- Email: [email protected]
Inflammatory pseudotumors of the lung
Inflammatory Pseudotumors of the Lung Robert J. Cerfolio, MD, Mark S. Allen, MD, Antonio G. Nascimento, MD, Claude Deschamps, MD, Victor F. Trastek, MD, Daniel L. Miller, MD, and Peter C. Pairolero, MD Section of General Thoracic Surgery and Division of Anatomic Pathology, Mayo Clinic and Mayo Foundation, Rochester, Minnesota
Background. Inflammatory pseudotumors of the lung are rare and often present a dilemma for the surgeon at time of operation. We reviewed our experience with patients who have this unusual pathology. Methods. Between February 1946 and September 1993, 56,400 general thoracic surgical procedures were performed at the Mayo Clinic. Twenty-three patients (0.04%) had resection of an inflammatory pseudotumor of the lung. There were 12 women and 11 men. Median age was 47 years (range, 5 to 77 years). Six patients (26%) were less than 18 years old. All pathologic specimens were rereviewed, and the diagnosis of inflammatory pseudotumor was confirmed. Eighteen patients (78%) were symptomatic which included cough in 12, weight loss in 4, fever in 4, and fatigue in 4. Four patients had prior incomplete resections performed elsewhere and underwent re-resection because of growth of residual pseudotumor. Wedge excision was performed in 7 patients, lobectomy in 6, pneumonectomy in 6, chest wall resection in 2, segmentectomy in 1, and bilobectomy in 1.
Complete resection was accomplished in 18 patients (78%). Median tumor size was 4.0 cm (range, 1 to 15 cm). There were no operative deaths. Follow-up was complete in all patients and ranged from 3 to 27 years (median, 9 years). Results. Overall 5-year survival was 91%. Nineteen patients are currently alive. Cause of death in the remaining 4 patients was unrelated to pseudotumor. The pseudotumor recurred in 3 of the 5 patients who had incomplete resection; 2 have had subsequent complete excision with no evidence of recurrence 8 and 9 years later. Conclusions. We conclude that inflammatory pseudotumors of the lung are rare. They often occur in children, can grow to a large size, and are often locally invasive, requiring significant pulmonary resection. Complete resection, when possible, is safe and leads to excellent survival. Pseudotumors, which recur, should be re-resected. (Ann Thorac Surg 1999;67:933– 6) © 1999 by The Society of Thoracic Surgeons
I
institution. These operations consisted of pulmonary, esophageal, and chest wall resections, as well as mediastionoscopies and video-assisted procedures. Twentythree patients (0.04%) had resection of an inflammatory pseudotumor of the lung. All pathologic slides were re-reviewed by a single pathologist and were confirmed to be inflammatory pseudotumor. The records were reviewed for age, sex, preoperative history, operative findings, and hospital course. Follow-up data were obtained from the patient’s clinic visits, correspondence from home health-care providers, and, when needed, telephone interview. Operative mortality was defined as any death within 30 days of the operation or during the hospitalization. Late mortality was defined as any subsequent death. Survival was estimated by the KaplanMeier method with the date of resection of the pseudotumor at our institution as the starting time [1]. The influence of variables on survival and growth of residual pseudotumor was analyzed using the proportional hazard model of Cox for continuous variables and the log-rank test for discrete variables [2, 3]. Values of p , 0.05 were considered statistically significant. All data are reported as median with ranges.
nflammatory pseudotumor of the lung is a benign tumor of the lung of unknown origin. Its nomenclature is inconsistent, and it has also been called plasma cell granuloma, fibrous histiocytoma, or fibroxanthoma because of its predominant cellular component. Little is known about its natural history or preferred treatment because it is rare and because its terminology is confusing. Pseudotumor presents a diagnostic dilemma because it is difficult to differentiate on histologic frozen section and especially difficult to distinguish from lymphoma or mediastinal fibrosis. Therefore, at thoracotomy it is often only biopsied or incompletely excised. To examine the clinical presentation, natural history, and effectiveness of varied treatments of inflammatory pseudotumors, we reviewed our experience with this unusual lesion.
Patients and Methods Patients and Data Collection Between February 1946 and September 1993, 56,400 general thoracic surgical procedures were performed at our
Accepted for publication Oct 21, 1998. Address reprint requests to Dr Cerfolio, Cardiothoracic Surgery, University of Alabama at Birmingham, 1900 University Blvd, THT 712, Birmingham, AL 35294.
© 1999 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
Clinical Findings The median age of the 23 patients (12 women, 11 men) was 47 years (range, 5 to 77 years). Six patients (26%) 0003-4975/99/$20.00 PII S0003-4975(99)00155-1
934
CERFOLIO ET AL INFLAMMATORY PSEUDOTUMORS OF LUNG
Ann Thorac Surg 1999;67:933– 6
Table 1. Patients Who Underwent Re-resection Date of Birth
Sex
Prior Operation Elsewhere
Nov 1969
Male
1979, biopsy only 1991, biopsy only 1984, pneumonectomy
Jan 1980
Female
June 1921
Male
May 1973
Male
Residual tumor on chest wall 1973, biopsy only 1975, right pneumonectomy Residual tumor on chest wall 1986, right upper lobectomy
Jan 1975
Female
Nov 1980
Female
Adjuvant Treatment
Mayo Surgery
Chemotherapy
1991, right pneumonectomy
Nothing
1987, chest wall resection plus margins 1988, chest wall resection 1975, chest wall resection
Radiation plus chemotherapy
Radiation
1987, completion right pneumonectomy
Residual tumor in lung None
None
None
None
1980, right upper lobectomy Residual tumor in lung 1982, right lower lobe wedge Residual tumor in lung 1984, chest wall resection plus right lower lobe wedge 1987, chest wall resection plus margins 1988, chest wall resection
were younger than 18 years. Eighteen patients (78%) were symptomatic, which included cough in 12, and weight loss, fever, and fatigue in 4 each. Significant medical conditions existed in 3 patients and included hypertension, chronic obstructive pulmonary disease, and an inflammatory brain tumor in 1 each. Two patients had a prior thoracotomy, and 2 patients had two prior thoracotomies as illustrated in Table 1. The indication for these 6 thoracotomies was an inflammatory pseudotumor. Two patients had radiation therapy and 1 had chemotherapy for treatment of residual pseudotumor after their initial thoracotomy. Before resection at our institution, chest roentgenogram and computed tomography of the chest, performed in all patients, demonstrated a pulmonary mass in 12 patients, nodules in 10, and infiltrate in 1.
Results Pneumonectomy was performed in 6 patients, bilobectomy in 1, lobectomy in 6, segmentectomy in 1, wedge resection in 7, and chest wall resection in 2. All patients had mediastinal and intrapulmonary lymph node resection, and no nodes were involved with pseudotumor in any patient. Complete resection of the mass was accomplished in 18 patients (78%). In 5 patients, who underwent incomplete resection, residual tumor was left on the chest wall in 2, in the lung in 2, and on the superior vena cava in 1. There were no operative deaths. Complications occurred in 4 patients (17%) and included retained secretions requiring bronchoscopy in 3, atrial fibrillation in 2, and severe confusion in 1. Median hospital stay was 8
days (range, 5 to 23 days). Follow-up was complete in all patients and ranged from 3.1 to 30.3 years (median, 13 years). Overall, 5-, 8-, and 10-year survival was 91.3%, 91.3%, and 77.7%, respectively. Nineteen patients are currently alive with a median follow-up of 16 years. Cause of death in the remaining 4 patients were not related to pseudotumor and were renal cell cancer, colon cancer, pulmonary fibrosis, and unknown in 1 each. The patient with an unknown cause of death had a normal chest roentgenogram 6 months before death, which occurred 10 years after his pseudotumor was completely resected. The patient who died of pulmonary fibrosis had progressive hypoxia and no evidence of tumor recurrence. No variables significantly affected survival. The residual tumor enlarged in 3 of the 5 patients who had incomplete resections. Two patients who had positive chest wall margins have undergone re-resection; they had complete excision (as seen in Table 1) and have no evidence of recurrence 8 and 9 years later. The third patient, who had a positive margin on the superior vena cava, has an enlarging mass on chest roentgenogram and has refused further treatment. The other 2 patients have been followed up for 4 and 9 years and have no evidence of growth of residual tumor. The median tumor size of the resected mass was 4.0 cm (range, 1 to 15 cm). There were plasma cells and spindle cells observed histologically in all patients. Mitosis (as defined by greater than 4 mitotic figures on average per high-power field) was seen in 6, nuclear atypia in 5, and necrosis in 3. Tumors were defined as having atypia or necrosis if any microscopic field showed this characteristic. The pathologist described the tumor as locally invasive in 12 patients. This was defined as microscopic
Ann Thorac Surg 1999;67:933– 6
CERFOLIO ET AL INFLAMMATORY PSEUDOTUMORS OF LUNG
935
Table 2. Pathologic Characteristics of All Patients Type of Inflammatory Pseudotumor Noninvasive Invasive p value
No. of Patients
Median Age (y)
Symptomatic (N)
Pneum or Lobec (N)
Mitosis (N)
Nuclear Atypia (N)
Growth Residual Tumor (N)
11 12
53.9 38.9 0.068
6 12 0.037
3 9 0.039
1 5 0.317
1 4 0.317
0 3 0.317
evidence of invasion of vessels by the inflammatory process. Table 2 depicts the pathologic characteristics of all 23 patients.
Comment Because of the numerous synonyms used for inflammatory pseudotumor of the lung (plasma cell granuloma, xanthoma, fibroxanthoma, histiocytoma, plasmacytoma, solitary mast cell tumor, and pseudoneoplastic pneumonia of the lung) little is known about its true incidence, natural history, and response to different treatments. To ensure that all patients in our series had inflammatory pseudotumors, all pathologic slides were rereviewed by a single pathologist (A.G.N.). The classic histiologic features [2–17] of spindle-shaped mesenchymal cells and inflammatory cells made up mostly of plasma cells were confirmed and other possible diagnoses ruled out. Inflammatory pseudotumor is a nonneoplastic process characterized by unregulated growth of inflammatory cells. The cause of this inflammatory response remains unknown. There are several theories [6, 8], most of which postulate an unchecked immunologic response to a viral or foreign antigen–antibody reaction. Inflammatory pseudotumor is not limited to the lung and can grow in other organ systems such as the brain or liver [18]. In our series the incidence of inflammatory pseudotumor of the lung was 0.04%. This is significantly lower than the 0.7% reported by Golbert and Pletven [19] and probably represents the small number of pediatric pulmonary resections performed at our institution as well as our strict pathologic criteria. Because inflammatory pseudotumor is so rare and small biopsies show inflammatory cells only, it is a diagnostic dilemma in any organ system for both the clinician and pathologist [20]. Although uncommon anywhere, it probably develops most commonly in the lung or trachea [21]. It also has an increased predilection to develop in children. The review by Bahadori and Liebow [2] states that it is the most common isolated primary tumor of the lung in children younger than 16 years. Similarly, Kauffman and Stout [22] described 3 patients with inflammatory pseudotumor of the lung in a review of 39 patients. In that series, 29% of the patients were younger than 20 years and 8% were younger than 10 years. Many case reports have described these unusual tumors in children. Inflammatory pseudotumors clearly develop more commonly in children than in adults. In our series, 6 patients (26%) were younger than 18 years. Seventy-eight percent of our patients were symptom-
atic. This incidence is higher than in other series and probably represents that fact that more than half our patients had a type of pseudotumor that we call locally invasive, which we believe is more often associated with symptoms than the noninvasive type (Table 2). Moreover, because this is a surgical series it may be more likely for a symptomatic patient to undergo resection for diagnosis and treatment than an asymptomatic one. Also we included cough (present in 12 of our 18 symptomatic patients) as a symptom whereas others had not. In our study chest roentgenograms showed a large mass in 12 patients. The diagnosis of an inflammatory pseudotumor when attempted by transbronchial biopsy or transthoracic needle biopsy often leads to confusion between it and fibrohistiocytic neoplasms, nodular sclerosing Hodgkin’s lymphoma, primary lung cancer, or mediastinal fibrosis. These preoperative procedures are unnecessary. We do not believe that they change patient management, and we do not recommend their use in a new, solitary, indeterminate mass. We recommend complete resection for both diagnosis and treatment. We recommend full pulmonary function testing, staging computed tomographic scans of the chest and upper abdomen, and liver function testing before operation. Concomitant inflammatory pseudotumors have been described, and we had 1 patient with a lung and brain inflammatory pseudotumor. At the time of thoracotomy, the mass should be wedge-resected, if possible, and sent for frozen section. Wedge resection is adequate treatment if removal is complete. Lobectomy should be performed if it is required for complete resection and if the patient’s pulmonary reserve is adequate. If pneumonectomy is required for complete resection, true-cut needle biopsy or incisional biopsy should be performed to establish pathologic confirmation before resection. Frozen sections performed at the time of operation are often indeterminate. The differential diagnosis includes lymphoma, sarcoma, and fibrosis. Usually the pathologist is able to eliminate a neoplastic process. If uncertainty remains on frozen section and if the mass can be safely removed, we recommend complete resection. As seen in our series and in the report by Bahadori and Liebow [2] of death owing to mediastinal compression, an inflammatory pseudotumor may continue to grow and lead to significant morbidity. Our data demonstrate that there may be two distinct types of pseudotumors as seen in Table 2. We have divided the tumors in our series into one of two categories on the basis of the presence or absence of local invasion as described by the pathologist. The pathologic
936
CERFOLIO ET AL INFLAMMATORY PSEUDOTUMORS OF LUNG
characteristic of invasion of vessels has been described previously [23, 24]. The first type we call noninvasive inflammatory pseudotumor of the lung. It is more likely to occur in an asymptomatic patient and is characterized by a small lesion that does not invade surrounding structures and is usually easily removed by wedge resection. The second type we call invasive inflammatory pseudotumor of the lung. It usually occurs in a younger patient with systemic symptoms of fever, fatigue, or weight loss. It is large and may invade local mediastinal structures or the chest wall. It usually requires a lobectomy or pneumonectomy for complete removal and may also require a concomitant chest wall resection. Grossly it is an invasive tumor that grows through tissue planes and pathologically it is characterized by nuclear atypia and a high number of mitotic figures. Although complete resection with pulmonary conservation remains our goal, many patients required lobectomy or pneumonectomy in our series. As seen in the report by Mandelbaum and colleagues [25], there is a low incidence of recurrence with long-term follow-up with minimal lung resection with complete removal of the mass. Complete resection remains the key to prevent recurrence. Because so many patients in our series had prior incomplete resections and because 12 patients had invasive-type tumors, our resections were larger than those reported in other series. Radiation has been reported to be a successful treatment for this tumor in some patients [26]. However, in our series, of the 4 patients who had prior attempts at resection performed elsewhere, 2 had radiation and 1 had chemotherapy (Table 1). All of these patients had continued growth of their residual tumor and underwent reresection. Brain irradiation was implemented in the 1 patient with a concomitant inflammatory pseudotumor of the brain. Interestingly, he has had no evidence of recurrence of either inflammatory pseudotumor 10 years after pneumonectomy or 11 years after brain irradiation. There have also been a few reports of spontaneous regression [25] and resolution with steroids. Patients with recurrent disease who can should undergo reresection. Our long-term follow-up documents that complete resection leads to long disease-free intervals. There was minimal morbidity or mortality in the 4 patients who had prior thoracotomies elsewhere or in the 3 patients who had repeat thoracotomies. We conclude that inflammatory pseudotumors of the lung are rare. They often occur in children, can grow to a large size, and are often locally invasive, requiring significant pulmonary resection. Complete resection, when possible, is safe and leads to excellent survival. Pseudotumors, which recur, should be reresected.
References 1. Kaplan EL, Meier. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457– 81.
Ann Thorac Surg 1999;67:933– 6
2. Bahadori M, Liebow A. Plasma cell granulomas of the lung. Cancer 1973;31:191–208. 3. Algere JA, Denst J. Xanthogranuloma as a coin lesion of the lung. Dis Chest 1958;33:427–31. 4. Buell R, Wang NS, Seemayer TA, Ahmed M. Endobronchial plasma cell granuloma (xanthomatous pseudotumor): a light and electron microscopic study. Hum Pathol 1976;7:411–26. 5. Berardi RS, Lee SS, Chen HP, Stines GJ. Collective review: inflammatory pseudotumors of the lung. Surg Gynecol Obstet 1983;156:89–96. 6. Herman PG, Hillman B, Pinkus G, Harris GC. Unusual noninfectious granulomas of the lung. Radiology 1976;121: 287–92. 7. Kuzela DC. Ultrastructural study of a postinflammatory “tumor” of the lung. Cancer 1975;36:149–56. 8. Shirakusa T, Miyanzaki N, Kitagawa T, Sugiyama K. Ultrastructural study of pulmonary plasma cell granuloma— report of a case. Br J Dis Chest 1979;73:289–96. 9. Titus JL, Harrison EG, Clagget OT, et al. Xanthomatous and inflammatory pseudotumors of the lung. Cancer 1962;15: 522–38. 10. Wentworth P, Lynch MJ, Fallis JC, et al. Xanthomatous pseudotumor of the lung: a case report with electron microscope and lipid studies. Cancer 1968;22:345–55. 11. Brunn H. Two interesting benign lung tumors of contradictory histopathology: remarks on the necessity of maintaining chest tumor registry. J Thorac Surg 1939;9:119–31. 12. Carter D, Eggleston JC. Tumors of the lower respiratory tract. Fascicle #17. Washington DC: Armed Forces Institute Pathology, 1974:300–7. 13. Brown WJ, Johnson LC. Postinflammatory “tumors” of the pleura. Three cases of pleural fibroma of the interlobal fissure. Mil Surg 1951;109:415–24. 14. Grossman RE, Bemis EL, Pemberton AH, Narodick BG. Fibrous histiocytoma or xanthoma of the lung with bronchial involvement. J Thorac Cardiovasc Surg 1973;65:653–7. 15. Long FL, Nott DB, MacArthur EB. Xanthomatous tumor of the lung with identification of lipid content. Aust Ann Med 1970;4:362–5. 16. Sherwin RP, Kern WH, Jones JC. Solitary mast cell granuloma (histocytoma) of the lung. Cancer 1965;18:634– 41. 17. Umiker WO, Iverson L. Postinflammatory “tumors” of the lung. J Thorac Surg 1954;28:55– 62. 18. Chan YF, White J, Brash H. Metachronous pulmonary and cerebral inflammatory pseudotumor in a child. Pediatr Pathol 1994;14:805–15. 19. Golbert ZV, Pletven SD. On pulmonary “pseudotumors.” Neoplasma 1967;14:189–98. 20. Copin MC, Gosselin BH, Ribet ME. Plasma cell granuloma of the lung: difficulties and prognosis. Ann Thorac Surg 1996; 61:1477– 82. 21. Tan-Liu NS, Matsubara O, Grillo HC, Mark EJ. Invasive fibrous tumor of the tracheobronchial tree: clinical and pathological study of seven cases. Hum Pathol 1989;20: 180– 4. 22. Kauffman SL, Stout AP. Histocytic tumors (fibrous xanthoma and histiocytoma) in children. Cancer 1961;14:469– 82. 23. Warter A, Satge D, Roeslin N. Angioinvasive plasma cell granulomas of the lung. Cancer 1987;59:435– 43. 24. Kubicz ST, Paradowska W, Czarnowska-Nastula B, Baworiczowa D. Pseudo-tumor of the lung in children. Ann Radiol (Paris) 1975;18:447–52. 25. Mandelbaum I, Brasher RE, Hull MT. Surgical treatment and course of pulmonary pseudotumor (plasma cell granuloma). J Thorac Cardiovasc Surg 1981;82:77– 82. 26. Hoover SV, Granston AS, Koch DF, Hudson TR. Plasma cell granuloma of the lung: response to radiation therapy: report of a single case. Cancer 1977;39:123–5.
Background. Inflammatory pseudotumors of the lung are rare and often present a dilemma for the surgeon at time of operation. We reviewed our experience with patients who have this unusual pathology. Methods. Between February 1946 and September 1993, 56,400 general thoracic surgical procedures were performed at the Mayo Clinic. Twenty-three patients (0.04%) had resection of an inflammatory pseudotumor of the lung. There were 12 women and 11 men. Median age was 47 years (range, 5 to 77 years). Six patients (26%) were less than 18 years old. All pathologic specimens were rereviewed, and the diagnosis of inflammatory pseudotumor was confirmed. Eighteen patients (78%) were symptomatic which included cough in 12, weight loss in 4, fever in 4, and fatigue in 4. Four patients had prior incomplete resections performed elsewhere and underwent re-resection because of growth of residual pseudotumor. Wedge excision was performed in 7 patients, lobectomy in 6, pneumonectomy in 6, chest wall resection in 2, segmentectomy in 1, and bilobectomy in 1.
Complete resection was accomplished in 18 patients (78%). Median tumor size was 4.0 cm (range, 1 to 15 cm). There were no operative deaths. Follow-up was complete in all patients and ranged from 3 to 27 years (median, 9 years). Results. Overall 5-year survival was 91%. Nineteen patients are currently alive. Cause of death in the remaining 4 patients was unrelated to pseudotumor. The pseudotumor recurred in 3 of the 5 patients who had incomplete resection; 2 have had subsequent complete excision with no evidence of recurrence 8 and 9 years later. Conclusions. We conclude that inflammatory pseudotumors of the lung are rare. They often occur in children, can grow to a large size, and are often locally invasive, requiring significant pulmonary resection. Complete resection, when possible, is safe and leads to excellent survival. Pseudotumors, which recur, should be re-resected. (Ann Thorac Surg 1999;67:933– 6) © 1999 by The Society of Thoracic Surgeons
I
institution. These operations consisted of pulmonary, esophageal, and chest wall resections, as well as mediastionoscopies and video-assisted procedures. Twentythree patients (0.04%) had resection of an inflammatory pseudotumor of the lung. All pathologic slides were re-reviewed by a single pathologist and were confirmed to be inflammatory pseudotumor. The records were reviewed for age, sex, preoperative history, operative findings, and hospital course. Follow-up data were obtained from the patient’s clinic visits, correspondence from home health-care providers, and, when needed, telephone interview. Operative mortality was defined as any death within 30 days of the operation or during the hospitalization. Late mortality was defined as any subsequent death. Survival was estimated by the KaplanMeier method with the date of resection of the pseudotumor at our institution as the starting time [1]. The influence of variables on survival and growth of residual pseudotumor was analyzed using the proportional hazard model of Cox for continuous variables and the log-rank test for discrete variables [2, 3]. Values of p , 0.05 were considered statistically significant. All data are reported as median with ranges.
nflammatory pseudotumor of the lung is a benign tumor of the lung of unknown origin. Its nomenclature is inconsistent, and it has also been called plasma cell granuloma, fibrous histiocytoma, or fibroxanthoma because of its predominant cellular component. Little is known about its natural history or preferred treatment because it is rare and because its terminology is confusing. Pseudotumor presents a diagnostic dilemma because it is difficult to differentiate on histologic frozen section and especially difficult to distinguish from lymphoma or mediastinal fibrosis. Therefore, at thoracotomy it is often only biopsied or incompletely excised. To examine the clinical presentation, natural history, and effectiveness of varied treatments of inflammatory pseudotumors, we reviewed our experience with this unusual lesion.
Patients and Methods Patients and Data Collection Between February 1946 and September 1993, 56,400 general thoracic surgical procedures were performed at our
Accepted for publication Oct 21, 1998. Address reprint requests to Dr Cerfolio, Cardiothoracic Surgery, University of Alabama at Birmingham, 1900 University Blvd, THT 712, Birmingham, AL 35294.
© 1999 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
Clinical Findings The median age of the 23 patients (12 women, 11 men) was 47 years (range, 5 to 77 years). Six patients (26%) 0003-4975/99/$20.00 PII S0003-4975(99)00155-1
934
CERFOLIO ET AL INFLAMMATORY PSEUDOTUMORS OF LUNG
Ann Thorac Surg 1999;67:933– 6
Table 1. Patients Who Underwent Re-resection Date of Birth
Sex
Prior Operation Elsewhere
Nov 1969
Male
1979, biopsy only 1991, biopsy only 1984, pneumonectomy
Jan 1980
Female
June 1921
Male
May 1973
Male
Residual tumor on chest wall 1973, biopsy only 1975, right pneumonectomy Residual tumor on chest wall 1986, right upper lobectomy
Jan 1975
Female
Nov 1980
Female
Adjuvant Treatment
Mayo Surgery
Chemotherapy
1991, right pneumonectomy
Nothing
1987, chest wall resection plus margins 1988, chest wall resection 1975, chest wall resection
Radiation plus chemotherapy
Radiation
1987, completion right pneumonectomy
Residual tumor in lung None
None
None
None
1980, right upper lobectomy Residual tumor in lung 1982, right lower lobe wedge Residual tumor in lung 1984, chest wall resection plus right lower lobe wedge 1987, chest wall resection plus margins 1988, chest wall resection
were younger than 18 years. Eighteen patients (78%) were symptomatic, which included cough in 12, and weight loss, fever, and fatigue in 4 each. Significant medical conditions existed in 3 patients and included hypertension, chronic obstructive pulmonary disease, and an inflammatory brain tumor in 1 each. Two patients had a prior thoracotomy, and 2 patients had two prior thoracotomies as illustrated in Table 1. The indication for these 6 thoracotomies was an inflammatory pseudotumor. Two patients had radiation therapy and 1 had chemotherapy for treatment of residual pseudotumor after their initial thoracotomy. Before resection at our institution, chest roentgenogram and computed tomography of the chest, performed in all patients, demonstrated a pulmonary mass in 12 patients, nodules in 10, and infiltrate in 1.
Results Pneumonectomy was performed in 6 patients, bilobectomy in 1, lobectomy in 6, segmentectomy in 1, wedge resection in 7, and chest wall resection in 2. All patients had mediastinal and intrapulmonary lymph node resection, and no nodes were involved with pseudotumor in any patient. Complete resection of the mass was accomplished in 18 patients (78%). In 5 patients, who underwent incomplete resection, residual tumor was left on the chest wall in 2, in the lung in 2, and on the superior vena cava in 1. There were no operative deaths. Complications occurred in 4 patients (17%) and included retained secretions requiring bronchoscopy in 3, atrial fibrillation in 2, and severe confusion in 1. Median hospital stay was 8
days (range, 5 to 23 days). Follow-up was complete in all patients and ranged from 3.1 to 30.3 years (median, 13 years). Overall, 5-, 8-, and 10-year survival was 91.3%, 91.3%, and 77.7%, respectively. Nineteen patients are currently alive with a median follow-up of 16 years. Cause of death in the remaining 4 patients were not related to pseudotumor and were renal cell cancer, colon cancer, pulmonary fibrosis, and unknown in 1 each. The patient with an unknown cause of death had a normal chest roentgenogram 6 months before death, which occurred 10 years after his pseudotumor was completely resected. The patient who died of pulmonary fibrosis had progressive hypoxia and no evidence of tumor recurrence. No variables significantly affected survival. The residual tumor enlarged in 3 of the 5 patients who had incomplete resections. Two patients who had positive chest wall margins have undergone re-resection; they had complete excision (as seen in Table 1) and have no evidence of recurrence 8 and 9 years later. The third patient, who had a positive margin on the superior vena cava, has an enlarging mass on chest roentgenogram and has refused further treatment. The other 2 patients have been followed up for 4 and 9 years and have no evidence of growth of residual tumor. The median tumor size of the resected mass was 4.0 cm (range, 1 to 15 cm). There were plasma cells and spindle cells observed histologically in all patients. Mitosis (as defined by greater than 4 mitotic figures on average per high-power field) was seen in 6, nuclear atypia in 5, and necrosis in 3. Tumors were defined as having atypia or necrosis if any microscopic field showed this characteristic. The pathologist described the tumor as locally invasive in 12 patients. This was defined as microscopic
Ann Thorac Surg 1999;67:933– 6
CERFOLIO ET AL INFLAMMATORY PSEUDOTUMORS OF LUNG
935
Table 2. Pathologic Characteristics of All Patients Type of Inflammatory Pseudotumor Noninvasive Invasive p value
No. of Patients
Median Age (y)
Symptomatic (N)
Pneum or Lobec (N)
Mitosis (N)
Nuclear Atypia (N)
Growth Residual Tumor (N)
11 12
53.9 38.9 0.068
6 12 0.037
3 9 0.039
1 5 0.317
1 4 0.317
0 3 0.317
evidence of invasion of vessels by the inflammatory process. Table 2 depicts the pathologic characteristics of all 23 patients.
Comment Because of the numerous synonyms used for inflammatory pseudotumor of the lung (plasma cell granuloma, xanthoma, fibroxanthoma, histiocytoma, plasmacytoma, solitary mast cell tumor, and pseudoneoplastic pneumonia of the lung) little is known about its true incidence, natural history, and response to different treatments. To ensure that all patients in our series had inflammatory pseudotumors, all pathologic slides were rereviewed by a single pathologist (A.G.N.). The classic histiologic features [2–17] of spindle-shaped mesenchymal cells and inflammatory cells made up mostly of plasma cells were confirmed and other possible diagnoses ruled out. Inflammatory pseudotumor is a nonneoplastic process characterized by unregulated growth of inflammatory cells. The cause of this inflammatory response remains unknown. There are several theories [6, 8], most of which postulate an unchecked immunologic response to a viral or foreign antigen–antibody reaction. Inflammatory pseudotumor is not limited to the lung and can grow in other organ systems such as the brain or liver [18]. In our series the incidence of inflammatory pseudotumor of the lung was 0.04%. This is significantly lower than the 0.7% reported by Golbert and Pletven [19] and probably represents the small number of pediatric pulmonary resections performed at our institution as well as our strict pathologic criteria. Because inflammatory pseudotumor is so rare and small biopsies show inflammatory cells only, it is a diagnostic dilemma in any organ system for both the clinician and pathologist [20]. Although uncommon anywhere, it probably develops most commonly in the lung or trachea [21]. It also has an increased predilection to develop in children. The review by Bahadori and Liebow [2] states that it is the most common isolated primary tumor of the lung in children younger than 16 years. Similarly, Kauffman and Stout [22] described 3 patients with inflammatory pseudotumor of the lung in a review of 39 patients. In that series, 29% of the patients were younger than 20 years and 8% were younger than 10 years. Many case reports have described these unusual tumors in children. Inflammatory pseudotumors clearly develop more commonly in children than in adults. In our series, 6 patients (26%) were younger than 18 years. Seventy-eight percent of our patients were symptom-
atic. This incidence is higher than in other series and probably represents that fact that more than half our patients had a type of pseudotumor that we call locally invasive, which we believe is more often associated with symptoms than the noninvasive type (Table 2). Moreover, because this is a surgical series it may be more likely for a symptomatic patient to undergo resection for diagnosis and treatment than an asymptomatic one. Also we included cough (present in 12 of our 18 symptomatic patients) as a symptom whereas others had not. In our study chest roentgenograms showed a large mass in 12 patients. The diagnosis of an inflammatory pseudotumor when attempted by transbronchial biopsy or transthoracic needle biopsy often leads to confusion between it and fibrohistiocytic neoplasms, nodular sclerosing Hodgkin’s lymphoma, primary lung cancer, or mediastinal fibrosis. These preoperative procedures are unnecessary. We do not believe that they change patient management, and we do not recommend their use in a new, solitary, indeterminate mass. We recommend complete resection for both diagnosis and treatment. We recommend full pulmonary function testing, staging computed tomographic scans of the chest and upper abdomen, and liver function testing before operation. Concomitant inflammatory pseudotumors have been described, and we had 1 patient with a lung and brain inflammatory pseudotumor. At the time of thoracotomy, the mass should be wedge-resected, if possible, and sent for frozen section. Wedge resection is adequate treatment if removal is complete. Lobectomy should be performed if it is required for complete resection and if the patient’s pulmonary reserve is adequate. If pneumonectomy is required for complete resection, true-cut needle biopsy or incisional biopsy should be performed to establish pathologic confirmation before resection. Frozen sections performed at the time of operation are often indeterminate. The differential diagnosis includes lymphoma, sarcoma, and fibrosis. Usually the pathologist is able to eliminate a neoplastic process. If uncertainty remains on frozen section and if the mass can be safely removed, we recommend complete resection. As seen in our series and in the report by Bahadori and Liebow [2] of death owing to mediastinal compression, an inflammatory pseudotumor may continue to grow and lead to significant morbidity. Our data demonstrate that there may be two distinct types of pseudotumors as seen in Table 2. We have divided the tumors in our series into one of two categories on the basis of the presence or absence of local invasion as described by the pathologist. The pathologic
936
CERFOLIO ET AL INFLAMMATORY PSEUDOTUMORS OF LUNG
characteristic of invasion of vessels has been described previously [23, 24]. The first type we call noninvasive inflammatory pseudotumor of the lung. It is more likely to occur in an asymptomatic patient and is characterized by a small lesion that does not invade surrounding structures and is usually easily removed by wedge resection. The second type we call invasive inflammatory pseudotumor of the lung. It usually occurs in a younger patient with systemic symptoms of fever, fatigue, or weight loss. It is large and may invade local mediastinal structures or the chest wall. It usually requires a lobectomy or pneumonectomy for complete removal and may also require a concomitant chest wall resection. Grossly it is an invasive tumor that grows through tissue planes and pathologically it is characterized by nuclear atypia and a high number of mitotic figures. Although complete resection with pulmonary conservation remains our goal, many patients required lobectomy or pneumonectomy in our series. As seen in the report by Mandelbaum and colleagues [25], there is a low incidence of recurrence with long-term follow-up with minimal lung resection with complete removal of the mass. Complete resection remains the key to prevent recurrence. Because so many patients in our series had prior incomplete resections and because 12 patients had invasive-type tumors, our resections were larger than those reported in other series. Radiation has been reported to be a successful treatment for this tumor in some patients [26]. However, in our series, of the 4 patients who had prior attempts at resection performed elsewhere, 2 had radiation and 1 had chemotherapy (Table 1). All of these patients had continued growth of their residual tumor and underwent reresection. Brain irradiation was implemented in the 1 patient with a concomitant inflammatory pseudotumor of the brain. Interestingly, he has had no evidence of recurrence of either inflammatory pseudotumor 10 years after pneumonectomy or 11 years after brain irradiation. There have also been a few reports of spontaneous regression [25] and resolution with steroids. Patients with recurrent disease who can should undergo reresection. Our long-term follow-up documents that complete resection leads to long disease-free intervals. There was minimal morbidity or mortality in the 4 patients who had prior thoracotomies elsewhere or in the 3 patients who had repeat thoracotomies. We conclude that inflammatory pseudotumors of the lung are rare. They often occur in children, can grow to a large size, and are often locally invasive, requiring significant pulmonary resection. Complete resection, when possible, is safe and leads to excellent survival. Pseudotumors, which recur, should be reresected.
References 1. Kaplan EL, Meier. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457– 81.
Ann Thorac Surg 1999;67:933– 6
2. Bahadori M, Liebow A. Plasma cell granulomas of the lung. Cancer 1973;31:191–208. 3. Algere JA, Denst J. Xanthogranuloma as a coin lesion of the lung. Dis Chest 1958;33:427–31. 4. Buell R, Wang NS, Seemayer TA, Ahmed M. Endobronchial plasma cell granuloma (xanthomatous pseudotumor): a light and electron microscopic study. Hum Pathol 1976;7:411–26. 5. Berardi RS, Lee SS, Chen HP, Stines GJ. Collective review: inflammatory pseudotumors of the lung. Surg Gynecol Obstet 1983;156:89–96. 6. Herman PG, Hillman B, Pinkus G, Harris GC. Unusual noninfectious granulomas of the lung. Radiology 1976;121: 287–92. 7. Kuzela DC. Ultrastructural study of a postinflammatory “tumor” of the lung. Cancer 1975;36:149–56. 8. Shirakusa T, Miyanzaki N, Kitagawa T, Sugiyama K. Ultrastructural study of pulmonary plasma cell granuloma— report of a case. Br J Dis Chest 1979;73:289–96. 9. Titus JL, Harrison EG, Clagget OT, et al. Xanthomatous and inflammatory pseudotumors of the lung. Cancer 1962;15: 522–38. 10. Wentworth P, Lynch MJ, Fallis JC, et al. Xanthomatous pseudotumor of the lung: a case report with electron microscope and lipid studies. Cancer 1968;22:345–55. 11. Brunn H. Two interesting benign lung tumors of contradictory histopathology: remarks on the necessity of maintaining chest tumor registry. J Thorac Surg 1939;9:119–31. 12. Carter D, Eggleston JC. Tumors of the lower respiratory tract. Fascicle #17. Washington DC: Armed Forces Institute Pathology, 1974:300–7. 13. Brown WJ, Johnson LC. Postinflammatory “tumors” of the pleura. Three cases of pleural fibroma of the interlobal fissure. Mil Surg 1951;109:415–24. 14. Grossman RE, Bemis EL, Pemberton AH, Narodick BG. Fibrous histiocytoma or xanthoma of the lung with bronchial involvement. J Thorac Cardiovasc Surg 1973;65:653–7. 15. Long FL, Nott DB, MacArthur EB. Xanthomatous tumor of the lung with identification of lipid content. Aust Ann Med 1970;4:362–5. 16. Sherwin RP, Kern WH, Jones JC. Solitary mast cell granuloma (histocytoma) of the lung. Cancer 1965;18:634– 41. 17. Umiker WO, Iverson L. Postinflammatory “tumors” of the lung. J Thorac Surg 1954;28:55– 62. 18. Chan YF, White J, Brash H. Metachronous pulmonary and cerebral inflammatory pseudotumor in a child. Pediatr Pathol 1994;14:805–15. 19. Golbert ZV, Pletven SD. On pulmonary “pseudotumors.” Neoplasma 1967;14:189–98. 20. Copin MC, Gosselin BH, Ribet ME. Plasma cell granuloma of the lung: difficulties and prognosis. Ann Thorac Surg 1996; 61:1477– 82. 21. Tan-Liu NS, Matsubara O, Grillo HC, Mark EJ. Invasive fibrous tumor of the tracheobronchial tree: clinical and pathological study of seven cases. Hum Pathol 1989;20: 180– 4. 22. Kauffman SL, Stout AP. Histocytic tumors (fibrous xanthoma and histiocytoma) in children. Cancer 1961;14:469– 82. 23. Warter A, Satge D, Roeslin N. Angioinvasive plasma cell granulomas of the lung. Cancer 1987;59:435– 43. 24. Kubicz ST, Paradowska W, Czarnowska-Nastula B, Baworiczowa D. Pseudo-tumor of the lung in children. Ann Radiol (Paris) 1975;18:447–52. 25. Mandelbaum I, Brasher RE, Hull MT. Surgical treatment and course of pulmonary pseudotumor (plasma cell granuloma). J Thorac Cardiovasc Surg 1981;82:77– 82. 26. Hoover SV, Granston AS, Koch DF, Hudson TR. Plasma cell granuloma of the lung: response to radiation therapy: report of a single case. Cancer 1977;39:123–5.