Role of flexible bronchoscopy in the diagnosis of pulmonary infiltrates in pediatric patients with cancer Dennis C. Stokes, MD, Jerry L. Shenep, MD, David Parham, MD, Paula M. Bozeman, MD, William Marienchek, MD, a n d Paul W. M a c k e r t From the Cardiopulmonary-Critical Care Division and the Departments of Infectious Diseases and Pathology, St, Jude Children's Research Hospital, St, Joseph Hospital, and the Departments of Pediatrics and Pathology, University of Tennessee, Memphis We reviewed 60 consecutive flexible bronchoscopies done during a 36-month period in 48 pediatric cancer patients with undiagnosed pulmonary infiltrates. Diagnostic procedures during bronchoscopy included 40 brushings, 50 bronchoalveolar lavages, and 6 transbronchial and mucosal biopsies. A total of 16 specific diagnoses were made by bronchoscopy (27% diagnostic yield), including infection (12), pulmonary leukemia (3), and lymphoma (I). The largest proportion of specific diagnoses came from lavage (14/50) and the smallest from brushings (I/40). Biopsies were also useful for selected patients. The low overall yield for bronchoscopy was probably due to the routine use of empiric broad-spectrum antibiotics and antifungal therapy, as well as trimethoprim-sulfamethoxazole prophylaxis for Pneumocystis carinii pneumonitis. Subsequent specific diagnoses were obtained by other procedures (open biopsy, needle aspiration, or autopsy) for 10 patients with negative bronchoscopy results and 3 patients with diagnostic bronchoscopies. These additional diagnoses included 7 infections (Pneumocystis carinii [I], Candida tropicalis [I], cytomegaIovirus [I], and Aspergillus [4]), and 6 other diagnoses with nonspecific histologic findings. A positive bronchoscopy result may be useful, but negative bronchoscopy findings do not justify delaying other diagnostic procedures or discontinuing antibiotic and antifungal therapy in children with cancer and pulmonary infiltrates.(J PEDIATR1989;115:561-7)
The diagnosis of pulmonary infiltrates of unknown cause in the pediatric patient with c a n c e r - - t h e presence of such infiltrates is termed "pneumopathy X " at our institution remains a difficult clinical problem. Because noninvasive studies generally are not diagnostic, clinicians are faced
Presented in part at the annual meeting of the American Thoracic Society, Las Vegas, Nevada, May 10, 1988. Supported in part by National Cancer Institute Cancer Center (CORE) grant No. P30 CA21765 and the American Lebanese Syrian Associated Charities. Submitted for publication April 4, 1989; accepted May 10, 1989. Reprint requests: Dennis C. Stokes, MD, St. Jude Children's Research Hospital, 332 N. Lauderdale, P.O. Box 318, Memphis, TN 38101-0318. 9/20/13813
with the choice of using empiric antimicrobial therapy or performing invasive procedures such as open lung biopsy, needle aspiration or biopsy, or bronchoscopy. Flexible
AIDS CMV HIV TMP-SMZ
Acquired immunodeficiency syndrome Cytomegalovirus Human immunodeficiency virus Trimethoprim-sulfamethoxazole
See related article, p. 589.
bronchoscopy has proved useful in the diagnosis of several types of pulmonary disorders, but its role in the diagnosis of "pneumopathy X " is less clear. I3 561
562
Stokes et al.
Available studies of bronchoscopy in adult patients with cancer may not always be generalizable to the pediatric population because of age-related differences in underlying diseases and technical differences in the use of bronchoscopy in children. 4-v Previous bronchoscopy series in immunocompromised children have not included sufficient patient numbers to clarify how bronchoscopy results--particularly negative findings--should be applied to clinical decision making in these patients. We therefore reviewed 60 consecutive bronchoscopies performed over a 3-year period at a pediatric oncology center to evaluate the role of bronchoscopy in patients with pulmonary infiltrates. METHODS Patients. The primary diagnoses included acute lymphocytic leukemia (23%), acute nonlymphocytic leukemias (33%), lymphomas (19%), solid tumors (8%), and other hematologic-oncologic disorders (17%). Two patients had human immunodeficiency virus infections in addition to their malignancy, and three had previously undergone allogeneic bone marrow transplantation. Clinicalfindings. Radiographs of the chest revealed diffuse bilateral pulmonary processes in 36% of the patients, and focal infiltrates, either lobar (38%) or multilobar (27%), in the remainder. Other clinical features were typical for a pediatric oncology center population: 50% had neutropenia (median absolute neutrophil count 810/mm3; range, 0 to 21,800/mm 3) and 31% had thrombocytopenia (median platelet count, 72,000/mm3; range, 11,000 to 622,000/mm3). Transfusions were generally performed to bring the platelet count to 50,000/mm 3 or more at bronchoscopy. At the time of bronchoscopy, 27% were receiving oxygen; only five bronchoscopies were performed on patients receiving mechanical ventilation. No patient had significant uremia, pulmonary hypertension, or clinically significant clotting abnormalities other than thrombocytopenia. Antibiotic and antifungal therapy. The majority of patients (73%) were receiving broad-spectrum antibiotics at the time of bronchoscopy (usually vancomycin-amikacinticarcillin8 or oxacillin-cefotaxime, plus erythromycin and trimethoprim-sulfamethoxazole), and 38% were receiving antifungal therapy with amphotericin B. Almost all patients at our institution who are in significant risk categories receive T M P - S M Z as prophylaxis against Pneumocystis carinii pneumonia; two of these patients had allergies to TMPS M Z and were not receiving prophylaxis. Procedures,
Indication for bronchoscopy. The patients in th~s series all had bronchoscopies performed between January 1985 and December 1988 (36 months). Sixty-nine bronchoscopies were performed in 56 patients: 11 patients had more
The Journal of Pediatrics October 1989
than one bronchoscopy during this period and two patients had three bronchoscopies. Nine (13%) of these bronchoscopies were limited diagnostic or therapeutic bronchoscopies performed as therapy for atelectasis or to examine the upper airway; although some of these procedures provided useful information, they were excluded from subse.. quent analysis. The indication for bronchoscopy in the remaining 60 procedures (48 patients) was the identifcation of a new pulmonary infiltrate, detected by chest radiography, that could not be diagnosed by noninvasive methods. Technique. The Olympus model BF-4B2 bronchofiber~ scope (Olympus Corp., Lake Success, N.Y.) (outside diameter, 4.9 mm) was used in half of the procedures (me-. dian patient age, 17 years; range, 9.5 to 29 years) and the pediatric model BF-3C10 (outside diameter, 3.6 ram) in the remainder (median patient age, 7.9 years; range, 2.5 to 26 years). Light general anesthesia was used in 27% of the patients (median age, 7.5 years). In 65% of the patients (median age, 15 years), sedation with intravenously administered meperidine plus diazepam or midazolam was used in addition to topical anesthesia with 1% lidocaine to the nose, vocal cords, and trachea. A few patients also received 4% lidocaine aerosol by intermittent positive-pressure breathing. The transnasal approach was used in all patients without intubation except one; in patients receiving general anesthesia or mechanical ventilation, bronchoscopy was performed through an endotracheal or tracheostomy tube with a q--shaped adapter. Diagnostic approaches. Simple tracheal or bronchial washings (_<10 to 15 ml) were performed during four bronchoscopies (5%). Brushings by either regular disposable cytology brushes (model BC12C, Olympus Corp.) or wax-protected, double-sheath brushes (No. 130 microbiology brush, Mill-Rose Laboratories, Mentor, Ohio) were done in 40 (67%) of 60 cases. Mucosal or transbronchial lung biopsies with standard biopsy forceps (model FB 19C, Olympus Corp.) were done in 6 (10%) of 60 cases. Bronchoalveolar lavage with nonbacteriostatic 0.9% saline solution was performed during 50 (83%) of the 60 procedures. The median lavage volume was 50 ml (range, 25 to 80 ml, in 10 to 15 ml aliquots) with the smaller bronchoscope and 80 ml (range, 40 to 140 ml, in 15 to 20 ml aliquots) with the larger bronchoscope. Pathologic examination and cultures. Biopsy specimens were processed for routine hematoxylin-and-eosin stained section and for the following stains for bacteria, fungi, P. carinii, and acid-fast organisms: Grocott-Gomori methenamine silver nitrate, Brown-Hopps-Gram, periodic acid Schiff, and Kinyoun carbolfuchsin acid-fast stains. Brushings were processed for cytologic examination with the May-Grtinwald-Giemsa stain and the Grocott-Gomori stain. Protected sheath brushes were submitted for bacterial
Volume 115 Number 4
Bronchoscopy in diagnosis o f pulmonary infiltrates
563
T a b l e I. Results of bronchoscopy: washings, brushings, biopsies, a n d b r o n c h o a l v e o l a r lavage Yield Procedure
No.
%
Diagnosis
Washings Brushings Biopsies Mucosal Transbronchial Bronchoalveolartavage
0/4 1/40
2
CMV
1/3 1/3 14/50
33 33 28
Non-Hodgkin lymphoma Presumptive mucormycosis*, % Bacterial: S. aureus, Pseudomonas aeruginosa Fungal Presumptive mucormycosis*, Blastomycosis Presumptive aspergillosis* Presumptive aspergillosis* + Candida albicans C. albicans
Protozoan: P. carinii (2) Viral: CMV~ Mixed Crytococcus neoformans + CMV~ P. carinii + C. albicansw
Noninfectious Leukemia (2) Leukemia + Echovirus type 11 *Histologicfindingsonly,negativeculture. IjSamebro•choscopy. ;Same patient,differentbronchoscopy. w had P. carinii on previousbronchoscopy.
and viral cultures, although quantitative bacterial cultures were not routinely grown. Each lavage specimen was divided and inoculated into culture media for bacteria, viruses, fungi, mycoplasma, Legionella spp., acid-fast bacteria, and Chlamydia trachomatis. Air-dried cytocentrifuge preparations from the lavage fluid were stained with routine and special stains and with immunofluorescent stains for cytomegalovirus, herpes simplex virus, varicella-zoster virus, adenovirus, respiratory syncytial virus, influenza A and B, parainfluenzae viruses 1 to 3, and C. trachomatis. RESULTS Complications. Only two complications were seen. Pneumothorax (10% to 15%) developed in one patient but required no therapy. A run of bigeminal beats developed in a second patient early in the procedure but resolved without additional treatment. Bronchoscopy findings. The overall yield for all bronchoscopies was 16 (27%) of 60 cases, and a specific diagnosis was made by bronchoscopy in 16 (33%) of 48. The yield and specific diagnoses made on the basis of each of the procedures performed at bronchoscopy are shown in Table I. Washings and brushings. None of the four washings showed positive results. The overall yield for brushings, including both the regular cytology brushings and the wax-
protected brushings in which specimens were processed for both cytologic studies and cultures, was 1 (2.5%) of 40, a single culture isolate of CMV. Biopsies. One (33%) of three mucosal biopsy specimens was positive for a tymphoma invading the mucosal wall. For a second mucosal biopsy, performed to determine ciliary morphologic characteristics in a patient with recurrent pneumonias, findings were normal. The third biopsy specimen showed chronic inflammation in a patient in whom a simultaneous transbronchial biopsy demonstrated mucormycosis. The three transbronchial lung biopsies were helpful, although only one provided a specific etiologic diagnosis. All were done in older patients with well-localized pulmonary infiltrates that were easily accessible by fluor0scopy-guided biopsy. One patient with non-Hodgkin lymphoma had invasive mucormycosis, and a second patient had nonspecific findings compatible with radiation-induced pneumonitis and fibrosis. A third patient with lymphoma and a history of lung irradiation and severe ]Vocardia pneumonitis had acute pulmonary decompensation; a biopsy specimen showed severe interstitial fibrosis without evidence of recurrent Nocardia pneumonitis. Bronchoalveolar lavage. Lavage yielded a specific diagnosis in 28% of the bronchoscopies in which it was
564
Stokes et al.
The Journal of Pediatrics October 1989
T a b l e II. A d d i t i o n a l diagnoses m a d e by p r o c e d u r e other t h a n b r o n c h o s c o p y in 13 p a t i e n t s Subsequent findings Initial bronchoscopy diagnosis
Final clinical diagnosis
Method
CMV; CMV + Cryptococcus neoformans* P. carinii; P. carinii + C. albicans* CMV None None None None None None None None None None *Two procedures,same illness. tHistologicfindingsonly,culture negative. $Open lung biopsydone at same time as bronchoscopy. w positiveHIV serologicfindings.
Same + A. flavus Same + CMV (extensive) Same + Presumptive aspergillosist P. carinii Lymphocytic interstitial pneumonia Nonspecific pneumonitis Nonspecific pneumonitis Radiation pneumonitis Azathioprine lung injury Candida tropicalis Presumptive aspergillosis A. flavus Adenovirus type 4
Autopsy Autopsy Needle aspirate Open biopsy$ Open biopsy$, w Open biopsy$ Open biopsy$ History; open biopsy:I: History; open biopsy Computed tomography; spleen aspirate Lung aspirate Autopsy Autopsy
performed (Table I). Diagnoses made by lavage included 1 l infections: 1 bacterial, 5 fungal (Candida albicans, blastomycosis, mucormycosis, aspergillosis), 1 viral, 2 P. carinii, and 2 mixed infections. Hyphal elements consistent with Asperigillus or Mucor were identified in three lavage samples, but the cultures were negative; these diagnoses must therefore be considered presumptive because other fungi may have a similar appearance. All three patients were receiving empiric amphotericin B therapy at the time of bronchoscopy. Organisms originating from the upper respiratory tract, including C. albicans, c~-hemolytic streptococci, and Staphylococcus epidermidis, were common in lavage specimens. Although these organisms are potential pathogens in the immunocompromised host; their presence in lavage or brush cultures was not considered significant in the absence of supportive clinical findings (positive blood cultures, presence of abundant Candida pseudohyphae elements suggesting extensive disease, or computed tomographic evidence of typical fungal !esions in liver or spleen in addition t ~ lung). 9 Only three patients were thought to have significant pulmonary infections with C. albicans, which was present with more significant pathogens (P. carinii, Aspergillus) in two cases. Final clinical diagnoses. A diagnosis made at bronchoscopy was not necessarily the patient's final diagnosis. In 11 of the 16 bronchoscopies with positive results, the study did provide the final clinical diagnosis. However, in three patients (five bronchoscopies), a bronchoscopic diagnosis was later modified by additional clinical information (Table 11). In one patient who had had a bone marrow transplan-
tation, CMV was isolated from one bronchoscopy specimen, CMV and Cryptococcus neoformans were isolated from another specimen during the same illness. At autopsy, however, Aspergillus flavus was also isolated. A patient with Hodgkin disease and acquired immunodeficiency syndrome had P. carinii and P. carinii with C. albicans in two bronchoscopy specimens; autopsy showed both P. carinii and extensive CMV pneumonia. A third patient from whom CMV was isolated initially by bronchoscopy also had presumptive aspergillosis identified by needle aspiration from a progressive lung lesion. In 31 (73%) of the 44 instances in which no specific diagnosis was obtained by bronchoscopy, no specific diagnosis was ever made, and the patients' subsequent clinical courses could be considered satisfactory, although empiric antibiotic and antifungal therapy usually was continued. However, in 10 cases, a specific diagnosis or histol0gic study of the lung process was subsequently obtained by another diagnostic procedure (Table II); three infections were diagnosed by open lung biopsy or autopsy (P. carinii, adenovirus type 4, and A. flavus) and two were diagnosed by needle aspirate (Candida tropicalis and presumptive Aspergiltus). Other histologic diagnoses made by open biopsy in patients with negative bronchoscopy findings included lymphoid interstitial pneumonia in a patient with acute lymphocytic leukemia and previously undiagnosed HIV infection, interstitial and lymphoid pneumonitis in a patient receiving azathioprine (compatible with reports of pulmonary reactions to that drug1~ and nonspecific alveolar damage in a patient with rhabdomyosarcoma and a recent history of lung irradiation. Two other patients had nonspe-
Volume 115 Number 4
Bronehoscopy in diagnosis o f pulmonary infiltrates
565
Bronchoscopy SJCRH, 1985-88 N - 60 (48 patients)
Specific Diagnosis N - 16 (14 pts)
No Specific Diagnosis N - 44 (40 pts) Diagnosis by Repeat Bronchoscopy N - 3 (2 pts)
Final Diagnosis N - 11 (11 pts)
I
Diagnosis by Other Procedure N - 5 (3 pts)
Diagnosis by Other Procedure N - 10 (10 pts)
No Diagnosis; Satisfactory Course N - 31 (28 pts)
I Treatable Cause Missed N - 8 pts - PCP
I
No Rx Indicated N = 5 pts
(1)
- Aspergillus (4) - Candida tropicalis (1) - CMV
(1)
- LIP ( 1 )
Figure. Summary of results of 60 flexible fiberoptic bronchoscopies performed on 48 pediatric oncology patients. SJCRH, St. Jude Children's Research Hospital; PCP, Pneumocystis carinii pneumonia; LIP, lymphoid interstitial pneumonitis.
cific pneumonitis at open biopsy for which no etiologic agent could be identifed. The overall outcome of the 60 bronchoscopies is summarized in the Figure. In 42 (70%) of the 60 procedures, either bronchoscopy provided a final clinical diagnosis of "pneumopathy X" or the patient had a satisfactory clinical course despite negative bronchoscopy findings (with continuing antibiotic therapy, antifungal therapy, or both). In 15 (25%) of the 60 bronchoscopies, or in 13 (27%) of 48 patients, diagnoses were missed by bronchoscopy (Table II. Figure). In this latter group, seven infections were potentially treatable, including P. carinii, Aspergillus, C. tropicalis, and CMV infection. Only the patient with AIDS and CMV infection had not begunspecific therapy before death: the others had all continued treatment with antipneumocystis or antifungal therapy despite negative bronchoscopy findings. In an additional patient with lymphoid interstitial pneumonia, lung biopsy provided the first clue to the diagnosis of HIV infection, resulting in treatment with zidovudine. Five patients had subsequent diagnoses made by lung biopsy for which no specific therapy was possible (nonspecific pneumonias, adenovirus infection).
DISCUSSION Previous exprience with bronchoscopy in children has demonstrated the diagnostic utility of bronchoscopy for a variety of disorders, including CMV and P. carinii pneumonias. 1H4 Frankel et alJ 3 recently described the use of bronchoscopy and bronchoalveolar lavage in the specific diagnosis of pneumonia in sxx of seven immunocomproraised children. Four of the seven had P. carinii pneumonia, including one heart transplant patient and three oncology patients. Pattishall et al. 14 performed bronchoscopy and lavage an 14 immunocompromised children, including organ transplant and cancer patients, and made a diagnosis in 10 (71%), including six with P. carinii pneumonia Although the reports cited above suggest a high diagnostic yield for bronchoscopy in immunocompromised children. it is difficult to reach conclusions regarding the role of bronchoscopy in pediatric oncology patients with pulmonary infiltrates because of the small number of patients, the large percentage with P. carinii pneumonia, and the absence of open-lung biopsy results. Our experience in a consecutive series of 48 pediatric oncology patients seen over a 3-year
566
S t o k e s et al.
period indicates the need for considerable caution in interpreting the results of bronchoscopy in this population. A positive diagnosis made by bronchoscopy may be useful, as in patients with P. earinii and fungal pneumonias. However, these bronchoscopies represent a minority of the total number performed. In severely immunocompromised oncology and bone marrow transplant patients, the isolation of potential pathogens such as CMV and C. albicans cannot automatically be assumed to be the sole cause of the pulmonary infiltrate. Moreover, negative bronchoscopy findings cannot be used as the basis for stopping empiric antibiotic and antifungal therapy because of the possibility that aspergillosis and other treatable infections will he missed. Several factors may have contributed to the low yield in our series. Our patients were generally begun on broadspectrum antibiotic therapy with the first appearance of fever (often before radiologic evidence of pneumonia), which made isolation of bacterial pathogens very unlikely. Thorpe el al., 15 in a study of immunocompromised adults, demonstrated that lavage was likely to yield aerobic bacterial pathogens only when performed before the start of antibiotic therapy. Earlier application of bronchoscopy may increase the yield of bacterial pathogens, even though most bacterial pneumonias would have a satisfactory response to empiric antimicrobial therapy. Many of our patients also received amphotericin B for persistent fever after 7 to 10 days of antibiotic therapy. 16 In aspergillosis, the most common final diagnosis missed by bronchoscopy, the pathogen may be more difficult to culture or to identify in lavage specimens from pretreated patients or from those who are in an early stage of the infection. 17 An unknown percentage of bronchoscopies with negative findings undoubtedly would have represented "true" negative results if open lung biopsies had been performed. Open biopsies were performed in six patients with negative bronchoscopy findings, and nonspecific findings compatible with bronchoscopy results were found in five of these six. Agerelated factors may also be important in making a specific diagnosis; the majority of the specific diagnoses were made with the larger bronchoscope in patients with a median age of 18 years. The ability to perform biopsies and the larger lavage volumes obtained may be important contributors to improved yields in older patients. Many studies of bronchoscopy of the immunocompromised host include a high proportion of patients with P. carinii pneumonia], 13.14, 18, 19 This tends to bias results because of the relative efficiency of bronchoscopy in the diagnosis of this infection, particularly in patients with AIDS, who have large numbers of organisms in sputum and lavage specimens. 19 In oncology centers where T M p - S M Z prophylaxis against P. carinii is routinely used, this organism
The Journal of Pediatrics October 1989
is an uncommon cause of pulmonary infiltrates.2~ 21 None of the three patients with P. carinii identified in our series, including a patient with a negative findings on bronchoscopy, had received prophylaxis. Because the yield of bron~ choscopy versus open biopsy for P. carinii in oncology patients has not been established, it is inadvisable to discontinue antipneumocystis therapy on the basis of negative bronchoscopy findings in those oncology patients with difo fuse pneumonia who have not received T M P - S M Z prophylaxis. The use of transbronchial lung biopsy has been reported rarely in pediatric patients. 22 Biopsy was useful in carefully selected adolescent patients in this series even when a specific diagnosis was not obtained, because it provided information on the histologic features of the lung process. Experience in immunocompromised adults suggests that more widespread use of transbronchial biopsy would have a limited impact on clinical decision making because of the frequency of false negative results, and would undoubtedly increase the rate of complications.23' 24 Bronchial brushings had a very low diagnostic yield in this series, in contrast to some reports of adult patients. 4, 7 Consistent use of quantitative brush cultures may help improve the yield of brushing for bacterial pathogens. 7 There are no previous data on the use of brushings in pediatric pa h tients, other than their use for diagnosis of bacterial infections in patients with mechanically ventilated lungs. 25 Pulmonary involvement with leukemia, identified by bronchoalveolar lavage in three patients here, is an uncommon but well-documented finding. 2628 In one of these patients, echovirus type 11 was also isolated; echoviruses cause diffuse pneumonia in neonates, 29 and this patient may have had a primary echovirus pneumonia in addition to relapsing leukemia. Should a pediatric oncology patient with "pneumopathy X" undergo bronchoscopy--a relatively low-yield but safe procedure--or go directly to open lung biopsy, a procedure with significant morbidity that may not dictate a change in therapy? 30-32 The answer to this question can be provided only by careful prospective studies of bronchoscopy and open biopsy in multiple centers. Our experience indicates that the overall yield of bronchoscopy is low in pediatric oncology patients, and the results must be interpreted cautiously. A positive bronchoscopy result may be useful, particularly when P. carinii is identified, but negative bronchoscopy findings do not justify delaying other diagnostic procedures or discontinuing antibiotic and antifungal therapy in children with cancer and pulmonary infiltrates. We acknowledge the hematology-oncology staff members who participated in the care of these patients, the staff of the microbiology and surgical pathology laboratories, Lennie Lott, RN, Rosemary Crawford, RN, and the editorial assistance of Christy
Volume 115 Number 4
Wright. We thank Drs. Walter Hughes and Joseph Mirro, Jr., for reviewing the manuscript.
REFERENCES 1. Wood RE, Postma D. Endoscopy of the airway in infants and children. J PEDIA'rR 1988;112:1-6. 2. Fitzpatrick SB, Marsh B, Stokes D, Wang K-P. indications for flexible fiberoptic bronchoscopy on pediatric patients. Am J Dis Child 1983;137:595-7. 3. Wood RE. Spelunking in the pediatric airways: explorations with the flexible fiberoptic bronchoscope. Pediatr Clin North Am 1984;31:785-99. 4. Williams D, Yungbluth M, Adams G, Glassroth J. The role of fiberoptic bronchoscopy in the evaluation of immunocompromised hosts with diffuse pulmonary infiltrates. Am Rev Resp Dis 1985;131:880-5. 5. Chuang MT, Tierstein AS, Gribetz AR, Berman LB, Nieburgs HE. Flexible fiberoptic bronchoscopy in the diagnosis of pulmonary complications of lymphoma. Cancer Detect Prey 1981 ;4:347-50. 6. Lauver GL, Hasan FM, Morgan RB, Campbell SC. The usefulness of fiberoptic bronchoscopy in evaluating new pulmonary lesions in the compromised host. Am J Med 1979;66: 580-5. 7. Xaubet A, Torres A, Marco F, Puig-Dela Bellacoxa J, Faus R, Agusti-Vidal A. Pulmonary infiltrates in immunocompromised patients: diagnostic value of telescoping plugged catheter and bronchoalveolar lavage. Chest 1989;95:130-5. 8. Shenep JL, Hughes WT, Roberson PK, et al. Vancomycin, ticarcillin, and amikacin compared with ticarcillin-clavulanate and amikacin in the empirical treatment of febrile, neutropenic children with cancer. N Engl J Med 1988;319:1053-8. 9. Bartley DL, Hughes WT, Parvey LS, Parham D. Computed tomography of hepatic and splenic fungal abscesses in leukemic children. Pediatr Infect Dis 1982;1:317-21. 10. Bedrossian CM, Sussman J, Conklin RH, Kahan B. Azathioprine-associated interstitial pneumonitis. Am J Clin Pathol 1984;82:148-54. 11. Leigh MW, Henshaw NG, Wood RE. Diagnosis of Pneumocystis carinii pneumonia in pediatric patients using bronchoscopic bronchoalveolar lavage. Pediatr Infect Dis 1985;4:40810. 12. McCray PB, Wagener JS, Howe CWS. Bronchoscopic diagnosis of cytomegalovirus pneumonia following pediatric bone marrow transplantation. Am J Pediatr Hematol Oncol 1986;8:338-41. 13. Frankel LR, Smith DW, Lewiston N J. Bronchoalveolar lavage for diagnosis of pneumonia in the immunocompromised child. Peidatrics 1988;81:785-8. 14. Pattishall EN, Noyes BE, Orenstein DM. Use of bronchoalveolar lavage in immunocompromised children with pneumonia. Pediatr Pulmonol 1988;5:1-5. 15. Thorpe JE, Baughman RP, Frame PT, Wesseler TA, Staneck JL. Bronchoalveolar lavage for diagnosing acute bacterial pneumonitis. J Infect Dis 1987;155:855-61.
Bronchoscopy in diagnosis o f pulmonary infiltrates
567
16. Young LS. Empirical antimicrobial therapy in the neutropenic host. N Engl J Med 1986;315:580-1. 17. Albelda SM, Talbot GH, Gerson SL, Miller WT, 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. 18. Stover DE, Zaman MB, Hajdu SI, Lunge M, Gold J, Armstrong D. Bronchoalveolar lavage in the diagnosis of diffuse pulmonary infiltrates in the immunosuppressed host. Ann Intern Med 1984;101:1-7. 19. Broaddus C, Dake MD, Stulbarg MS, et al. Bronchoalveolar lavage and transbrohcnial biopsy for the diagnosis of pulmonary infections in the acquired immunodificiency syndrome. Ann Intern Med 1985;102:747-52. 20. Hughes WT. Five-year absence of Pneumocystis carinii pneumonitis in a pediatric oncology center. J Infect Dis 1984; 150:305-6. 21. Hughes WT, Rivera G, Schell M J, Thornton D, Lott L. Successful intermittent chemoprophylaxis for Pneumocystis carinii pneumonitis. N Engl J Med 1987;316:1627-32. 22. Fitzpatrick SB, Stokes DC, Marsh B, Wang K-P. Transbronchial lung biopsy in pediatric and adolescent patients. Am J Dis Child 1985;139:46-9. 23. Haponik E, Summer WR, Terry PB, Wang K-P. Clinical decision making with transbronchial lung biopsies: the value of nonspecific histologic examination. Am Rev Resp Dis 1982;125:524-9. 24. Nishio JN, Lynch JP. Fiberoptic bronchoscopy in the immunocompromised host: the significance of a "nonspecific" transbronchial biopsy. Am Rev Respir Dis 1980;121:307-12. 25. Zucker A, Pollack M, Katz R. Blind use of the double-lumen plugged catheter for diagnosis of respiratory tract infections in critically ill children. Crit Care Med 1984;12:867-70. 26. Rossi GA, Balbi B, Risso M, Repetto M, Ravazzoni C. Acute myelomonocytic leukemia: demonstration of pulmonary involvement by bronchoalveolar lavage. Chest 1985;87:259-60. 27. Kelleher JF, Miale TD, Donnelly WH. Respiratory distress secondary to pulmonary leukemia. Am J Dis Child 1981; 135:716-8. 28. Wells R J, Weetman RM, Ballantine TVN, Grosfeld JL, Baehner RL. Pulmonary leukemia in children presenting as diffuse interstitial pneumonia. J PEDIATR 1980;96:262-4. 29. Cheeseman SH, Hirsch MS, Keller EW, Keim DE. Fatal neonatal pneumonia caused by echovirus type 9. Am J Dis Child 1977;131:1169. 30. McCabe RE, Brooks RG, Marks JB, Remington JS. Open lung biopsy in patients with acute leukemia. Am J Med 1985;78:609-16. 31. McKenna R J, Mountain CF, McMurtrey MJ. Open lung biopsy in immunocompromised patients. Chest 1984;86:671-4. 32. Potter D, Pass HI, Brower S, et al. Prospective randomized study of open lung biopsy versus empirical antibiotic therapy for acute pneumonitis in nonneutropenic cancer patients. Ann Thorac Surg 1985;40:422-8.