Ann Thorac Surg 2002;74:185-190
© 2002 The Society of Thoracic Surgeons
Original article: general thoracic
Surgical considerations for pulmonary actinomycosis
Shunsuke Endo, MD*a,
Fumio Murayama, MDa,
Tsutomu Yamaguchi, MDa,
Shin-ichi Yamamoto, MDa,
Shin-ichi Otani, MDa,
Noriko Saito, MDa,
Yasunori Sohara, MDa
a Division of Thoracic Surgery, Department of Surgery, Jichi Medical School, Tochigi, Japan
Accepted for publication March 19, 2002.
* Address reprint requests to Dr Endo, Division of Thoracic Surgery, Department of Surgery, Jichi Medical School, Minamikawachi-machi, Kawachi-gun, Tochigi 329-0498, Japan
e-mail: tcvshun{at}jichi.ac.jp
 |
Abstract
|
|---|
Background. Diagnosis and treatment of pulmonary actinomycosis is difficult without surgical intervention.
Methods. Thirteen patients (10 men, 3 women; mean age, 62 years) underwent pulmonary resection and were given a pathologic diagnosis of pulmonary actinomycosis at our institution between 1976 and 2001. To clarify when pulmonary actinomycosis should be suspected in patients and the role of surgical intervention, we reviewed preoperative clinical characteristics, computed tomography findings, surgical indication, operative procedure, postoperative clinical course, and outcome.
Results. Ten patients (77%) had poor oral hygiene. Twelve patients (92%) were symptomatic, and 10 patients (77%) had hemoptysis. The mean interval between radiographic identification of the abnormality and surgical intervention was 8 months (interquartile range, 3.25 to 8 months). Computed tomography findings in all cases included radiologic opacity with air bronchogram or a low attenuation area. Lung cancer was diagnosed initially because of computed tomography findings of spiculation or pleural indentation, and operation was required in 8 patients (62%). The others were diagnosed with chronic pneumonia, and surgical intervention became necessary because of recurrent hemoptysis or prolonged illness. Six patients underwent lobectomy; the others underwent partial resection or segmentectomy. Neither complication nor recurrence has occurred.
Conclusions. When patients, particularly those with poor oral hygiene, show radiologic opacity with an air bronchogram or low attenuation area on the computed tomography scan, pulmonary actinomycosis should be considered and penicillin should be administered as diagnostic therapy. Surgical intervention may be necessary when frequent hemoptysis has no resolution or lung neoplasm cannot be ruled out.
|
Introduction
|
|---|
Actinomycosis is a chronic, suppurative granulomatous infection usually caused by Actinomyces israelii. Actinomyces israelii is of relatively low virulence and is found normally in the human oropharynx, particularly in persons with poor oral hygiene [1]. Actinomycosis classically involves cervicofacial (55% of cases), abdominopelvic (20% of cases), thoracic (15% of cases), and multiple organs (10% of cases). Thoracic involvement is uncommon, and if it occurs it may be misdiagnosed as bronchopulmonary disease such as chronic infection or lung cancer on the basis of the chronic clinical course, the poor response to antibiotic treatment, and the radiographic findings [2]. The histopathologic finding of yellowish sulfur granules is often necessary for differential diagnosis of actinomycosis because clinical and microbiologic studies cannot always detect the actual bacteria or the primary infection site [1]. A definitive diagnosis is not usually possible without surgical intervention [3–6]. Some patients with pulmonary actinomycosis, however, respond to long-term treatment with penicillin and resolve without surgical intervention [7–10].
We reviewed our experience with cases of pulmonary actinomycosis to clarify the best preoperative strategy and the role of surgical intervention.
|
Patients and methods
|
|---|
Patients
We identified 13 patients with pulmonary actinomycosis who had undergone pulmonary resection in our division between 1974 and 2001. Radiologic opacities had been found in all 2,045 patients. Patients were included in this study on the basis of pathologic findings. Characteristics of the study patients are shown in Table 1.
The study group consisted of 10 men and 3 women whose median age at the time of surgical intervention was 61 years (interquartile range, 50 to 67 years). Six of the 10 men were current smokers; 2 patients (1 man, 1 woman) were former smokers, and 4 patients (2 men, 2 women) had never smoked. Five of the men were addicted to alcohol. Oral hygiene was poor in 10 patients (77%), ie, gingivitis, tonsillitis, or sinusitis was present. Four patients had diabetes mellitus, and 5 had lung disease, ie, bronchiectasis (n = 2), old pleuritis (n = 1), emphysema (n = 1), and sarcoidosis (n = 1).
Follow-up data
Follow-up data were obtained from outpatient charts or by direct interview with each patient or a family member. The data collection was completed in December 2001.
Analysis
We reviewed symptoms, laboratory findings, computed tomography (CT) findings, the period from radiographic identification of the abnormality to surgical intervention, initial treatment, surgical indications, operative procedure, postoperative clinical course, and surgical outcomes in the patient group.
|
Results
|
|---|
Symptoms occurred in 12 (92%) of the 13 patients. Hemoptysis was experienced by 10 patients (77%). The abnormality was detected during a roentgenogram screening in the 1 patient who had no symptoms.
Sputum culture of 11 patients showed abnormal flora including gram-negative rods, anaerobic bacteria, and fungi, but no actinomyces were detected. Transbronchial biopsy yielded sulfur granules in only 1 patient. The erythrocyte sedimentation rate was elevated to 40 mm/h in 6 patients, whereas serum C-reactive protein was increased, albeit only slightly, in 12 patients.
Ten patients were diagnosed as having lung abscess including tuberculoma, and 9 of these patients received antimicrobial treatment. Radiologic findings resolved partially in 3 patients; 1 patient (patient no. 3) was diagnosed preoperatively by means of transbronchial biopsy as having actinomycosis, but this patient had an allergic reaction to the penicillin so it had to be discontinued. The time from the radiographic identification of the abnormality to surgical intervention ranged from 3 months to 40 years (median, 8 months; interquartile range, 3.25 to 8 months).
The CT findings are shown in Table 2.
The lower lobe was affected in 6 patients (46%); no preferential site for pulmonary actinomycosis was observed. Average diameter of the focal lesions was 4.1 cm. Tumor was diagnosed in 9 patients, and segmental atelectasis, infiltration shadow, or cystic formations in the other 4 (Fig 1). Air bronchogram was the most common finding (11 patients, 85%) followed by a low attenuation area (10 patients, 77%). Eight patients (62%) had pleural thickening adjacent to the focal lesion. Interlobular fissure occurred in 4 patients, and no pleural effusion was seen. Eight of 9 patients with CT features of tumor showed spiculation or pleural indentation at the edge of the focal lesion leading to a misdiagnosis of lung cancer (Fig 2).
Four patients who underwent arteriography showed hypervascularization of the ipsilateral bronchial and internal thoracic arteries. Embolization treatment failed in 1 of these patients.
View larger version (94K):
[in this window]
[in a new window]
|
Fig 1. Computed tomography scans of the patients who were diagnosed as having chronic pneumonia. (A) Lung windows of patient no. 2 showing multiple cystic formations in the left basal segment. (B) Lung windows of patient no. 3 showing atelectasis of the left posterior basal segment.
|
|
View larger version (89K):
[in this window]
[in a new window]
|
Fig 2. Computed tomography scans of the patients who were diagnosed as having lung cancer. (A) Lung windows of patient no. 11, showing a tumor with air bronchogram, spiculation, and pleural indentation in the right upper lobe. (B) Soft tissue windows of patient no. 13 showing the tumor with air bronchogram and a low attenuation area in the left anterior segment.
|
|
Indications for surgical intervention, surgical procedures, postoperative clinical course, and outcome are summarized on Table 3.
Four patients who were diagnosed as having chronic pneumonia required surgical intervention for recurrent or lethal hemoptysis, and for the patient (patient no. 3) who discontinued penicillin because of drug allergy, operation was indicated because of prolonged illness. Eight patients underwent operation because of the diagnosis of lung cancer, and 1 patient underwent urgent left upper lobectomy because of lethal hemoptysis.
The surgical interventions included single lobectomy (4 patients), combined lobectomy (2 patients), segmentectomy (3 patients), and partial resection (4 patients). Median operation time was 185 minutes (interquartile range, 129 to 227 minutes). Blood loss ranged from 90 to 2810 mL (median, 170 mL; interquartile range, 41 to 518 mL), and 3 patients required blood transfusion.
No operative complication or death occurred. Chronic inflammation continued after operation and spread to the superior segment in 1 patient (patient no. 9), who had undergone thoracoscopic partial resection of the left apicoposterior segment, and this was resolved with administration of penicillin. Postoperative antibiotic treatment was continued for 1 to 6 months (mean, 2.5 months) in 10 patients. There has been no recurrence of radiographic abnormality since operation (median postoperative period, 4 years; interquartile range, 0.75 to 8.8 years) in any case.
|
Comment
|
|---|
Most cases of pulmonary actinomycosis are caused by A. israelii [1]. The Latin root of Actinomyces means fungus, and the organisms in this genus cause clinical infections that simulate fungal disease. They are, however, classified as bacteria for several reasons, including the facts that these organisms have a cell wall and they respond to antibiotic rather than to antifungal drugs. The organism is a slow-growing, filamentous, gram-positive bacterium that shows true branching. It generally grows anaerobically, but it can be microaerophilic [11]. It is believed to enter the thorax by way of the bronchial tree, either by inhalation of contaminated aerosol particles or by aspiration of contaminated matter from the upper digestive tract [1, 6]. It is sometimes transmitted through an esophageal penetration and occasionally by direct extension from cervicofacial or abdominopelvic disease [11]. The bacteria are always present synergistically with other opportunistic bacteria [1].
Pulmonary actinomycosis extends to the pleura [12], pericardium [13], and chest wall [14]; occasionally it spreads hematogenously [15] because of delayed diagnosis. These clinical features have become uncommon, partly because of generally adequate oral hygiene practices and extensive use of antibiotics [1], whereas different clinical features of pulmonary actinomycosis are occasionally demonstrated with the spread of mass screening with or without CT [7]. Therefore, pulmonary actinomycosis experienced in our division was retrospectively reviewed in this study. The 0.6% (13 of 2,045) incidence of actinomycosis we encountered is in keeping with the 0.6% of a previous report [4].
Most of our patients had poor oral hygiene. Smoking and underlying lung disease appeared not to have a causal relation to development of this disease. No sputum culture showed Actinomyces. As reported previously [11], positive cultures are obtained in only about 50% of patients who actually have cervicofacial actinomycosis. The high rate of false negatives is caused by overgrowth of synergistic bacteria, improper anaerobic culturing, or prior antibiotic treatment that suppresses viable organisms, as indicated in our series. Furthermore, because actinomycosis is found in 30% to 50% of normal saliva specimens, the diagnosis of thoracic actinomycosis cannot be made from bronchial brushings or sputum studies. Transbronchial biopsy and needle aspiration biopsy can be used to confirm diagnosis [9], but the reported accuracy of the biopsy method is low owing to inadequate aspiration [2]. Thus, the most reliable method for accurate diagnosis of pulmonary actinomycosis is thoracotomy.
Symptoms of low-grade fever, cough, chest pain, and hemoptysis are common but not life-threatening; the actinomyces pathogens are not especially virulent [1]. Perilesional fibrosis in association with remarkable neovascularization may have contributed to the high incidence of hemoptysis in our study patients. Blood chemistry in most of our patients suggested chronic infection. Pleural fistulas and sepsis were uncommon in our patients, although these have been reported often in other series [12–15]. The presence of fistulas and sepsis may contribute to early detection and extensive use of antibiotics. The chronic clinical features described herein can lead to misdiagnosis of pulmonary actinomycosis as chronic pneumonia or malignant tumor; the lesion mimics lung cancer. In our present series, CT findings showed air bronchogram, a low attenuation area, or both. An air bronchogram in the involved lung parenchyma, ie, "the open bronchus sign," may help to differentiate this lesion from malignancy in which the bronchi, except in cases of bronchioloalveolar cell carcinoma, are usually obstructed or compressed by the tumor mass [2, 16]. In other reports, a central low attenuation area containing "sulfur granules" is considered the key to accurate diagnosis because of the protean manifestations of the organism [17]. However, 8 of our patients (62%) appeared to have a tumor with spiculation or pleural indentation, leading to misdiagnosis as lung cancer. Septal fibrosis between the pulmonary acini is causative.
The widespread use of antibiotics in known and even in undiagnosed cases, has substantially modified the clinical and radiologic pattern of actinomycosis. Penicillin remains the main drug of choice, despite other effective drugs [2]. In this study, however, drugs were discontinued for a short period. First, inappropriate antibiotics were selected. Second, malignancy could not be ruled out, as radiologic opacity did not resolve readily. The interval from radiologic appearance to surgical intervention was approximately 6 months in the patients who were misdiagnosed as having lung cancer. This is a sufficient amount of time to determine whether antibiotics will be effective in eradicating the disease. Once an accurate diagnosis is made, most patients are cured with long-term administration of penicillin [2, 7–10]. In our hospital, 2 patients who were diagnosed on the basis of transbronchial biopsy recovered completely after treatment with penicillin for 10 months without surgical intervention.
Five of our patients eventually underwent operation because of recurrent or lethal hemoptysis. These patients had a relatively long interval to surgical intervention, ranging from 5 months to 40 years. Diagnostic penicillin therapy may be appropriate. Surgical management has resulted in reduced mortality rates in comparison with rates in patients who are managed conservatively [18], but active pulmonary hemorrhage at the time of operation results in increased mortality rate [19]. When response to antibiotics is poor and hemoptysis is not resolvable, surgical intervention should be considered. Bronchial arterial embolization is reported to be effective in patients with hemoptysis [10], but embolization failed in a patient who had multiple feeding branches.
Surgical outcome is usually good in actinomycosis patients [3–6]. Surgical procedures should entail segmentectomy or lobectomy, not debridement. However, surgical intervention with pulmonary actinomycosis should be limited to diagnostic purposes [2]. One of our patients had continuing infection after partial resection, infection that was eradicated by 6 months’ treatment with penicillin. Surgical intervention was successful in our other patients, although some took penicillin as a prophylactic measure after operation. Postoperative administration of penicillin, especially in patients with limited resection for diagnosis, is recommended [3].
In conclusion, pulmonary actinomycosis should be considered when patients at risk, ie, with poor oral hygiene, have chronic pneumonia or a tumor mimicking lung cancer and show an increased level erythrocyte sedimentation rate and air bronchogram with a low attenuation area on CT. Penicillin should be administered in these patients for a few weeks before operation. Surgical intervention is indicated when a patient has frequent hemoptysis and the lesion does not respond to drug therapy.
|
References
|
|---|
- Tedder M., Wolfe W.G. Actinomycosis and nocardial infections. Chest Surg Clin North Am 1993;3:653-670.
- Hsieh M.-J., Liu H.-P., Chang J.-P., Chang C.-H. Thoracic actinomycosis. Chest 1993;104:366-370.[Abstract/Free Full Text]
- Tastepe A.I., Ulasan N.G., Liman S.T., Demircan S., Uzar A. Thoracic actinomycosis. Eur J Cardiothorac Surg 1998;14:578-583.[Abstract/Free Full Text]
- Rizzi A., Rocco G., Della Pona C., et al. Pulmonary actinomycosis: surgical considerations. Monaldi Arch Chest Dis 1996;51:369-372.[Medline]
- Shannon H.M., Wightman A.J., Carey F.A. Pulmonary actinomycosis: a master of disguise. J Infect 1995;31:165-169.[Medline]
- Takanami I., Takeuchi K., Naruke M. Association of esophageal achalasia and pulmonary actinomycosis infection simulating bronchial neoplasm. J Thorac Cardiovasc Surg 1999:199-200.
- Lee-Chiong T.L., Horowitz B.L., Perazella M.A., Brackett J.W. Thoracic actinomycosis: forgotten but not gone. J Respir Dis 1994;15:1062-1072.
- Dalhoff K., Wallner S., Finck C., Gatermann S., Wiebmann K.J. Endobronchial actinomycosis. Eur Respir J 1994;7:1189-1191.[Abstract]
- Bakhtawar I., Schaefer R.F., Salian N. Utility of Wang needle aspiration in the diagnosis of actinomycosis. Chest 2001;119:1966-1968.[Abstract/Free Full Text]
- Hamer D.H., Schwab L.E., Gray R. Massive hemoptysis from thoracic actinomycosis successfully treated by embolization. Chest 1992;101:1442-1443.[Abstract/Free Full Text]
- Bennhoff D.F. Actinomycosis: diagnostic and therapeutic considerations and a review of 32 cases. Laryngoscope 1984;94:1198-1217.[Medline]
- Merdler C., Greif J., Burke M., Sasson E., Campus A. Primary actinomycotic empyema. South Med J 1983;76:411-412.[Medline]
- Datta J.S., Raff M.J. Actinomycotic pleuropericarditis. Am Rev Respir Dis 1974;110:338-341.[Medline]
- Jara F.M., Toledo-Pereyra L.H., Magilligan D.J., Jr Surgical implications of pulmonary actinomycosis. J Thorac Cardiovasc Surg 1979;78:600-604.[Abstract]
- Kuijper E.J., Wiggerts H.O., Jonker G.J., Schaal K.P., deGans J. Disseminated actinomycosis due to Actinomyces meyeri and Actinobacillus actinomycetemcomitans. Scand J Infect Dis 1992;24:667-672.[Medline]
- Kwong J.S., Muller N.L., Godwin J.D., Aberle D., Grymaloski M.R. Thoracic actinomycosis: CT findings in eight patients. Radiology 1992;183:189-192.[Abstract/Free Full Text]
- Cheon J.-E., Im J.-G., Kim M.Y., Lee J.S., Choi G.M., Yeon K.M. Thoracic actinomycosis: CT findings. Radiology 1998;209:229-233.[Abstract/Free Full Text]
- Sehhat S., Oreizie M., Moinedine K. Massive pulmonary hemorrhage: surgical approach as choice of treatment. Ann Thorac Surg 1978;25:12-15.[Abstract]
- Conlan A.A., Hurwitz S.S., Krige L., Nicolaou N., Pool R. Massive hemoptysis: review of 123 cases. J Thorac Cardiovasc Surg 1983;85:120-124.[Abstract]
This article has been cited by other articles:
|
|
|
|
 
J. Choi, W.-J. Koh, T. S. Kim, K. S. Lee, J. Han, H. Kim, and O J. Kwon
Optimal Duration of IV and Oral Antibiotics in the Treatment of Thoracic Actinomycosis
Chest,
October 1, 2005;
128(4):
2211 - 2217.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
R. H. Wong, A. D. Sihoe, K. Thung, I. Y. Wan, M. B. Ip, and A. P. Yim
Actinomycosis: An Often Forgotten Diagnosis
Asian Cardiovasc Thorac Ann,
June 1, 2004;
12(2):
165 - 167.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
