Pulmonary Nodular Lesions in Bone Marrow Transplant Recipients Impact of Histologic Diagnosis on Patient Management and Prognosis

Transcription

1 Anatomic Pathology / PULMONARY NODULAR LESIONS IN BONE MARROW TRANSPLANT RECIPIENTS Pulmonary Nodular Lesions in Bone Marrow Transplant Recipients Impact of Histologic Diagnosis on Patient Management and Prognosis H. Evin Gulbahce, MD, 1 Stefan E. Pambuccian, MD, 1 Jose Jessurun, MD, 1 Paul Woodard, MD, 4 Marie E. Steiner, MD, 2 J. Carlos Manivel, MD, 1 Stephen Hite, MD, 3 Norma K.C. Ramsay, MD, 2 and K. Scott Baker, MD 2 Key Words: Bone marrow transplant; Lung; Nodules Abstract Bone marrow transplantation is associated with numerous pulmonary complications, which may manifest as nodules. We studied 33 bone marrow transplant (BMT) recipients in whom pulmonary nodular lesions (PNLs) developed during a 5-year period and who underwent open lung biopsy (OLB) for diagnosis. Of 33 patients with PNL, 15 (45%) had pulmonary cytolytic thrombi (PCT), a recently described condition characterized histologically by occlusive vascular lesions and hemorrhagic infarcts and clinically by a favorable outcome. Clinical symptoms and radiologic abnormalities disappeared during a period of a few weeks. None of the patients died of PCT; 10 were alive at last contact. The second most common cause of PNL (8/33 [24%]) was Aspergillus infection, which was the cause of death in 6. OLB is an effective way of obtaining diagnostic tissue in BMT recipients with PNLs. Histologic examination is accurate in determining the cause of PNLs and identifying lesions that have a favorable outcome and those that require a change in treatment. Bone marrow transplantation frequently is performed to treat hematologic and nonhematologic malignant neoplasms, marrow failure, and certain congenital disorders. Infectious and noninfectious pulmonary complications are seen commonly in 30% to 60% of bone marrow transplant (BMT) recipients, and pulmonary nodules develop in some of these patients. 1-3 With high-resolution computed tomography (CT) scanning, even smaller pulmonary nodules can be detected. In immunocompromised patients, these lesions may represent diagnostic and therapeutic challenges because their cause may be different from that in immunocompetent patients. In immunocompromised hosts, lung nodules commonly are caused by bacterial or fungal infections or posttransplant lymphoproliferative disorders (PTLDs); these patients often are evaluated by transbronchial or open lung biopsy (OLB) to institute proper treatment. 4,5 We reviewed OLB specimens obtained for the evaluation of pulmonary nodular lesions (PNLs) in BMT recipients to determine their causes and the impact of the pathologic diagnosis on management and prognosis. Materials and Methods Review of the Blood and Marrow Transplant database at the University of Minnesota, Minneapolis, for the period January 1993 to December 1998 revealed 1,228 patients who underwent hematopoietic stem cell transplantation. Of these patients, 88 (7.17%) had pulmonary nodules that led to 37 OLBs (42%). All but 3 patients underwent thoracotomy to obtain tissue samples from peripheral lung nodules. In 3 cases, the cause of PNLs was determined at autopsy. Downloaded from Am J Clin Pathol 2004;121:

2 Gulbahce et al / PULMONARY NODULAR LESIONS IN BONE MARROW TRANSPLANT RECIPIENTS PNLs are described as single or multiple circumscribed lesions that are identifiable radiologically. Localized consolidations with reasonably sharp borders also were included. Formalin-fixed, paraffin-embedded, and H&E-stained slides from 3 autopsies and 33 biopsy specimens from a total of 33 patients (3 patients underwent 2 biopsies each) were available for review. Cytologic material collected within 2 weeks of the OLB was available for 12 patients: 8 patients underwent bronchoalveolar lavage (BAL); 1 underwent fineneedle aspiration (FNA) biopsy; for 2, imprint cytologic examination was performed at the time of the OLB; and for 1, imprint cytology and BAL specimens were available for review. Gomori methenamine silver stain for fungi was performed on all cytology specimens. The primary diagnoses, transplant type, and follow-up information were obtained from the Blood and Marrow Transplant database. Results Clinical and Radiologic Findings Of these patients, 25 were male and 8 were female. The average age was 13.8 years (range, 1-49 years). Twenty-five patients were younger than 18 years. Two patients underwent autologous and 31 patients underwent allogeneic bone marrow transplantation. The primary diagnoses were as follows: acute lymphoblastic leukemia, 14; acute myeloid leukemia, 3; chronic myeloid leukemia, 4; aplastic anemia, 4; immunodeficiency, 2; Fanconi anemia, 1; myelodysplastic syndrome, 1; rhabdomyosarcoma, 1; choriocarcinoma, 1; Image 1 Computed tomography scan of a patient with a pulmonary nodular lesion. Open lung biopsy showed pulmonary cytolytic thrombi. metabolic disorder, 1; and malignant lymphoma, 1. All patients had persistent fever and underwent chest radiography, CT scanning, or both followed by OLB. OLB was performed because of pulmonary nodular opacities seen on chest radiograph or CT an average of 170 days (range, 27-1,166 days) after transplantation Image 1. Most patients (all patients with pulmonary cytolytic thrombi [PCT]) had multiple nodules. None had diffuse interstitial infiltrates. In 3 patients, the histologic diagnosis was made at autopsy. Pathologic Examination and Follow-up The recently described entity, PCT, accounted for 15 (45%) of 33 cases Table 1. All but 1 of these patients were younger than 18 years. The PCT had distinct histologic features characterized by occlusive vascular lesions and hemorrhagic infarcts Image 2 and Image 3. The thrombi consisted of intensely basophilic, amorphous material that sometimes extended into the adjacent tissue through the damaged vascular wall. Entrapped WBCs frequently were identified in the thrombi. Of these 15 patients, 10 were alive after OLB (average, 52 months; range, months). Five patients with PCT died after OLB (average, 13 months; range, 0-44 months) of infection (Aspergillus species, 2; adenovirus, 1; cytomegalovirus, 1) and chronic graft-vs-host disease (GVHD; 1). These complications occurred after resolution of the clinical symptoms and radiologic abnormalities in all but 1 case. In this case, PCT were diagnosed at autopsy; the patient had cytomegalovirus (CMV) pneumonitis with diffuse alveolar damage leading to death. Eight of 33 patients with PNL had Aspergillus infection, which was the cause of death in 6 Image 4. Two patients with Aspergillus infection died of recurrent disease, 6 and 18 months after OLB, respectively. Other pulmonary conditions associated with PNLs and their follow-up (in parentheses) were as follows: 2 patients had idiopathic interstitial pneumonitis (both alive, 60 and 79 months); 2 had intra-alveolar hemorrhage (1 died of sepsis, 3 months; 1 alive, 80 months); 2 had organizing pneumonitis of unknown cause (both died of recurrent disease, 5 and 14 months); 1 had pulmonary thromboembolus (died of interstitial pneumonitis at time of diagnosis of PNL); 1 had metastatic choriocarcinoma (died of metastatic disease, 23 months after OLB) Image 5 ; 1 had hemorrhagic infarct of the lung (alive, 90 months); and 1 had a foreign body embolus (alive, 79 months). The BAL specimens were negative for fungi and Pneumocystis carinii in 8 patients, including 2 patients in whom Aspergillus infections were identified in the subsequent OLB specimen. One BAL specimen showed budding yeast in a patient in whom PCT were diagnosed by histologic examination. One FNA biopsy specimen was positive for Aspergillus species, which was confirmed on OLB. Imprint cytologic 206 Am J Clin Pathol 2004;121: Downloaded 206 from

3 Anatomic Pathology / ORIGINAL ARTICLE Table 1 Pathologic Findings and Outcome for Patients With PNLs Alive (Average Time Died of Unrelated Cause Pathologic Diagnosis After OLB, mo) Died of PNL (Average Time After OLB, mo) PCT (n = 15) 10 (52) 0 5 (13) Fungal infection (n = 8) (12) Interstitial pneumonitis (n = 2) 2 (70) 0 0 Intra-alveolar hemorrhage (n = 2) 1 (80) 0 1 (3) Organizing pneumonitis (n = 2) (10) Thromboembolism (n = 1) (0) Metastatic choriocarcinoma (n = 1) (23) Hemorrhagic infarct (n = 1) 1 (90) 0 0 Foreign body embolus (n = 1) 1 (79) 0 0 OLB, open lung biopsy; PCT, pulmonary cytolytic thrombi; PNL, pulmonary nodular lesion. examinations were performed at the time of OLB in 2 cases; both were negative for fungi and showed reactive atypia of the epithelial cells. In these cases, organizing pneumonitis and PCT, respectively, were diagnosed on histologic sections. For 1 case in which BAL and imprint cytology were performed, neither showed fungi, viral inclusions, or atypia; interstitial pneumonitis was diagnosed from the OLB specimen. Discussion In immunocompetent people, the differential diagnosis of pulmonary nodules includes primary and metastatic malignant neoplasms and, less commonly, nonmalignant conditions such as infections and inflammatory diseases. Multiple nodules favor metastases, commonly originating from solid organ tumors. 6 Secondary involvement of the lungs by lymphoma and leukemia is less likely. 7 In immunocompromised patients such as solid organ transplant recipients and patients with hematologic malignant neoplasms, pulmonary nodules are caused mostly by infections and less by PTLDs. 5,8 Bone marrow transplantation is becoming a common form of treatment for numerous malignant and nonmalignant diseases and is associated with pulmonary complications in 30% to 60% of the cases, accounting for 40% of transplantrelated mortality. 2,9 We reviewed 33 OLB specimens and 3 autopsy samples from 33 BMT recipients with PNL. Image 2 Histologic appearance of a nodule caused by pulmonary cytolytic thrombi showing a large hemorrhagic infarct; arrows point to occluded blood vessels (H&E, original magnification 2). Inset, A vessel occluded by intensely basophilic fibrillar material with entrapped nuclear fragments (H&E, original magnification 60). Image 3 Hemorrhagic infarct caused by pulmonary cytolytic thrombi (H&E, original magnification 10). Inset, An occluded vessel (H&E, original magnification 60). Downloaded from Am J Clin Pathol 2004;121:

4 Gulbahce et al / PULMONARY NODULAR LESIONS IN BONE MARROW TRANSPLANT RECIPIENTS Image 4 Hemorrhagic infarct caused by invasive aspergillosis (H&E, original magnification 10). Insets, A vessel occluded by Aspergillus hyphae (left, H&E, original magnification 40; middle, Gomori methenamine silver [GMS], original magnification 40; right, GMS, original magnification 100). Image 5 Metastatic choriocarcinoma manifesting as a post bone marrow transplantation pulmonary nodular lesion (H&E, original magnification 10; inset, H&E, original magnification 40). The recently described entity, PCT, was the most common cause of PNL in BMT recipients Of 33 patients, 15 (45%) had PCT; 14 cases were diagnosed by examination of the OLB specimen, and 1 was diagnosed at autopsy. Of the 15 PCT cases, 12 have been reported previously in a series from our institution. 10 PCT were seen only after allogeneic bone marrow transplantation and were characterized by multiple peripheral lung nodules. We did not encounter PCT in autologous BMT or solid organ transplant recipients. PCT have a unique histologic appearance characterized by vascular occlusion and hemorrhagic infarcts. Although the hemorrhagic infarcts are somewhat similar to those seen in angioinvasive fungal infections, neither the cultures nor the special tissue stains were positive for fungi in any of these cases. Of 15 patients with PCT, 13 (87%) had active GVHD at the time of OLB; GVHD had developed in 1 previously and in 1 subsequently. Obliterative bronchiolitis (constrictive bronchiolitis obliterans), a small airway disease, has been associated with GVHD and is thought to represent chronic GVHD in the lungs of BMT recipients. 13 Although lymphocytic bronchitis has been considered to represent acute GVHD in the lungs, the correlation between systemic acute GVHD and lymphocytic bronchitis has not been consistent. 9,14,15 Because PCT were seen exclusively in allogeneic BMT recipients at the time most patients have GVHD in other organs, it is possible that this entity represents a manifestation of acute GVHD in the lungs in which the target cell is the endothelium. The fact that all of our patients with PCT diagnosed by examination of OLB specimens showed clinical and radiologic improvement after increased immunosuppression also favors GVHD as the underlying pathogenic mechanism. Although 5 of these patients died an average of 13 months after the emergence of nodular lesions, they died of unrelated causes after resolution of the lung nodules. The absence of PCT described in previous series of pulmonary lesions in BMT recipients might be due to classification of these lesions as hemorrhage or hemorrhagic infarcts. Fungal infection, specifically Aspergillus infection, had the highest mortality among patients with PNL. Of 8 patients with Aspergillus infection, 6 died; in 2 of them, multiorgan involvement was detected at autopsy. PNLs are atypical radiologic manifestations for some of the conditions seen in the present study, such as alveolar hemorrhage and interstitial pneumonitis. Despite the lack of typical viral inclusions, some of these cases might represent the resolving phase of a viral infection such as CMV pneumonitis, which occasionally might manifest as PNLs. PTLD after bone marrow transplantation might manifest as pulmonary nodules. 16 The cumulative incidence of PTLD after bone marrow transplantation at 10 years has been reported between 0% and 1%, and most post bone marrow transplantation PTLDs develop within 6 months after transplantation Patients with chronic GVHD who receive T cell depleted and unrelated-donor or HLA-mismatched related-donor transplants are at increased risk of developing 208 Am J Clin Pathol 2004;121: Downloaded 208 from

5 Anatomic Pathology / ORIGINAL ARTICLE PTLD. Of 33 patients, 4 (12%) received T cell depleted marrow, and 9 (29%) of 31 allogeneic BMT recipients had fewer than 6 HLA matches, which placed them at increased risk of developing PTLD. However, none of the PNLs were related to PTLD. Both children and adults are included in our study. PCT were seen almost exclusively in children. No difference was noted for the other conditions associated with PNL for different age groups. However, the number of patients in other groups is too small to permit statistical analysis. The incidence of pulmonary nodules in BMT recipients in our institution was 7.17%. This is similar to the incidence of similar lesions after liver (7.1%), heart (10%), and lung (12.5%) transplantation. 5,8,20 In our institution, patients with fever and pulmonary nodules receive empiric treatment with broad-spectrum antibacterial and antifungal agents. Also, patients who are seronegative for CMV receive blood products from seronegative donors, and patients who are seropositive for CMV receive prophylactic treatment with acyclovir. The condition of 38% (33/88) of our patients with fever and PNL did not improve with empiric therapy. Since radiologic findings could not distinguish between different conditions causing nodular lesions, OLB was done. The diagnostic yield of OLB in BMT recipients with pulmonary infiltrates has been low, and the benefits have been questioned. 21,22 Snyder et al 22 found that 60% of OLBs yielded a specific diagnosis, but only 18% of patients demonstrated a clinical improvement attributable to OLBdirected antimicrobial alterations. No major intraoperative complications were reported. However, in this high-risk population of patients, 45% died within 1 month of the OLB. 22 In that study, however, all pulmonary infiltrates were included. In a different series of patients with hematologic malignant neoplasms (24 of 63 were BMT recipients), a specific diagnosis after the OLB could be made more commonly when focal rather than diffuse radiographic abnormalities were present (79% vs 36%). 23 Changes in therapy were made in 57% of patients, with increased survival at 1 and 3 months in those with a specific diagnosis. 23 In another study of BMT recipients, when only nodular infiltrates (5 of 12 patients) were considered, OLB resulted in the initiation or modification of antimicrobial agent administration in 2 of 5 patients and the diagnosis and treatment of PTLD in 1 of 5 patients with PNL. 24 The decision to perform OLB in immunocompromised, usually thrombocytopenic and severely ill patients should be individualized. Only one of the patients included in the present study died of complications of surgery. Three additional patients died within a week of OLB; in all 3, multiorgan involvement by Aspergillus was the cause of death. In BMT recipients with PNLs, obtaining a tissue sample for histologic diagnosis might be necessary because in some cases of fungal infection, the BAL and serologic test results are negative. 5 OLB also might lead to the diagnosis of other specific conditions and, therefore, help determine which patients do not require treatment for an infection. In a group of 31 patients with hematologic malignant neoplasms with clinically and radiologically suspected invasive pulmonary Aspergillus infection, only 17 (55%) were confirmed to have fungal infection by OLB. 25 In our study, 45% of the patients who underwent an OLB (15/33) had PCT, which might represent a manifestation of acute GVHD in the lungs that does not require antifungal therapy and has a favorable outcome. These patients were treated empirically with antifungal medications that were discontinued in many after the histologic diagnosis of PCT was established. BAL is a useful diagnostic tool in the evaluation of bone marrow transplantation related pulmonary complications. 26 However in our patients with nodular lesions, 8 underwent BAL within 2 weeks of the OLB; none of the BAL specimens were informative. Two patients with negative BAL results had Aspergillus organisms identified in tissue sections. FNA biopsy also has been reported to be useful in cases of hematologic malignant neoplasms. 27 One of our patients had an Aspergillus infection diagnosed by FNA biopsy, which was confirmed on subsequent examination of the OLB tissue sample. An OLB often is indicated for BMT recipients with PNLs that do not respond to empiric therapy. PNLs might be caused by a variety of conditions, some of which require prompt, specific treatment, while others may permit discontinuation of the empiric therapy. OLB might be more useful in the management of pediatric BMT recipients with PNLs because of the predominance of a diagnosis of PCT in this age group. From the Departments of 1 Laboratory Medicine and Pathology, 2 Pediatrics, and 3 Radiology, University of Minnesota, Fairview- University Medical Center, Minneapolis; and 4 Hematology/ Oncology, Division of Stem Cell Transplantation, St Jude Children s Research Hospital, Memphis, TN. Presented in part at the 90th annual meeting of United States and Canadian Academy of Pathology, Atlanta, GA, March 3-9, Address reprint requests to Dr Gulbahce: Dept of Laboratory Medicine and Pathology, University of Minnesota, Mayo Bldg MMC 76, FUMC, 420 Delaware St SE, Minneapolis, MN References 1. Cordonnier C, Bernaudin JF, Bierling P, et al. Pulmonary complications occurring after allogeneic bone marrow transplantation: a study of 130 consecutive transplanted patients. Cancer. 1986;58: Downloaded from Am J Clin Pathol 2004;121:

6 Gulbahce et al / PULMONARY NODULAR LESIONS IN BONE MARROW TRANSPLANT RECIPIENTS 2. Jochelson M, Tarbell NJ, Freedman AS, et al. Acute and chronic pulmonary complications following autologous bone marrow transplantation in non-hodgkin s lymphoma. Bone Marrow Transplant. 1990;6: Quabeck K. The lung as a critical organ in marrow transplantation. Bone Marrow Transplant. 1994;14(suppl 4):S19-S Crawford SW, Hackman RC, Clark JG. Biopsy diagnosis and clinical outcome of persistent focal pulmonary lesions after marrow transplantation. Transplantation. 1989;48: End A, Helbich T, Wisser W, et al. The pulmonary nodule after lung transplantation: cause and outcome. Chest. 1995;107: Andrea S, Paolo C, Ascanelli S, et al. Significance of a single pulmonary nodule in patients with previous history of malignancy. Eur J Cardiothorac Surg. 2001;20: Quint LE, Park CH, Iannettoni MD. Solitary pulmonary nodules in patients with extrapulmonary neoplasms. Radiology. 2000;217: Munoz P, Palomo J, Guembe P, et al. Lung nodular lesions in heart transplant recipients. J Heart Lung Transplant. 2000;19: Cooke KR, Krenger W, Hill G, et al. Host reactive donor T cells are associated with lung injury after experimental allogeneic bone marrow transplantation. Blood. 1998;92: Gulbahce HE, Manivel JC, Jessurun J. Pulmonary cytolytic thrombi: a previously unrecognized complication of bone marrow transplantation. Am J Surg Pathol. 2000;24: Woodard JP, Gulbahce E, Shreve M, et al. Pulmonary cytolytic thrombi: a newly recognized complication of stem cell transplantation. Bone Marrow Transplant. 2000;25: Morales IJ, Anderson PM, Tazelaar HD, et al. Pulmonary cytolytic thrombi: unusual complication of hematopoietic stem cell transplantation. J Pediatr Hematol Oncol. 2003;25: Urbanski SJ, Kossakowska AE, Curtis J, et al. Idiopathic small airways pathology in patients with graft-versus-host disease following allogeneic bone marrow transplantation. Am J Surg Pathol. 1987;11: Beschorner WE, Saral R, Hutchins GM, et al. Lymphocytic bronchitis associated with graft-versus-host disease in recipients of bone-marrow transplants. N Engl J Med. 1978;299: Sloane JP, Depledge MH, Powles RL, et al. Histopathology of the lung after bone marrow transplantation. J Clin Pathol. 1983;36: Tolar J, Coad JE, Ramsay NK, et al. Lymphoproliferative disorder presenting as pulmonary nodules after bone marrow transplantation. Bone Marrow Transplant. 2001;28: Curtis RE, Travis LB, Rowlings PA, et al. Risk of lymphoproliferative disorders after bone marrow transplantation: a multi-institutional study. Blood. 1999;94: Socie G, Curtis RE, Deeg HJ, et al. New malignant diseases after allogeneic marrow transplantation for childhood acute leukemia. J Clin Oncol. 2000;18: Muti G, De Gasperi A, Cantoni S, et al. Incidence and clinical characteristics of posttransplant lymphoproliferative disorders: report from a single center. Transpl Int. 2000;13(suppl 1):S382-S Paterson DL, Singh N, Gayowski T, et al. Pulmonary nodules in liver transplant recipients. Medicine (Baltimore). 1998;77: Shorter NA, Ross AJ III, August C, et al. The usefulness of open-lung biopsy in the pediatric bone marrow transplant population. J Pediatr Surg. 1988;23: Snyder CL, Ramsay NK, McGlave PB, et al. Diagnostic openlung biopsy after bone marrow transplantation. J Pediatr Surg. 1990;25: 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(3 pt 1): Shaikh ZH, Torres HA, Walsh GL, et al. Open lung biopsy in bone marrow transplant recipients has a poor diagnostic yield for a specific diagnosis. Transpl Infect Dis. 2002;4: Kim K, Lee MH, Kim J, et al. Importance of open lung biopsy in the diagnosis of invasive pulmonary aspergillosis in patients with hematologic malignancies. Am J Hematol. 2002;71: 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: Wong PW, Stefanec T, Brown K, et al. Role of fine-needle aspirates of focal lung lesions in patients with hematologic malignancies. Chest. 2002;121: Am J Clin Pathol 2004;121: Downloaded 210 from