTB (nondiagnostic histopathologic condition in 91 percent of the TBB) further reduces the diagnostic utility of the lung biopsy The purified protein derivative (PPD) status and HIV serology of our patients were not consistently available for all patients. Furthermore, skin and serologic test results are often not routinely available prior to the clinically driven decision to proceed with diagnostic FOB. Also, the high incidence of anergy among HIVinfected patients makes a PPD testing a less reliable marker for the presence of tuberculosis infection. 11 Interestingly, in the high-risk group, the presentation of pulmonary disease due to MTB most often occurred simultaneously with other opportunistic pathogens (Table 1). Although Candida infection was most commonly detected, 19 percent (5I27) had concurrent pce These data underscore the importance and value of early diagnostic FOB with submission of appropriate specimens to detect the wide array of pulmonary infections typically occurring in these patients with complex conditions. Our data may seem at variance with previous studies where pulmonary MTB was frequently diagnosed as the earliest evidence of host immunodeficiency I-f However, patients were selected if they had documented MTB and underwent diagnostic FOB. Thus, our population is
biased toward a group that was more likely to undergo FOB because of a greater suspicion of having PCP as the primary diagnosis. The 19 percent incidence of concurrent MTB and PCP in this series is similar to that reported by others. 6 In both groups of patients, bronchoscopy did not statistically improve immediate or ultimate MTB diagnosis over sputum alone. Although the utility of bronchoscopy in patients with smear-negative sputum is often debated in the literature,t2,13 it was not the intention of this study to address this issue. We sought to determine which bronchoscopic specimens are sufficient to properly evaluate suspected AIDS patients for pulmonary MTB infections. If nonbronchoscopic specimens, such as induced-sputum obtained as part of the evaluation for PC~ are employed as the first diagnostic tool, then independent considerations must be given to other possible concurrent pulmonary infectious complications such as tuberculosis. In summary, our results suggest that more invasive diagnostic sampling with bronchial brushing and TBB does not contribute to the microscopic, bacteriologic, or histopathologic diagnosis of AIDS-associated pulmonary MTB. Furthermore, the risk-benefit ratio for all invasive procedures should be optimized in these frequently hypoxemic and seriously ill patients. The information gained by TBB must be weighed against a more prolonged procedure, risk of bleeding and pneumothorax, ftuoroscopic radiation exposure, and cost-effectiveness. The combination of sputum, washings, and BAL appropriately stained and cultured for mycobacteria appears sufficient. REFERENCES
1 Pitchenilc AE, Cole CR, Bussell B~ Fischl MA, Spira Tj, Snider DE Jr. Tuberculosis, atypical mycobacteriosis, and the acquired immunodeficiency syndrome among Haitian and nonHaitian patients in South Florida. Ann Intern Moo 1984; 101:641-45 2 Sunderam G, McDonald 8J, Maniatis 1: Oleske J, Kapila R, Reichman LB. Tuberculosis as a manifestation of the acquired immunodeficiency syndrome (AIDS). JAMA 1986; 246:362-66 3 Pitchenilc AE, Burr J, Suarez M, Fertel D, Gonzalez G, Moas C. Human T-celllymphotrophic virus-III (HTLV-III) seropositivity and related disease among 71 consecutive patients in whom tuberculosis was diagnosed. Am Rev Bespir Dis 1987; 135:875-79 4 Handwerger S, Mildvan D, Senie R, McKinley FW Tuberculosis and the acquired immunodeficiency syndrome at a New YorkCity hospital: 1978-1985. Chest 1987; 91:176-80 5 Pitchenilc AE, Rubinson HA. The radiographic appearance of tuberculosis in patients with the acquired immune deficiency syndrome (AIDS) and pre-AIDS. Am Rev Respir Dis 1985; 131:393-96 6 Modilevsky T, Sattler FR, Barnes PF. Mycobacterial disease in patients with immunodeficiency virus infection. Arch Intern Moo 1989; 149:2201-05 7 Stover DE, White DA, Romano PA, GeDene RA. Diagnosis of pulmonary disease in acquired immune deficiency syndrome (AIDS): role of bronchoscopy and bronchoalveolar lavage. Am CHEST I 101 I 5 I MA~
1992
1213
Rev Respir Dis 1984; 130:659-62 8 Orenstein M, Webber C, Cash M, Heurich AE. Value of bronchoalveolar lavage in the diagnosis of pulmonary infection in acquired immune deficiency syndrome. Tborax 1986; 41:34549 9 Stenson ~ Aranda C, Beleaqua FA. Transbroncbial biopsy culture in pulmonary tuberculosis. Chest 1983; 83:883-84 10 Malabonga VM, Basti J, Kambolz SL. Utility of broncboscopic sampling tecbniques for cryptoeoccal disease in AIDS. Cbest 1991;99:370-72
11 Cbaisson RE, Schecter GF. Theuer C~ Rutherford G~ Ecbenberg DF, Hopewell PC. Tuberculosis in patients with acquired immunodeficiency syndrome: clinical features, response to therapy, and survival. Am Rev Respir Dis 1987; 136:570-74 12 WallaceJM, Deutscb AL, Harrell JB, Moser KM. Broncboscopy and transbronchial biopsy in evaluation of patients with suspected active tuberculosis. Am J Moo 1981; 70:1189-94 13 Kvale PA, Johnson MC, Wroblewski DA. Diagnosis oftubereulosis: routine cultures of broncbial washings are not indicated. Cbest 1979; 76:140-42
Postgraduate Course: Surgery of the Airways Massachusetts General Hospital and Harvard Medical School will hold this course July 2021 at the Massachusetts General Hospital, Boston. For information, contact: Thoracic Surgical Unit, Massachusetts General Hospital, Boston 02114 (617:627-2806).
1214
FOBin Diagnosis of PulmonaryTB in Patients at Risk for AIDS (Mito at al)