|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Division of Pulmonary and Critical Care Medicine (Drs. Parambil and Ryu) and Department of Laboratory Medicine and Pathology (Dr. Aubry), Mayo Clinic, Rochester, MN; and Department of Pathology (Dr. Myers), University of Michigan, Ann Arbor, MI.
Correspondence to: Jay H. Ryu, MD, FCCP, Division of Pulmonary and Critical Care Medicine, Desk East 18, Mayo Clinic, 200 First St SW, Rochester, MN 55905; e-mail: ryu.jay{at}mayo.edu
Abstract
Background: Diffuse alveolar damage (DAD) is a relatively common histopathologic finding at autopsy, particularly in patients dying with ARDS, and can result from a variety of causes. The spectrum of causes and associated prognostic implications for DAD diagnosed by surgical lung biopsy are unclear.
Methods: We identified 58 consecutive patients with DAD diagnosed by surgical lung biopsy over a 7-year period, January 1996 through December 2002. The presenting clinicoradiologic features, causes, and clinical course of these patients were studied.
Results: The median age was 61 years, 48% were women, and 60% were immunocompromised. Ninety percent of patients fulfilled the criteria for ARDS at the time of surgical lung biopsy. Chest radiography demonstrated bilateral parenchymal infiltrates, while CT revealed predominantly ground-glass and consolidative opacities. Infections were the most common cause of DAD (22%). Other causes were noninfectious pulmonary complications of hematopoietic stem-cell or solid-organ transplantation (17%), connective tissue diseases (16%), acute exacerbation of idiopathic pulmonary fibrosis (12%), drugs (10%), and radiation therapy (2%). Twelve patients (21%) had acute interstitial pneumonia (ie, no identifiable cause or predisposing condition for DAD). Overall hospital mortality was 53%, with the highest mortality (86%) occurring among patients for whom DAD represented acute exacerbation of idiopathic pulmonary fibrosis.
Conclusion: Our study showed that infections and acute interstitial pneumonia are the most common causes of DAD diagnosed by surgical lung biopsy. Hospital mortality rate associated with DAD may vary depending on the underlying cause.
Key Words: interstitial lung disease interstitial pneumonia pulmonary fibrosis respiratory failure
Diffuse alveolar damage (DAD) is a relatively common histopathologic pattern seen on surgical lung biopsies and can be caused by infections, drugs, inhalational injuries, connective tissue diseases, and other forms of insult.1234 It is the most common histology seen in patients with ARDS.2345678 When DAD occurs without an identifiable cause or underlying condition, it is referred to as acute interstitial pneumonia (AIP).910
Previous studies568 on the causes of DAD were based on autopsy data and may not necessarily reflect the spectrum of causes encountered clinically in patients who undergo diagnostic lung biopsy. In addition, DAD found at autopsy may relate to common preterminal events in the clinical course of critically ill patients such as shock, disseminated intravascular coagulation, sepsis, or oxygen toxicity.11112131415 In this study, we sought to identify the causes and underlying clinical conditions in a consecutive series of patients with DAD diagnosed on surgical lung biopsy, and to assess associated prognostic implications.
Materials and Methods
Patient Selection
A computer-aided search was conducted to identify all adults (
18 years old) seen at the Mayo Clinic in Rochester, MN, during a 7-year period from January 1, 1996 to December 31, 2002, with a diagnosis of DAD on surgical lung biopsy performed at our institution. We identified 58 patients, all of whom underwent surgical lung biopsy for diffuse lung infiltrates of undetermined cause. The Mayo Foundation Institutional Review Board approved this study. Patients who did not authorize the use of their medical records for research were excluded from this study.
Fifteen of 58 patients included in this report have been described in previous publications.1617 These 15 patients include 6 of 7 patients with idiopathic pulmonary fibrosis and all 9 patients with connective tissue diseases.
Histopathology
Surgical lung biopsy slides were retrieved and reviewed independently by two pulmonary pathologists (J.L.M. and M.C.A.) without knowledge of clinical or radiologic data. In all cases, the diagnosis of DAD was assigned by both pathologists. Lung biopsy specimens were obtained by thoracotomy in 42 patients (72%) and by video-assisted thoracoscopic surgery in 16 patients (28%). The histopathologic diagnosis of DAD was made using criteria outlined in the American-European consensus statement on idiopathic interstitial pneumonias.3 Gomori methenamine-silver stain for fungi was performed in 57 cases, auramine-rhodamine stain for acid-fast bacilli in 34 cases, and Calcofluor white fluorochrome stain for Pneumocystis in 17 cases.
Clinical, Laboratory, and Radiologic Findings
Data extracted from the medical records included demographics, clinical presentation, physical findings, laboratory results, radiologic findings, diagnosis, treatment, and clinical course. Presenting signs and symptoms were recorded from the first encounter at the Mayo Clinic that led to a diagnosis of DAD. Immunocompromised status was defined by corticosteroid therapy (prednisone
15 mg/d or equivalent) or chemotherapy, HIV infection, leukemia, and hematopoietic stem-cell or solid-organ transplantation managed with immunosuppressive therapy. ARDS and acute lung injury were defined using the American-European Consensus Conference criteria.18 Diagnostic criteria for ARDS included acute onset of bilateral pulmonary infiltrates and severe hypoxemia (PaO2/fraction of inspired oxygen [FIO2] ratio
200 mm Hg) regardless of applied positive end-expiratory pressure level in the absence of congestive heart failure as an etiologic factor (pulmonary artery occlusion pressure
18 mm Hg or no clinical evidence of heart failure). Acute lung injury was similarly defined except for the criterion of less severe hypoxemia (PaO2/FIO2, 201 to 300 mm Hg). Cause of DAD was assigned by consensus of the two of the authors (J.G.P., J.H.R.) based on the review of all available clinical, laboratory, radiologic, and biopsy data.
Statistical Analysis
Data are summarized using mean ± SD, median, and range for continuous variables, and frequency and percentage for categorical variables. Age and APACHE (acute physiology and chronic health evaluation) II score associated with etiologic categories were compared using the Kruskal-Wallis test. We used logistic regression analysis to examine the independent effect of etiologic subgroup on hospital mortality while controlling for APACHE II score. Two-tailed p values < 0.05 were considered statistically significant.
Results
Demographics and Clinical Presentation
The median age of the patients was 61 years; 28 patients (48%) were women (Table 1
). All patients had respiratory symptoms, most commonly exertional dyspnea (100%), followed by cough (76%), fever (59%), chest pain (14%), and hemoptysis (7%). On physical examination, the most common finding was inspiratory crackles (98%), followed by expiratory wheezes (38%) and digital clubbing (12%). Digital clubbing was noted in four patients with idiopathic pulmonary fibrosis and three patients with connective tissue disease-related interstitial lung disease (all three had usual interstitial pneumonia). The duration of symptomatic illness ranged from 5 to 28 days (median, 10 days) prior to presentation.
|
Fifty-six patients (97%) were hospitalized for a median duration of 8 days (range, 3 to 12 days) prior to surgical lung biopsy; 48 of these patients (83% of total) were receiving invasive mechanical ventilation in the ICU. Fifty-two patients (90%) met the American-European Consensus Conference criteria for ARDS, while 4 patients (7%) met the criteria for acute lung injury. Two remaining patients (3%) were evaluated in the outpatient clinic prior to their surgical lung biopsy. One of these patients presented with persistent cough and patchy pulmonary infiltrates 3 months after completing radiation therapy to the right breast (lumpectomy for stage I breast cancer). The other patient presented with chronic polyarthralgia, cough, and intermittent fevers along with bilateral patchy pulmonary infiltrates and she was eventually discovered to have an undifferentiated connective tissue disease.
Radiologic Findings
Chest radiography was performed on all patients on presentation and revealed bilateral parenchymal infiltrates; 8 patients (14%) had pleural effusions. CT of the chest was performed on 54 patients (93%), 1 to 6 days (median, 3 days) before surgical lung biopsy. CT findings consisted most commonly of bilateral areas of ground-glass opacities (100%) and airspace consolidation (93%) [Fig 1
]. Nine patients (17% of those scanned) had underlying fibrotic changes consisting of subpleural honeycombing and traction bronchiectasis; these included six patients with idiopathic pulmonary fibrosis and three patients with connective tissue disease-related usual interstitial pneumonia. In one remaining patient with idiopathic pulmonary fibrosis, CT scan revealed reticular opacities in the peripheral lung zones associated with traction bronchiectasis but no honeycombing; surgical lung biopsy revealed evidence of underlying usual interstitial pneumonia. Additional findings included a parenchymal cavity in a patient who proved to have a lung abscess and a mediastinal abscess in another patient. Pleural effusions were seen in 13 patients (24% of those scanned) by CT.
|
Blood cultures yielded organisms in five cases (9%) and included one patient each with lung abscess (Staphylococcus aureus, Pseudomonas aeruginosa), mediastinal abscess (S aureus), central venous catheter-related sepsis (Micrococcus luteus), fungal (Cryptococcus neoformans), and viral (cytomegalovirus) pneumonia, respectively.
Serologic findings were positive in four patients (7%), with results finalized after surgical lung biopsy in all. These included one case each of influenza A, influenza B, herpes simplex virus type 1, and Mycoplasma pneumoniae.
Bronchoscopy Results
Bronchoscopy with BAL was performed prior to surgical lung biopsy in all patients 0 to 8 days (median, 4 days) following hospital admission. Neutrophils were the predominant cell type identified in the BAL fluid, with a mean percentage of 62 ± 22%. Bronchoscopic lung biopsies were performed in 17 patients (29%) and revealed features suggestive of organizing pneumonia in 5 patients (9%), diffuse alveolar damage in 3 patients (5%), nonspecific interstitial inflammation and fibrosis in 3 patients (5%), alveolar hemorrhage in 1 patient (2%), and was unrevealing in the remaining 5 patients (9%).
Microbiologic analysis of BAL fluid yielded positive results in eight patients (14%) and included five cases of viral pneumonia (one case each of respiratory syncytial virus, influenza A, influenza B, Herpes simplex virus type 1, and cytomegalovirus) and one case each of fungal pneumonia (Cryptococcus neoformans), mycobacterial pneumonia (Mycobacterium avium complex), and polymicrobial lung abscess (S aureus, Pseudomonas aeruginosa, and Klebsiella pneumoniae). In five of these eight patients, positive culture results (influenza A, Herpes simplex virus type 1, cytomegalovirus, and C neoformans, and M avium complex) were known before surgical lung biopsy; the surgical lung biopsy was performed in these patients because the clinical relevance of the microbial organism to the ongoing pulmonary process was questioned or another undiagnosed process was suspected because of progressive lung infiltrates despite appropriate antimicrobial therapy.
Surgical Lung Biopsy and Complications
All patients underwent a surgical lung biopsy for further evaluation of diffuse pulmonary infiltrates with biopsy specimens obtained from one lobe (31 patients), two lobes (25 patients), or three lobes (2 patients); 2 patients (3%) also underwent surgical treatment including a lobectomy for lung abscess and sternotomy with debridement, irrigation, and drainage for mediastinal abscess.
Surgical lung biopsy yielded clues to the cause for DAD in six cases (10%). These included three patients with acute exacerbation of idiopathic pulmonary fibrosis in whom the underlying chronic lung disease had not been previously diagnosed and three cases of infections. Infections included two patients with histologic evidence of cytomegalovirus pneumonia despite negative preoperative viral blood culture and BAL specimen findings, and one patient with associated necrotizing granulomas containing rare acid-fast bacilli in whom cultures grew M avium complex. All stains for fungi and Pneumocystis were negative.
Other findings on surgical lung biopsy included focal acute bronchopneumonia in nine patients, including one patient with serological evidence of influenza infection. Patchy areas of infarct-like necrosis were present in three patients; none had clinical evidence of venous thromboembolism, including negative CT chest angiography results in one patient. Parenchymal hemorrhage without capillaritis was seen in two patients. Amyloid deposits were seen in a patient with multiple myeloma.
Postoperative complications were noted in 13 patients (22%) and included 12 patients (20%) with prolonged air leak and 1 patient (2%) with chylothorax. These complications resolved in all 13 patients with conservative management. There were no deaths directly attributable to the surgical procedure.
Causes of DAD
The causes of DAD are displayed in Table 2
. Infectious causes were the most common and occurred in 13 patients (22%), 10 of whom were immunocompromised. Those with sepsis included one patient each with lung abscess, mediastinal abscess, and central venous catheter-related infection, respectively. In 16 patients (28%), DAD could only be attributed to the underlying disorder including connective-tissue diseases (16%) and IPF (12%). Noninfectious pulmonary complication of hematopoietic stem-cell or solid-organ transplantation occurred in 10 patients (17%). Other causes were drugs (10%) that included bleomycin, cytosine arabinoside, gemcitabine, cocaine and amiodarone, as well as radiation therapy (2%). Twelve patients (21%) had AIP with no identifiable underlying cause or predisposing condition.
|
Outcomes and Follow-up
Thirty-one patients died in the hospital (53%): 30 of 52 patients with ARDS (58% mortality), and 1 of 6 non-ARDS patients (17% mortality rate). Hospital mortality associated with various causes of DAD is shown in Table 2. Primary mechanism of death was refractory respiratory failure in 24 patients and multiorgan failure in the remaining 7 patients. Hospital mortality rate was highest (86%) for those with acute exacerbation of idiopathic pulmonary fibrosis, but there was no statistically significant difference in mortality rates across groups (p = 0.279 by logistic regression analysis). The APACHE II scores were significantly different (p = 0.038 by the Kruskal-Wallis test) among the etiologic groups (Table 2), with the highest score being associated with those with idiopathic pulmonary fibrosis, although the median ages were similar (p = 0.225 by the Kruskal-Wallis test). There was no statistically significant difference in mortality rates across groups when adjusted for APACHE II scores (p = 0.444 by logistic regression analysis). Analysis of hospital mortality rates according to underlying conditions is shown on Table 3
.
|
Discussion
This study evaluated the causes and prognostic implication of DAD diagnosed on surgical lung biopsy and found infections and AIP to be the most common causes. Other major causes of DAD in our patients were noninfectious pulmonary manifestations associated with hematopoietic stem-cell or solid-organ transplantation, connective tissue diseases, idiopathic pulmonary fibrosis, and drugs, each accounting for 10 to 17% of cases. This spectrum of causes for DAD diagnosed on surgical lung biopsy differs considerably from that for DAD diagnosed at autopsy. For example, Ferguson and colleagues8 found multiorgan failure to be the primary cause of death in 79% of their patients diagnosed to have DAD at autopsy. In the study by Esteban et al,6 the causes for DAD diagnosed at autopsy were not separately analyzed, but sepsis syndrome was listed as the most common risk factor for ARDS among their patients. Compared to these autopsy studies that were likely encumbered by confounding terminal processes at the time of death, we were able to identify discrete causes of DAD in the majority of our patients as well as cases of AIP. This difference in the spectrum of causes was, in large part, due to case selection issues: patients in the current study underwent a surgical lung biopsy for diagnostic evaluation of diffuse lung infiltrates, whereas the autopsy studies consisted of patients dying of ARDS usually in the presence of traditional risk factors associated with this syndrome.
Infections including pneumonia and sepsis are the most common causes of DAD, as noted in our study.3567 Nearly all of our patients with DAD related to infections were immunocompromised. Infections accounted for one half of the cases in which the surgical lung biopsy yielded an etiologic diagnosis.
Pulmonary complications are frequently seen at autopsy of patients with hematopoietic stem-cell transplant, and DAD is the most common histopathologic finding.1519 DAD in these patients can be associated with infectious or noninfectious etiology.15192021 Major clinical forms of noninfectious pulmonary complications in these patients include engraftment syndrome, diffuse alveolar hemorrhage syndrome, idiopathic pneumonia syndrome, and obliterative bronchiolitis.192021 Idiopathic pneumonia syndrome was seen in 12% of our patients and is defined as diffuse lung injury occurring after hematopoietic stem-cell transplantation for which an infectious cause is not identified.1922 This clinical entity likely represents a heterogeneous group of disorders resulting from acute lung injury caused by conditioning chemotherapy, radiation, and possibly occult infection. Idiopathic pneumonia syndrome is poorly responsive to corticosteroid therapy.192021 Engraftment syndrome occurs at the time of neutrophil recovery in hematopoietic stem-cell recipients and is postulated to be triggered by the release of cytokines by these neutrophils.192021 Response to prompt use of systemic corticosteroids has been reported.192021 DAD can also be seen in acute rejection of transplanted lung as in our single case.232425
DAD occurring in patients with idiopathic pulmonary fibrosis and connective tissue diseases accounted for nearly 30% of our cases. In patients with idiopathic pulmonary fibrosis, DAD can be seen in those with "acute exacerbation" of IPF as defined by sudden worsening of symptoms, hypoxemia, and new lung infiltrates without evidence of infection or cardiac failure.16262728 This acute illness can occasionally be the initial clinical presentation in some patients with IPF.16 Similar acute phenomenon have been described in patients with connective tissue disease-related ILD who similarly are at risk for acute worsening of their chronic lung disease.17 DAD may also occur de novo in patients with connective tissue disease but without preexisting ILD.1726 DAD occurring in patients with IPF and connective tissue diseases is poorly responsive to corticosteroid therapy.
DAD is one of many histopathologic patterns seen in drug-induced lung disease and has been associated with many pharmacologic agents, most commonly chemotherapy drugs such as bleomycin.2930 Response to corticosteroid therapy in patients with drug-induced DAD is variable and may depend on the inciting drug, severity of the reaction, and promptness of the therapeutic intervention. In our study, the majority of our patients with drug-induced DAD survived, as did our patient with radiation-related DAD.
AIP, one of several idiopathic interstitial pneumonias, generally presents with progressive respiratory symptoms and respiratory insufficiency occurring over a course of days to weeks.391031 AIP is distinguished from other cases of DAD by the absence of an identifiable cause or preexisting disease that predisposes to DAD.31031 In our study, digital clubbing was limited to patients with acute exacerbation of underlying fibrotic lung disease and served as a helpful clue in separating them from patients with AIP. Chest radiography in patients with AIP reveals diffuse lung infiltrates, and organizing DAD is seen on surgical lung biopsy. The mortality rate associated with AIP has ranged from 13 to 74%.103132 In our case series, AIP was the second most common cause of DAD and was associated with a hospital mortality rate of 50%. A study33 from Korea suggests that an aggressive diagnostic approach and management including early surgical lung biopsy, mechanical ventilation with lung-protective strategies, and early institution of immunosuppressive agents may improve clinical outcomes in patients with AIP.
The overall hospital mortality rate associated with DAD in our study was 53% but varied widely depending on the underlying cause. The highest mortality rate was seen among patients for whom DAD represented acute exacerbation of idiopathic pulmonary fibrosis, followed by those with noninfectious complications of transplantation. However, we were unable to demonstrate statically significant difference in mortality rates across groups with or without adjustment for APACHE II scores. A modest number of patients in our study cohort likely impaired our ability to detect differences.
Although DAD is commonly found in patients with ARDS, these two terms are not synonymous. The American-European Consensus Conference definition of ARDS18 is relatively nonspecific, and the pulmonary pathology underlying ARDS is not limited to DAD.6734 Both DAD and ARDS are seen in a wide spectrum of clinical contexts and are overlapping entities. Nonetheless, the vast majority (90%) of our patients with DAD met the criteria for ARDS.
Our study analyzed patients undergoing surgical lung biopsy for evaluation of diffuse lung infiltrates and provides a different perspective on DAD compared with previous autopsy studies examining patients dying of ARDS. This likely led to underrepresentation of traditional risk factors for ARDS such as sepsis, trauma, shock, aspiration, cardiopulmonary bypass, and acute pancreatitis in this cohort because surgical lung biopsy is unlikely to have been undertaken in the presence of such well-recognized risk factors. It also likely explains a relatively high proportion of patients with AIP (ie, DAD of unknown cause) in our study. However, our results provide new clinical insights into the spectrum of causes for DAD diagnosed on surgical lung biopsy for evaluation of diffuse lung disease.
In conclusion, our study shows that infections and AIP are the most common causes of DAD diagnosed on surgical lung biopsy, and the underlying cause may influence the clinical outcome. Pathophysiologic mechanisms leading to DAD likely vary depending on the underlying cause and may require individually tailored approaches to treatment. Further research into the pathogenesis and optimal management of DAD will need to take into consideration the heterogeneity in the evolution of this lung lesion and associated clinical sequelae.
Footnotes
Abbreviations: AIP = acute interstitial pneumonia; APACHE = acute physiology and chronic health evaluation; DAD = diffuse alveolar damage; FIO2 = fraction of inspired oxygen
The authors have no conflicts of interest to disclose.
Received for publication January 11, 2007. Accepted for publication March 22, 2007.
References
This article has been cited by other articles:
![]() |
M. C. B. Godoy, D. Nonaka, B. G. Raphael, and I. Vlahos Diffuse Ground-Glass Opacities in a Patient With Hodgkin Lymphoma and Progressive Respiratory Failure Chest, July 1, 2008; 134(1): 207 - 212. [Full Text] [PDF] |
||||
![]() |
B. W. Kinder, K. K. Brown, M. I. Schwarz, J. H. Ix, A. Kervitsky, and T. E. King Jr Baseline BAL Neutrophilia Predicts Early Mortality in Idiopathic Pulmonary Fibrosis Chest, January 1, 2008; 133(1): 226 - 232. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |