Chest ACCP Education Calendar
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Guest Access | Sign In via User Name/Password
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (27)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Jain, P.
Right arrow Articles by Mehta, A. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Jain, P.
Right arrow Articles by Mehta, A. C.
(Chest. 2004;125:712-722.)
© 2004 American College of Chest Physicians

Role of Flexible Bronchoscopy in Immunocompromised Patients With Lung Infiltrates*

Prasoon Jain, MD, FCCP; Sunder Sandur, MD, FCCP; Yvonne Meli, RN; Alejandro C. Arroliga, MD, FCCP; James K. Stoller, MD, MS, FCCP and Atul C. Mehta, MD, FCCP

* From the Department of Medicine (Dr. Jain), Louis A. Johnson Veterans Affairs Medical Center, Clarksburg, WV; the Department of Pulmonary and Critical Care Medicine (Dr. Sandur), St. Mary’s Hospital, Waterbury, CT; and the Department of Pulmonary and Critical Care Medicine (Drs. Arroliga, Stoller, and Mehta, and Ms. Meli), The Cleveland Clinic Foundation, Cleveland, OH.

Correspondence to: Atul C. Mehta, MD, FCCP, Head, Section of Bronchology, Department of Pulmonary and Critical Care Medicine/A-90, The Cleveland Clinic Foundation, 9500 Euclid Ave, Cleveland, OH 44195; e-mail: mehtaa1{at}ccf.org


   Abstract
TOP
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Conclusion
 References
 
Study objective: To study the diagnostic role of flexible bronchoscopy (FB) in immunocompromised patients with pulmonary infiltrates.

Design: Prospective, observational study.

Setting: Tertiary care hospital.

Patients: A total of 104 consecutive non-HIV-infected immunocompromised patients with lung infiltrates in whom FB was performed.

Methods: The primary outcome measure was the diagnostic yield of FB, which was derived as the number of the diagnoses made using FB results divided by all final diagnoses. Final diagnoses were established using data from FB, surgical lung biopsy (SLB), and microbiology and serology testing, and by the clinical response to empiric therapy. We also studied the diagnostic yields of individual sampling procedures such as BAL, transbronchial biopsy (TBB), and protected-specimen brush (PSB) sampling.

Results: Overall, 128 diagnoses were made in 104 patients. The overall diagnostic yield of FB was 56.2% (95% confidence interval [CI], 47 to 64%). FB provided at least one diagnosis in 53 of 104 patients (51%; 95% CI, 40 to 62%). FB was more likely to establish the diagnosis when the lung infiltrate was due to an infectious agent (81%; 95% CI, 67 to 90%) than to a noninfectious process (56%; 95% CI, 43 to 67%; p = 0.011). The diagnostic yields of BAL (38%; 95% CI, 30 to 47%) and TBB (38%; 95% CI, 27 to 51%) were similar (p = 0.94). The diagnostic yield of PSB sampling was lower (13%; 95% CI 6 to 24%; p = 0.001) than that of BAL. The combined diagnostic yield of BAL and TBB (70%; 95% CI, 57 to 80%) was higher than that of BAL alone (p < 0.001). Finally, the diagnostic yield of FB with PSB sampling, BAL, and TBB was similar to that of FB with BAL and TBB. The complication rate from FB was 21% (95% CI, 15 to 31%). Minor bleeding (13%) and pneumothorax (4%) were the most common complications.

Conclusions: FB has a high diagnostic yield in immunocompromised patients with pulmonary infiltrates. Based on our results, we recommend performing TBB in these patients, whenever possible.

Key Words: BAL • bronchoscopy • immunocompromised patients


   Introduction
TOP
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Conclusion
 References
 
Pulmonary complications account for significant morbidity and mortality in immunocompromised patients.1 2 3 4 5 6 The optimal outcome in these patients depends on identifying the cause of the lung infiltrate and instituting specific therapy as soon as possible. Because several infectious and noninfectious diseases can present with similar clinical and radiologic features in immunocompromised patients with lung infiltrates, establishing the cause of pulmonary infiltrates is challenging in this setting.7 The current study was performed to better understand the value of flexible bronchoscopy (FB) in assessing the cause of pulmonary infiltrates in non-HIV- infected immunocompromised patients. We report the diagnostic yields of different sampling techniques including BAL, protected-specimen brush (PSB) sampling, and transbronchial biopsy (TBB) in these patients. This study also examines the frequency of complications of FB in immunocompromised patients with lung infiltrates.


   Materials and Methods
TOP
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Conclusion
 References
 
Patients
One hundred four consecutive immunocompromised patients with lung infiltrates seen on chest roentgenograms who were referred to the inpatient Pulmonary Consult Service of the Cleveland Clinic Foundation over a 2-year period comprised the patient population. Eligibility criteria included the following: (1) immunocompromised state; (2) the presence of either a focal or diffuse pulmonary infiltrate; (3) a decision by pulmonary consultant to perform FB; and (4) informed consent to undergo FB given by the patient.

The following conditions defined the immunocompromised state: hematologic malignancy (eg, leukemia, lymphoma, myelodysplastic syndrome, and multiple myeloma); chemotherapy in the last 6 months; long-term corticosteroid therapy (ie, >= 20 mg/d prednisone or its equivalent for >= 2 months) or high-dose corticosteroid therapy (>= 60 mg/d prednisone or its equivalent for >= 2 weeks within 3 months); bone marrow or solid organ transplantation; and the current use of immunosuppressive and cytotoxic medications. Patients were excluded from the study if: (1) they were HIV positive or (2) they had received a lung transplant. Patients were included if they were receiving antimicrobial therapy at the time of FB.

Data Collection
Demographic features and details of underlying causes of the immunocompromised state, current medications including antimicrobial therapy, and relevant clinical, laboratory, and radiologic findings were recorded at the time of FB using a standard data collection form. The pulmonary consultant, who was aware of the ongoing study, directed the patient management (eg, antibiotics, or the need for FB or surgical lung biopsy [SLB]). Cultures of sputum and blood, serology tests, and other laboratory tests were performed at the discretion of the managing physicians. The final diagnosis was determined using data from clinical presentation, serology, and microbiological tests, the results of FB and SLB, response to therapy, and postmortem examination.

FB
All FBs were performed in an inpatient bronchoscopy suite or in the ICU with extensive cardiopulmonary monitoring. The choice of sedating medications was at the discretion of the pulmonary physician. Neuromuscular blocking agents were not routinely employed in intubated patients undergoing FB. Local anesthesia was provided with sterile methylparaben-free 2% lidocaine.

Hemostatic goals during FB were as follows: platelet count, >= 20 x 109 cells/L when TBB was not performed, and >= 50 x 109 cells/L when TBB was performed; prothrombin time-international normalized ratio (PT-INR) less than or equal to two times normal; and partial thromboplastin time within the normal range. The choice of sampling procedure during FB (eg, BAL, PSB sampling, or TBB) was at the pulmonologist’s discretion. The standard sequence of sampling techniques used during FB was BAL, PSB sampling, and bronchial washings (BWs), followed by TBB.

BAL was performed according to American Thoracic Society guidelines.8 In patients with diffuse pulmonary infiltrates, BAL was performed either in the right middle lobe or lingula. For patients with focal lung infiltrates, BAL was performed from the pulmonary segment corresponding to the radiologic abnormality. BAL continued until 40 mL fluid was retrieved or until 200 mL was instilled, whichever came first, and the fluid was analyzed according to our standard protocol, as described previously.9 BW specimens, when obtained, were sent for acid-fast bacilli smear and culture, fungal smear and culture, and cytologic analysis.

PSB specimens were obtained using a Bartlett-Faling- Wimberly brush (Mill-Rose; Mentor, OH), according to a previously described technique.10 Up to 6 transbronchial biopsies were performed in standard fashion under fluoroscopic guidance.11

Definitions and Diagnostic Criteria
Infiltrates were defined as focal if they involved one lobe or less, and diffuse(ie, either unilateral or bilateral) if they were otherwise. Pulmonary lesions were classified as lung nodules (solitary or multiple, depending on the number) when the diameter was <= 3 cm, and were defined as pulmonary masses when the diameter was > 3 cm. Hypoxemia was considered to be mild when the PaO2/fraction of inspired oxygen (FIO2) ratio exceeded 301, to be moderate when the PaO2/FIO2 ratio was 201 to 300, and to be severe when the PaO2/FIO2 ratio was < 200.

Information obtained from BAL was considered to be diagnostic when any of the following organisms were identified: Pneumocystis carinii; Mycobacterium tuberculosis; Histoplasma capsulatum; Legionella species; influenza and parainfluenza viruses; herpes simplex virus; varicella-zoster virus; respiratory syncytial virus; and mycoplasma species.8 Sarcoidosis,12 hypersensitivity pneumonitis,13 and bronchiolitis obliterans organizing pneumonia14 were diagnosed according to previously described criteria. The diagnostic criteria for other processes are presented in Table 1 .8 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34


View this table:
[in this window]
[in a new window]

  		Table 1.. Diagnostic Criteria Used in the Current Study*

 
Outcome Measures
The primary outcome measure was the yield of FB for the diagnosis of the lung infiltrate. As several patients had more than one cause of lung infiltrates, the diagnostic yield of FB was calculated as the number of diagnoses provided by FB divided by all final diagnoses. Secondary outcome measures were the diagnostic yields of the individual sampling procedures, such as BAL, PSB sampling, and TBB, and of combinations of these techniques.

Complications
All complications of FB were recorded during and immediately after the FB. Potential complications of FB include bleeding, pneumothorax, bronchospasm, fever, cardiac arrhythmia, hypoxemia, and death. Bleeding, which was classified as minor (ie, < 100 mL or clinically insignificant) or major (ie, > 100 mL or clinically significant [eg, required an RBC transfusion, caused hemodynamic instability, or led to intubation to protect the airway]), was considered to be related to FB if it occurred within 24 h after the procedure. Pneumothorax was classified as major when aspiration (eg, chest tube or Heimlich valve) was needed, or minor when aspiration was not deemed to be needed.

Statistical Analysis
All statistical analyses were performed using a statistical software package (Sigmastat, version 2.0; SPSS Inc; Chicago, IL). The proportions in two independent groups were compared using either Z-statistics with continuity correction or {chi}2 test with continuity correction. The confidence intervals (CIs) for the difference in the percentage yields of individual sampling procedures and the combinations of procedures were calculated using standard statistical techniques. Univariate analysis was used to identify the association between the complications during FB and clinically relevant parameters such as TBB, PT-INR > 1.5, a serum creatinine level of > 1.5 mg/dL, a respiratory rate of > 20 breaths/min, supplemental oxygen therapy with an FIO2 of > 40% prior to bronchoscopy, a platelet count < 50 x 109 cells/L, and mechanical ventilation received prior to bronchoscopy. A two-sided p value of <= 0.05 was considered to be significant.


   Results
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Conclusion
 References
 
Patients
Table 2 presents the demographic features of the study group. Forty-eight patients (46.2%) had hematologic malignancy. Of these, 15 and 6 patients, respectively, had received allogenic and autologous bone marrow transplantation. Fifteen patients (14.4%) had received solid organ transplantation, 9 heart transplantation, 4 renal transplantation, and 2 liver transplants. The remaining patients had collagen vascular disorders, systemic vasculitis, nontuberculosis granulomatous diseases such as sarcoidosis and Crohn disease, and other conditions requiring long-term use of immunosuppressive agents. More than 75% of patients were receiving long-term therapy with systemic corticosteroids. Nearly 60% of patients were receiving or had recently completed cancer chemotherapy. Thirty percent of patients were receiving immunosuppressive medications for nonmalignant medical illnesses.


View this table:
[in this window]
[in a new window]

  		Table 2.. Baseline Characteristics of the Patients*

 
Two thirds of patients were febrile at the time of FB. Ten patients were receiving oxygen therapy at 100% FIO2 before FB. Arterial blood gas results were available in 58 patients, of whom 25 (43.1%) had severe hypoxemia (PaO2/FIO2 ratio, < 200), 23 (39.7%) had moderate hypoxemia (PaO2/FIO2 ratio, 200 to 300), and 10 (17.2%) had mild hypoxemia (PaO2/FIO2 ratio, > 300).

FB was performed for the evaluation of diffuse lung infiltrates in 52 patients (50%), for focal lung infiltrates in 34 patients (32.6%), for multiple lung nodules in 14 patients (13.5%), and for the evaluation of a solitary lung nodule in 1 patient (1%). Three patients had normal chest roentgenogram findings, and FB was performed to evaluate unexplained fever and hypoxemia. The median duration of lung infiltrates was 5 days (range, 1 to 365 days). Most patients were receiving therapy with antimicrobial agents prior to undergoing FB, with 83 (80%) receiving IV antibiotic therapy, 39 (37.5%) receiving antifungal medications, and 28 (27%) receiving antiviral agents.

FB
Bronchoscope insertion was through an endotracheal tube in 18 patients (17.3%), nasally in 49 patients (47.1%), and orally in the remainder (in whom uncorrected coagulopathy was present). The median duration of FB was 15 min (range, 3 to 60 min).

As indicated in Table 3 , most subjects underwent more than one sampling procedure during FB. BAL was the single most common sampling procedure (performed in 95.2%; 99 patients), followed by TBB in 43.2% (45 patients) and PSB sampling in 40.4% (42 patients). No patient who underwent TBB was receiving mechanical ventilation at the time of FB, and 13 of 45 patients (28.8%) undergoing TBB had a platelet count of < 150 x 109 cells/L.


View this table:
[in this window]
[in a new window]

  		Table 3.. Sampling Procedures Performed During 104 FBs

 
Final Diagnoses
Table 4 shows the 128 final diagnoses in 104 patients. FB provided 72 of 128 diagnoses (56.2%; 95% CI, 47 to 64%), with at least one diagnosis made by FB in 53 of 104 patients (51%; 95% CI, 42 to 60%), and more than one diagnosis made by FB in 19 patients (18.3%; 95% CI, 12 to 26%). As shown in Table 4 , the most common final diagnosis was lung infection (36.7%), most frequently bacterial (49% of infections), followed by viral causes (26%), fungal causes (21% of infections), and P carinii (4%). The bacterial infections included Pseudomonas aeruginosa (six patients), Staphylococcus aureus (four patients), M tuberculosis (two patients), atypical mycobacterial infection (three patients), Escherichia coli (two patients), Xanthomonas maltophilia (two patients), Legionella species (two patients), Enterococcus faecium (one patient), and Mycoplasma species (one patient). Fungal infections included invasive pulmonary aspergillosis (five patients), semi-invasive pulmonary aspergillosis (three patients), Candida (one patient), and histoplasmosis (one patient). Viral infections were due to cytomegalovirus (CMV) [nine patients] and herpes simplex virus (three patients).


View this table:
[in this window]
[in a new window]

  		Table 4.. Etiology of 128 Final Diagnoses in 104 Study Patients*

 
Of noninfectious causes of final diagnoses, diffuse alveolar hemorrhage (DAH) accounted for 14.8%, ARDS or acute lung injury for 8.6%, and radiation or chemotherapy-induced pneumonitis for 3.9%. Miscellaneous etiologies comprised 18%. The basis for final diagnoses was FB in 56.2% (72 patients), SLB in 13.3% (17 patients), and clinical course, culture (other than bronchoscopy specimens), and serology in 30.5% (39 patients). Included in the last category were 23 patients (18%) in whom the etiology of lung infiltrates was inconclusive, and the clinicians treated these patients on the basis of overall clinical evaluation.

FB was more likely to establish the diagnosis when the infiltrate was due to infection (81%; 95% CI, 67 to 90%) than to a noninfectious etiology (55.7%, 95% CI, 43 to 67%; p = 0.011). Overall, FB was the primary source of diagnosis in 78% of bacterial infections (including three cases of mycobacterial infections), 80% of fungal infections, 83% of viral infections, and 100% of P carinii pneumonia (PCP) infection in our patients. All except one patient with bacterial infections were receiving antibiotics at the time of bronchoscopy, and the median duration of lung infiltrate prior to FB was 5 days. Blood culture results (two patients), Legionella serology (two patients), and positive sputum cultures for the same organism (three patients) supported the diagnosis provided by FB in seven patients with bacterial infections. CMV antigenemia was present in six of eight patients (75%) with CMV pneumonitis. In patients with proven bacterial infection, FB results led to the addition of a new antibiotic agent in eight patients, to the withdrawal of antibiotic therapy in three patients, and to the addition of antituberculous treatment in three patients.

Candida species were isolated from bronchoscopy specimens in five patients, but only one of them had a lung biopsy, and it showed evidence of invasive Candida infection. In the remaining patients, Candida was thought to be due to contamination or colonization according to the study criteria. In addition, coagulase-negative Staphylococcus (three patients) and CMV (one patient) found on BAL specimens were judged to be due to airway colonization. Overall, 8 of 99 BAL procedures (8%) yielded potentially false-positive results.

The diagnostic yield of FB did not differ between focal lung infiltrates (56%; 95% CI, 40 to 72%) and diffuse lung infiltrates (59%; 95% CI, 48 to 70%; p = 0.91). Similarly, there was no difference in the diagnostic yield of FB for solid-organ transplantation patients (50%; 95% CI, 28 to 72%) vs patients with either hematologic malignancy or bone-marrow transplantation (50%; 95% CI, 37 to 63%; p = 0.77).

Table 5 summarizes the yields of individual sampling procedures performed during FB. Considered individually, BAL provided 38.4% of final diagnoses, TBB provided 37.7% of final diagnoses, and PSB sampling provided 12.7% of final diagnoses. The combination of BAL, PSB sampling, and TBB rendered a diagnosis in 86.4% of instances. BAL established a diagnosis in 65% of bacterial infections (95% CI, 43 to 84%), in 75% of viral infections (95% CI, 43 to 95%), and in 40% of fungal infections (95% CI, 17 to 69%). PSB sampling provided exclusive evidence of bacterial infection in three cases in which BAL cultures were negative. The diagnosis of DAH was made by BAL in all 17 patients in whom FB had established this diagnosis. Although TBB confirmed three cases of DAH, all such diagnoses already had been established with BAL. In the remaining two patients, SLB established the diagnosis of DAH. Six patients with DAH had a concomitant disease process, including bacterial infection (three patients), CMV pneumonia (one patient), lupus pneumonitis (one patient), and Wegener granulomatosis (one patient). Table 6 shows that BAL had a higher diagnostic yield than PSB sampling and that the addition of TBB to BAL enhanced the diagnostic yield (p < 0.001).


View this table:
[in this window]
[in a new window]

  		Table 5.. Diagnostic Yield of Sampling Procedures Performed During FB

 

View this table:
[in this window]
[in a new window]

  		Table 6.. Comparison of Diagnostic Yields of Individual and Combinations of Sampling Procedures

 
TBB provided 27 diagnoses and was the only source of diagnosis in 17 cases (63%) [Table 7 ]. All the exclusive diagnoses by TBB except two (Candida pneumonia, one patient; CMV infection, one patient) were noninfectious in origin.


View this table:
[in this window]
[in a new window]

  		Table 7.. Exclusive Diagnosis by TBB (n = 17)*

 
SLB was performed in 14 patients and established 17 diagnoses (Table 4) . Although 2 of the 17 diagnoses were suggested by FB, SLB provided the diagnosis exclusively in all instances. No instance of prolonged air leak or death was attributed to SLB in this series.

Complications of FB
Procedure-related complications (Table 8) occurred in 22 patients (21.2%; 95% CI, 15 to 31%). Bleeding was the most common complication (13.5%; 95% CI, 8 to 21%) but was minor in all except one patient who required transfusion with packed RBCs. A worsening of hypoxemia was observed in 9.6% of patients (95% CI, 6 to 17%). Two patients required short-term ventilatory support after FB due to respiratory failure. There were no deaths due to FB.

The complications after FB were encountered in 14 of 45 patients (31%; 95% CI, 20 to 46%) in whom transbronchial biopsies were performed compared with 8 of 59 patients (13.6%; 95% CI, 7 to 25%) in whom TBB was not performed. This difference was statistically significant (p = 0.036). Univariate analysis showed no association between the bronchoscopy-related complications and a PT-INR of < 1.5, a platelet count of < 50 x 109 cells/L, a serum creatinine level of > 1.5 mg/dL, a baseline respiratory rate of > 20 breaths/min, a need for an FIO2 of > 0.40 prior to FB, age > 70 years, and mechanical ventilation prior to bronchoscopy. Similarly, complications after TBB were not significantly associated with any demographic, clinical, radiologic, or laboratory characteristic.


   Discussion
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Conclusion
 References
 
This study confirms the important diagnostic role of FB in immunocompromised patients with lung infiltrates. Indeed, our main findings are as follows: (1) the overall diagnostic yield of FB was 56% in this study; (2) the diagnostic yield of FB was higher for infectious causes of lung infiltrates (81%) than for noninfectious causes of lung infiltrates (40%); (3) BAL and TBB had similar diagnostic yields, and the diagnostic yields of PSB sampling and BW were significantly lower than that of BAL; (4) the combined diagnostic yield of BAL and TBB was significantly higher than that of BAL alone; (5) SLB provided a diagnosis in most patients in whom FB results were nondiagnostic; (6) complications due to FB occurred in 21% of patients but were usually minor; and (7) procedure-related complications were more likely when TBB was performed.

The diagnostic yield of FB in immunocompromised patients with lung infiltrates has varied from 15 to 93% in prior studies.8 35 36 37 38 39 40 41 42 43 44 45 46 47 In this context, the diagnostic yield of FB in our study (56%) is similar to that reported by some investigators41 42 43 48 but lower than that reported by others.37 38 39 40 49 Heterogeneity in patient populations in different studies appears to be the most likely explanation for the wide variation in the reported yield of FB. For example, because FB has a higher diagnostic yield for infectious causes of lung infiltrates, studies that enroll a greater proportion of patients with lung infections38 39 40 50 report a higher diagnostic yield of FB than studies with a higher proportion of patients with noninfectious causes of lung infiltrates.35 36 Indeed, the 56% diagnostic yield of FB in our study appears high in the context that a large proportion of our study patients had noninfectious causes of lung infiltrates. Furthermore, the inclusion of a greater proportion of patients with PCP infection37 47 51 and the inclusion of HIV-infected patients52 53 also may have contributed to higher diagnostic yields of FB in some studies. We excluded HIV-infected patients from our study, and only two of our patients had PCP infection.

Another important explanation for the variation in the reported diagnostic yield of FB diagnosis is the manner in which diagnostic yield is calculated in different studies. For instance, in one study, patients in whom the specific etiology of lung infiltrate could not be identified were excluded from the denominator.37 Consequently, the reported diagnostic yield of FB in that study (83%) was higher than the actual diagnostic yield of FB. Other studies have used the number of FB procedures in the denominator to calculate the diagnostic yield. However, several immunocompromised patients with lung infiltrates had more than one cause of lung infiltrates. Therefore, it appears more appropriate to use all final diagnoses rather than the number of procedures in the denominator to calculate the diagnostic yield of FB. In calculating the diagnostic yield of FB, we used all final diagnoses, including 39 clinical diagnoses in the denominator. We therefore believe that our results more accurately reflect the diagnostic yield of FB in these patients than do some of the prior studies.

FB was the single most useful procedure for identifying an infectious cause of lung infiltrates in our study. FB identified the bacterial pathogens in 78% of patients in whom an underlying infectious agent was identified, fungal agents in 80%, and viral agents in 83%. However, we believe that its actual sensitivity for the diagnosis of infectious causes of lung infiltrates was lower, because FB might have failed to identify a causative organism in many cases that were subsequently included under "pulmonary infiltrates of unclear etiology." Still, our ability to identify and isolate bacterial pathogens was quite impressive considering that the majority of our patients were receiving antimicrobial therapy prior to undergoing FB. In this regard, our findings concur with those of prior studies in which FB had similar success with the isolation of bacterial pathogens in patients receiving antibiotic therapy.40 41 42 43 It is possible that the high success rate of FB in identifying bacterial pathogens despite antibiotic therapy may be related to the high bacterial load in the lower respiratory tract secretions or to the higher incidence of infection with virulent organisms that respond slowly to antimicrobial therapy in immunocompromised hosts.16 We believe that the optimal yield for bacterial pathogens also requires careful attention to technical details to prevent the contamination of BAL specimens from upper airway secretions, prompt transport of specimen to the laboratory, and the support of an advanced microbiology laboratory with experience in processing and interpreting semi-quantitative cultures. Among the noninfectious causes of lung infiltrates, FB had a high yield for the diagnosis of DAH (89%), as reported in previous studies.37 38 44

In contrast to the findings of Lanino et al,54 but in keeping with the results from White et al,36 the diagnostic yield of FB was similar for patients with focal and diffuse lung infiltrates in our study. We also failed to confirm the results of a prior study42 that reported a higher diagnostic yield of FB for solid organ transplant recipients compared with patients with hematologic malignancy.

Our data provide useful information on the diagnostic yields of individual sampling procedures during FB in immunocompromised patients with lung infiltrates. BAL had a good diagnostic yield for bacterial and viral infections, but BAL results identified only 4 of 10 fungal infections. Also, semi-quantitative BAL cultures identified Candida species in four additional patients, but these patients did not meet the criterion of Candida pneumonia because lung biopsy was not performed. Unfortunately, the clinical significance of the isolation of Candida species from BAL remains controversial, as contamination from upper airways or colonization of lower airway is difficult to distinguish from invasive Candida pneumonia.39 Further work is needed in this regard.

In keeping with the results of Rano and coworkers41 (who reported only one diagnosis obtained by PSB sampling in addition to the diagnosis provided by BAL in 125 patients), our results suggest that PSB sampling adds little to the diagnostic yield of BAL and TBB. Based on these data, we question the value of routine PSB sampling in addition to BAL in immunocompromised patients with lung infiltrates. Still, PSB sampling may be considered when the clinical features strongly suggest bacterial pneumonia as the cause of lung infiltrates. Similarly, our results do not support routine BWs in this setting, provided that a BAL sample is obtained and tested comprehensively.

The diagnostic yield of TBB has varied from 26 to 68% in prior studies.23 37 53 55 56 57 58 59 60 61 However, routine TBB is not performed in immunocompromised patients due to safety issues.49 Furthermore, the clinical value of information obtained by TBB in these patients is not well-established.36 In our study, TBB was performed in 43% of patients, and its diagnostic yield was 38%. TBB provided 17 exclusive diagnoses, most of which were noninfectious. The addition of TBB to BAL in our study significantly improved the diagnostic yield of FB, confirming the complementary role of BAL and TBB in this setting.37 42 53 Based on these results, we favor performing TBB when it is deemed safe in this setting.

The diagnostic yield of SLB has been estimated to be 46 to 85% in immunocompromised patients with lung infiltrates.62 63 64 65 66 67 68 However, because of a higher potential for complications and higher Scost, SLB is typically performed after BAL and/or TBB have failed to provide the diagnosis. Our study confirms that SLB provides specific diagnoses in most patients in whom FB is nondiagnostic.69 70 We encountered no major procedure-related complications in any patient who underwent SLB. Based on our results, we recommend early SLB when the initial FB is nondiagnostic and when there is no response to empiric treatment.

Most immunocompromised patients who require FB are critically ill and have several comorbid conditions that place them at a high risk for procedure-related complications. Complications during FB in our study were more likely to occur in patients who underwent TBB, which is expected in view of the more invasive nature of TBB compared with BAL or PSB sampling. However, none of our patients died due to bronchoscopy-related complications. In contrast to the results of Cunningham et al,58 renal insufficiency was not associated with more frequent procedure-related complications in our patients. Also, we did not confirm independent association between elevated PT-INR and bronchoscopy-related complications, as was previously reported.43 However, our results in this regard need to be interpreted with caution due to the limited number of patients with elevated PT-INRs (ie, > 1.5) who underwent FB.

Several limitations of our study should be noted. First, our conclusions about the safety of TBB may not generalize because TBB was not performed on every patient in this series. Although it is difficult to identify all factors that may have influenced the decision not to perform TBB, it is reasonable to presume that the assessment of procedure safety in the individual patient must have played an important role. Such bias would cause these results to underestimate the complication rate in an unselected series.

A second limitation of this study is that most patients were receiving antimicrobial agents when bronchoscopy was performed. Although direct evidence is not available in non-HIV immunocompromised patients, empiric antimicrobial therapy impairs the diagnostic yield of FB in HIV-infected patients71 and in immunocompetent individuals with nosocomial infection.72 While the use of antimicrobial agents prior to FB could have reduced its diagnostic yield in our patients, we recognize the important role of administering antimicrobial agents promptly in these patients.

A third limitation is that 30% of all final diagnoses in this series were based on clinical course or outcome. Furthermore, in 23 patients (18%), no firm diagnosis was established. Nevertheless, all except one patient in this group survived and showed an adequate response to empiric treatment. It is important to note that prior investigators have also reported37 45 a failure to establish a final diagnosis in 14 to 31% of immunocompromised patients with lung infiltrates.

Fourth, even though our study deals with a diagnostic test (FB), we are unable to determine the true sensitivity or specificity of FB due to the lack of a "gold standard" test in this setting. Even SLB cannot identify the etiology of lung infiltrates in every immunocompromised patient. For this reason, neither our results nor the results from prior studies can accurately establish the sensitivity, specificity, and predictive value of FB in these patients. Nevertheless, we believe that our study provides useful information to treating clinicians regarding the value of FB in this difficult clinical setting.

Finally, our patients had varying causes of immunosuppression (other than HIV infection) and a wide variety of lung processes. Patients with solid tumors receiving chemotherapy were underrepresented in our patient population. Furthermore, because the number of patients in each diagnostic category in our study was small, our data regarding the yield of FB for the individual diagnostic categories (eg, bacterial infections or DAH) must be interpreted with caution.


   Conclusion
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Conclusion
References
 
In conclusion, our study supports the role of FB as the initial procedure of choice for identifying the cause of lung infiltrates in non-HIV immunocompromised patients. In this series, the rate of serious complications of FB was low, and FB accurately established the etiology of the lung infiltrate in 56% of cases. Because our results suggest that BAL and TBB are complementary, we recommend performing TBB whenever it is deemed safe. In contrast, PSB sampling and BW add little to BAL and TBB. SLB should be performed when the FB is nondiagnostic, especially when the patient is not responding to empiric treatment.


   Footnotes
 
Abbreviations: BW = bronchial washing; CI = confidence interval; CMV = cytomegalovirus; DAH = diffuse alveolar hemorrhage; FB = flexible bronchoscopy; FIO2 = fraction of inspired oxygen; PCP = Pneumocystis carinii pneumonia; PSB = protected-specimen brush; PT-INR = prothrombin time-international normalized ratio; SLB = surgical lung biopsy; TBB = transbronchial biopsy

Received for publication December 11, 2002. Accepted for publication August 12, 2003.


   References
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Conclusion
 References
 

  1. Breuer, R, Lossos, IS, Berkman, N, et al (1993) Pulmonary complications of bone marrow transplantation. Respir Med 87,571-579[CrossRef][ISI][Medline]
  2. Rosenow, EC, III, Wilson, WR, Cockrill, FR, III Pulmonary diseases in the immunocompromised host: 1. Mayo Clin Proc 1985;60,473-487[ISI][Medline]
  3. Cordonnier, C, Bernaudin, J-F, Bierling, P, et al Pulmonary complications occurring after allogenic bone marrow transplantation: a study of 130 consecutive transplanted patients. Cancer 1986;58,1047-1054[CrossRef][ISI][Medline]
  4. Ramsey, PG, Rubin, RH, Tolkoff-Rubin, NE, et al The renal transplant patient with fever and pulmonary infiltrates: etiology, clinical manifestations, and management. Medicine (Baltimore) 1980;59,206-222[Medline]
  5. Jules-Elysee, K, Stover, DE, Yahalom, J, et al Pulmonary complications in lymphoma patients treated with high-dose therapy and autologous bone marrow transplantation. Am Rev Respir Dis 1992;146,485-491[ISI][Medline]
  6. Singer, C, Armstrong, D, Rosen, PP, et al Diffuse pulmonary infiltrates in immunosuppressed patients: prospective study of 80 cases. Am J Med 1979;66,110-120[CrossRef][ISI][Medline]
  7. Shelhamer, JH, Toews, GB, Masur, H, et al Respiratory disease in the immunosuppressed patient. Ann Intern Med 1992;117,415-431[ISI][Medline]
  8. American Thoracic Society. Clinical role of bronchoalveolar lavage in adults with pulmonary disease. Am Rev Respir Dis 1990;142,481-486[ISI][Medline]
  9. Lee, FYW, Mehta, AC Basic techniques in flexible bronchoscopy. Wang, KP Mehta, AC eds. Flexible bronchoscopy 1995,95-118 Blackwell Science. Cambridge, UK:
  10. Wimberley, WA, Faling, LJ, Bartlett, JG A fiberoptic bronchoscopy technique to obtain uncontaminated lower airway secretions for bacterial culture. Am Rev Respir Dis 1979;119,337-343[ISI][Medline]
  11. Ahmad, M, Livingston, DR, Golish, JA, et al The safety of outpatient transbronchial biopsy. Chest 1986;90,403-405[Abstract/Free Full Text]
  12. Koerner, SK, Sakowitz, AJ, Appelman, RI, et al Transbronchial lung biopsy for the diagnosis of sarcoidosis. N Engl J Med 1975;293,268-270[Abstract]
  13. Coleman, A, Colby, TV Histologic diagnosis of extrinsic allergic alveolitis. Am J Surg Pathol 1988;12,514-518[ISI][Medline]
  14. Poletti, V, Cazzato, S, Minicuci, N, et al The diagnostic value of bronchoalveolar lavage and transbronchial lung biopsy in cryptogenic organizing pneumonia. Eur Respir J 1996;9,2513-2516[Abstract]
  15. Thorpe, JE, Baughman, RP, Frame, PT, et al Bronchoalveolar lavage for diagnosing acute bacterial pneumonia. J Infect Dis 1987;155,855-861[ISI][Medline]
  16. Kahn, FW, Jones, JM Diagnosing bacterial respiratory infection by bronchoalveolar lavage. J Infect Dis 1987;155,862-869[ISI][Medline]
  17. Baselski, VS, El-Torkey, M, Coalson, JJ, et al The standardization of criteria for processing and interpreting laboratory specimens in patients with suspected ventilator associated pneumonia. Chest 1992;102,S571-S579
  18. Chenez, P, Aubas, P, Segondy, G, et al Approaches to diagnosis of viral pneumonias in immunocompromised host: importance of assaying cytopathogenic viral effects in bronchoalveolar lavage cells. Eur Respir J 1988;1,553-557[Abstract]
  19. Paradis, IL, Grgurich, WF, Dummer, S, et al Rapid detection of cytomegalovirus pneumonia from lung lavage cells. Am Rev Respir Dis 1988;138,697-702[ISI][Medline]
  20. Ruutu, P, Ruutu, T, Volin, L, et al Cytomegalovirus is frequently isolated in bronchoalveolar lavage fluid of bone marrow transplant recipients without pneumonia. Ann Intern Med 1990;112,913-916[ISI][Medline]
  21. Stout, JE, Yu, VL Legionellosis. N Engl J Med 1997;337,682-687[Free Full Text]
  22. Levy, H, Horak, DA, Tegtmeier, BR, et al The value of bronchoalveolar lavage and bronchial washings in the diagnosis of invasive pulmonary aspergillosis. Respir Med 1992;86,243-248[ISI][Medline]
  23. Albelda, SM, Talbot, GH, Gerson, SL, et al Role of fiberoptic bronchoscopy in the diagnosis of invasive pulmonary aspergillosis in patients with acute leukemia. Am J Med 1984;76,1027-1034[CrossRef][ISI][Medline]
  24. Gefter, WB, Weingrad, TR, Epstein, DM, et al Semi-invasive pulmonary aspergillosis: a new look at the spectrum of Aspergillus infections of the lung. Radiology 1981;140,313-321[Abstract/Free Full Text]
  25. Dubois, PJ, Myerowitz, RL, Allen, CM Pathoradiologic correlation of pulmonary candidiasis in immunosuppressed patients. Cancer 1977;40,1026-1036[CrossRef][ISI][Medline]
  26. Heurlin, N, Bergstrom, S-E, Winiarski, J, et al Fungal pneumonia: the predominant lung infection causing death in children undergoing bone marrow transplantation. Acta Paediatr 1996;85,166-172
  27. Baughman, RP Use of bronchoscopy in the diagnosis of infection in the immunocompromised host. Thorax 1994;49,5-9
  28. American Thoracic Society. Diagnosis and treatment of disease caused by nontuberculous mycobacteria. Am J Respir Crit Care Med 1997;156,S1-S25
  29. De Lassence, A, Fleury-Feith, J, Escudier, E, et al Alveolar hemorrhage: diagnostic criteria and results in 194 immunocompromised hosts. Am J Respir Crit Care Med 1995;151,157-163[Abstract]
  30. Poe, RH, Utell, MJ, Israel, RH, et al Sensitivity and specificity of the non-specific transbronchial lung biopsy. Am Rev Respir Dis 1979;119,25-31[ISI][Medline]
  31. Smith, GJ The histopathology of pulmonary reaction to drugs. Clin Chest Med 1990;11,95-117[ISI][Medline]
  32. Huang, M-S, Colby, TV, Goellner, JR, et al Utility of bronchoalveolar lavage in the diagnosis of drug-induced pulmonary toxicity. Acta Cytol 1989;33,533-538[ISI][Medline]
  33. Movsas, B, Raffin, TA, Epstein, AH, et al Pulmonary radiation injury. Chest 1997;111,1061-1067[Free Full Text]
  34. Bernard, GR, Artigas, A, Brigham, KL, et al The American-European consensus conference on ARDS: definitions, mechanisms, relevant outcomes and clinical trial coordination. Am J Respir Crit Care Med 1994;149,818-824[Abstract]
  35. Martin, WJ, II, Smith, TF, Sanderson, DR, et al Role of bronchoalveolar lavage in the assessment of opportunistic pulmonary infections: utility and complications. Mayo Clin Proc 1987;62,549-557[ISI][Medline]
  36. White, P, Bonacum, JT, Miller, CB Utility of fiberoptic bronchoscopy in bone marrow transplant patients. Bone Marrow Transplant 1997;20,681-687[CrossRef][ISI][Medline]
  37. Stover, DE, Zaman, MB, Hajdu, SI, et al Bronchoalveolar lavage in the diagnosis of diffuse pulmonary infiltrates in the immunosuppressed host. Ann Intern Med 1984;101,1-7[ISI][Medline]
  38. Glazer, M, Breuer, R, Berkman, N, et al Use of fiberoptic bronchoscopy in bone marrow transplantation recipients. Acta Haematol 1998;99,22-26[CrossRef][ISI][Medline]
  39. Marra, R, Pagano, L, Pagliari, G, et al The yield of bronchoalveolar lavage in the etiological diagnosis of pneumonia in leukemia and lymphoma patients. Eur J Haematol 1993;51,256-258[ISI][Medline]
  40. von Eiff, M, Zuhlsdorf, M, Roos, N, et al Pulmonary infiltrates in patients with hematologic malignancies: clinical usefulness of non-invasive bronchoscopic procedures. Eur J Haematol 1995;54,157-162[ISI][Medline]
  41. Rano, A, Agusti, C, Jimenez, P, et al Pulmonary infiltrates in non-HIV immunocompromised patients: a diagnostic approach using non-invasive and bronchoscopic procedures. Thorax 2001;56,379-387[Abstract/Free Full Text]
  42. Eriksson, B-M, Dahl, H, Wang, F-Z, et al Diagnosis of pulmonary infections in immunocompromised patients by fiberoptic bronchoscopy with bronchoalveolar lavage and serology. Scand J Infect Dis 1996;28,479-485[ISI][Medline]
  43. Dunagan, DP, Baker, AM, Hurd, DD, et al Bronchoscopic evaluation of pulmonary infiltrates following bone marrow transplantation. Chest 1997;111,135-141[Abstract/Free Full Text]
  44. Huaringa, AJ, Leyva, FJ, Signes-Costa, J, et al Bronchoalveolar lavage in the diagnosis of pulmonary complications of bone marrow transplant patients. Bone Marrow Transplant 2000;25,975-979[CrossRef][ISI][Medline]
  45. Pisani, RJ, Wright, AJ Clinical utility of bronchoalveolar lavage in immunocompromised hosts. Mayo Clin Proc 1992;67,221-227[ISI][Medline]
  46. Cordonnier, C, Bernaudin, J-F, Fleury, J, et al Diagnostic yield of bronchoalveolar lavage in pneumonitis occurring after allogenic bone marrow transplantation. Am Rev Respir Dis 1985;132,1118-1123[ISI][Medline]
  47. Young, JA, Hopkin, JM, Cuthbertson, WP Pulmonary infiltrates in immunocompromised patients: diagnosis by cytological examination of bronchoalveolar lavage fluid. J Clin Pathol 1984;37,390-397[Abstract/Free Full Text]
  48. Sternberg, RI, Baughman, RP, Dohn, MN, et al Utility of bronchoalveolar lavage in assessing pneumonia in immunosuppressed renal transplant recipients. Am J Med 1993;95,358-364[CrossRef][ISI][Medline]
  49. Springmeyer, SC, Hackman, RC, Holle, R, et al Use of bronchoalveolar lavage to diagnose acute diffuse pneumonia in the immunocompromised host. J Infect Dis 1986;154,604-610[ISI][Medline]
  50. Campbell, JH, Blessing, N, Burnett, AK, et al Investigation and management of pulmonary infiltrates following bone marrow transplantation: an eight-year review. Thorax 1993;48,1248-1251[Abstract]
  51. Kahn, FW, Jones, JM Analysis of bronchoalveolar lavage specimens from immunocompromised patients with a protocol applicable in the microbiology laboratory. J Clin Microbiol 1988;26,1150-1155[Abstract/Free Full Text]
  52. de Blic, J, Mckelvie, P, Bourgeosis, ML, et al Value of bronchoalveolar lavage in the management of severe acute pneumonia and interstitial pneumonitis in the immunocompromised child. Thorax 1987;42,759-765[Abstract]
  53. Cazzadori, A, Di Perri, G, Todeschini, G, et al Transbronchial biopsy in the diagnosis of pulmonary infiltrates in immunocompromised patients. Chest 1995;107,101-106[Abstract/Free Full Text]
  54. Lanino, E, Sacco, O, Kotitsa, Z, et al Fiberoptic bronchoscopy and bronchoalveolar lavage for the evaluation of pulmonary infiltrates after BMT in children. Bone Marrow Transplant 1996;2,117-120
  55. Matthay, RA, Farmer, WC, Odero, D Diagnostic fiberoptic bronchoscopy in the immunocompromised host with pulmonary infiltrates. Thorax 1977;32,539-545[Abstract]
  56. Puska, S, Hutcheon, MA, Hyland, RH Usefulness of transbronchial biopsy in immunosuppressed patients with pulmonary infiltrates. Thorax 1983;38,146-150[Abstract]
  57. Springmeyer, SC, Silvestri, RC, Sale, GE, et al The role of transbronchial biopsy for the diagnosis of diffuse pneumonias in immunocompromised marrow transplant recipients. Am Rev Respir Dis 1982;126,763-765[ISI][Medline]
  58. Cunningham, JH, Zavala, DC, Corry, RJ, et al Trephine air drill, bronchial brush, and fiberoptic transbronchial lung biopsies in immunosuppressed patients. Am Rev Respir Dis 1977;115,213-220[ISI][Medline]
  59. Feldman, NT, Pennington, JE, Ehrie, MG Transbronchial lung biopsy in the compromised host. JAMA 1977;238,1377-1379[Abstract]
  60. Nishio, JN, Lynch, JP, III Fiberoptic bronchoscopy in the immunocompromised host: the significance of a nonspecific transbronchial biopsy. Am Rev Respir Dis 1980;121,307-312[ISI][Medline]
  61. Pennington, JE, Feldman, NT Pulmonary infiltrates and fever in patients with hematologic malignancy: assessment of transbronchial biopsy. Am J Med 1977;62,581-587[CrossRef][ISI][Medline]
  62. Canham, EM, Kennedy, TC, Merrick, TA Unexplained pulmonary infiltrates in the compromised patient: an invasive investigation in a consecutive series. Cancer 1983;52,325-329[CrossRef][ISI][Medline]
  63. Kramer, MR, Berkman, N, Mintz, B, et al The role of open lung biopsy in the management and outcome of patients with diffuse lung disease. Ann Thorac Surg 1998;65,198-202[Abstract/Free Full Text]
  64. White, DA, Wong, PW, Downey, R The utility of open lung biopsy in patients with hematologic malignancies. Am J Respir Crit Care Med 2000;161,723-729[Abstract/Free Full Text]
  65. Jaffe, JP, Maki, DG Lung biopsy in immunocompromised patients: one institution’s experience and an approach to management of pulmonary disease in the compromised host. Cancer 1981;48,1144-1153[CrossRef][ISI][Medline]
  66. Catterall, JR, Mccabe, RE, Brooks, RG, et al Open lung biopsy in patients with Hodgkin’s disease and pulmonary infiltrates. Am Rev Respir Dis 1989;139,1274-1279[ISI][Medline]
  67. Cockerill, FR, III, Wilson, WR, Carpenter, HA, et al Open lung biopsy in immunocompromised patients. Arch Intern Med 1985;145,1398-1404[Abstract]
  68. Leight, GS, Jr, Michaelis, LL Open lung biopsy for the diagnosis of acute, diffuse pulmonary infiltrates in the immunosuppressed patient. Chest 1978;73,477-482[Free Full Text]
  69. Ellis, ME, Spence, D, Bouchama, A, et al Open lung biopsy provides a higher and more specific yield compared to bronchoalveolar lavage in immunocompromised patients. Scand J Infect Dis 1995;27,157-162[ISI][Medline]
  70. Toledo-Pereyra, LH, DeMeester, TR, Kinealey, A, et al The benefits of open lung biopsy in patients with previous non-diagnostic transbronchial lung biopsy: a guide to appropriate therapy. Chest 1980;77,647-650[Abstract/Free Full Text]
  71. De Gracia, J, Miravitlles, M, Mayordomo, C, et al Empiric treatments impair the diagnostic yield of BAL in HIV-positive patients. Chest 1997;111,1180-1186[Abstract/Free Full Text]
  72. Souweine, B, Veber, B, Bedos, JP, et al Diagnostic accuracy of protected specimen brush and bronchoalveolar lavage in nosocomial pneumonia: impact of previous antimicrobial treatments. Crit Care Med 1998;26,236-244[CrossRef][ISI][Medline]



This article has been cited by other articles:


Home page
 ChestHome page
A. Holley, D. Mayes, and R. Browning
A 40-Year-Old Man With Neutropenic Fever and Lobar Consolidation
Chest, March 1, 2008; 133(3): 816 - 819.
[Full Text] [PDF]

Home page
 ChestHome page
D. Stolz, A. Stulz, B. Muller, A. Gratwohl, and M. Tamm
BAL Neutrophils, Serum Procalcitonin, and C-Reactive Protein To Predict Bacterial Infection in the Immunocompromised Host
Chest, August 1, 2007; 132(2): 504 - 514.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (27)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Jain, P.
Right arrow Articles by Mehta, A. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Jain, P.