respiratory investigation 52 (2014) 160 –166
Contents lists available at ScienceDirect
Respiratory Investigation journal homepage: www.elsevier.com/locate/resinv
Original article
Pulmonary nocardiosis: A clinical analysis of 59 cases Yu Kuraharaa, Kazunobu Tachibanaa,b, Kazunari Tsuyuguchib, Masanori Akirac, Katsuhiro Suzukia,n, Seiji Hayashia a
Department of Internal Medicine, National Hospital Organization Kinki-Chuo Chest Medical Center, Osaka, Japan Clinical Research Center, National Hospital Organization Kinki-Chuo Chest Medical Center, Osaka, Japan c Department of Radiology, National Hospital Organization Kinki-Chuo Chest Medical Center, Osaka, Japan b
ar t ic l e in f o
abs tra ct
Article history:
Background: Pulmonary nocardiosis is a rare but severe infection caused by Nocardia
Received 15 August 2013
species. This study aimed at describing the clinical characteristics and prognosis of
Received in revised form
pulmonary nocardiosis.
5 September 2013
Methods: An observational, retrospective study was undertaken of patients diagnosed with
Accepted 26 September 2013
pulmonary nocardiosis over a 13-year period at the Kinki-Chuo Chest Medical Center,
Available online 7 November 2013
Osaka, Japan.
Keywords: Clinical characteristics Nocardia Pulmonary nocardiosis Radiographic findings Survival
Results: Seven patients with airway nocardial colonization and 59 patients with pulmonary nocardiosis were identified, one of whom had disseminated nocardiosis. Patients with pulmonary nocardiosis were predominantly male patients (73%), with a mean age of 66 (range, 15–88) years. New-onset cough and dyspnea were the most common manifestations (76%). Although 52 (88%) patients had at least one underlying pulmonary disease, most patients did not appear to be systemically immunocompromised. The predominant abnormality on chest computed tomography in pulmonary nocardiosis was airspace consolidation (52%), sometimes associated with cavitation. Multivariate Cox proportional-hazards analysis revealed the following significant and independent risk factors for overall mortality: age 468 years (hazard ratio [HR], 4.7; 95% confidence interval [CI], 1.6–14; p¼0.05), pulmonary aspergillosis (HR, 8.8; 95% CI, 2.4–33; p¼ 0.01), and trimethoprim/sulfamethoxazole (TMP-SMZ) resistance (HR, 4.3; 95% CI, 1.6– 11; p¼ 0.04). Conclusions: Clinicians should be aware that pulmonary nocardiosis can occur even in immunocompetent patients, especially those with an underlying pulmonary disease. In pulmonary nocardiosis, older age, pulmonary aspergillosis, and TMP-SMZ resistance are associated with increased risk of mortality. & 2013 The Japanese Respiratory Society. Published by Elsevier B.V. All rights reserved.
Abbreviations: CI, confidence interval; CT, computed tomography; HR, hazard ratio; TMP-SMZ, trimethoprim/sulfamethoxazole n Corresponding author. Tel.: þ81 72 252 3021; fax: þ81 72 251 1372. E-mail addresses: [email protected] (Y. Kurahara), [email protected] (K. Tachibana), [email protected] (K. Tsuyuguchi), [email protected] (M. Akira), [email protected], [email protected] (K. Suzuki), [email protected] (S. Hayashi). 2212-5345/$ - see front matter & 2013 The Japanese Respiratory Society. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.resinv.2013.09.004
respiratory investigation 52 (2014) 160 –166
1.
Introduction
The Nocardia genus of aerobic gram-positive bacteria causes a range of infectious diseases, including isolated pulmonary and skin infections and disseminated disease [1]. Pulmonary nocardiosis is a rare but severe infection that commonly presents as a subacute or chronic disease, mimicking lung cancer or other pulmonary infections such as mycosis or bacterial pneumonia. There are more than 90 Nocardia species, at least 33 of which cause disease in humans [2]. Because of changes in taxonomy and the difficulty inherent in the routine identification of Nocardia strains at the species level, there have been few published studies concerning pulmonary nocardiosis [3–8]. Of those that have been conducted, most have investigated only patients with immunosuppression and have included only a small number of patients. Moreover, the details of the clinical characteristics and prognosis of pulmonary nocardiosis remain unclear. The objective of the present study is to clarify the characteristics of pulmonary nocardial patients and to elucidate the factors that determine its etiology and prognosis.
tests, growth characteristics, and antimicrobial susceptibility patterns. Hence, the microbiological notation was rendered in its traditional form in this study. The susceptibility of the isolates was determined by the disc diffusion method. A total of 90 specimens from 66 patients were identified as cultureproven Nocardia.
2.3.
Materials and methods
2.1.
Study Subjects
We aimed at studying the clinical pattern of pulmonary nocardiosis in the Kinki-Chuo Chest Medical Center, Osaka, Japan, based on a retrospective review of cases from January 1999 to October 2012. The hospital is a 450-bed respiratory disease center designed to manage patients with various respiratory diseases. In the present study, patients with no evidence of pulmonary disease or symptoms of respiratory infection despite the isolation of Nocardia species were defined as having airway nocardial colonization. Pulmonary nocardiosis was defined as the presence of symptoms of respiratory infection consistent with pulmonary nocardiosis, with Nocardia species isolated from respiratory sample cultures on at least one occasion. Disseminated nocardiosis was defined as an infection in two noncontiguous organs or in the central nervous system. This study was approved by the Institutional Review Board at the Kinki-Chuo Chest Medical Center (Approval date: November 14, 2011; Approval number: 388).
2.2.
Radiographic assessment
Three pulmonary infectious disease experts (Y.K., K. Tachibana., and K.S.) and one chest radiologist (M.A.) retrospectively reviewed de-identified computed tomography (CT) films. The reviewers classified each case based on the presence of the following features: (1) airspace consolidation; (2) nodules (defined as those with diameters o3 cm); (3) cavity; (4) consolidation with cavity; (5) pleural effusion; and (6) indeterminate (we anticipated the possibility that some cases had an underlying pulmonary condition indistinguishable from pulmonary nocardiosis because of the complicated structure of lung fields). Any inconsistencies were resolved by consensus.
2.4.
2.
161
Study variables
The following data were collected: patient demographics, comorbidities, immunological state, clinical manifestations, radiographic findings, bacteriological reports, susceptibility to trimethoprim/sulfamethoxazole (TMP-SMZ), treatment course, and survival time from diagnosis.
2.5.
Statistical analysis
Statistical analysis was performed with SPSS version 15.0J software (SPSS Inc., Chicago, Illinois), and the Kaplan–Meier method was used for survival analysis. If the patient died or was lost to follow-up during the study period, this was recorded as censored data. A univariate assessment of selected risk factors was performed using a Cox proportional hazard model. To eliminate confounding factors in predicting mortality risk, variables with P values o0.05 on univariate analysis were then entered into a multivariate assessment. We used a Cox proportional hazard model with a step-up procedure, utilizing the likelihood ratio as the criterion for adding significant variables. Results are expressed as hazard ratios with corresponding 95% confidence intervals (CI). A P value of o0.05 was regarded as significant.
Microbiological identification
In order to diagnose pulmonary nocardiosis, samples were examined microscopically after Gram staining in order to determine the presence of microorganisms and establish specimen quality. The presumptive identification of Nocardia species was based on the microscopic characteristics of the isolates and the macroscopic morphology of the colonies. Although 16S ribosomal RNA sequence-based identification of Nocardia species has recently become the gold standard, this technology was not available in our laboratories during the study period. Therefore, Nocardia isolates were identified using a combination of traditional standard biochemical
3.
Results
3.1.
Characteristics and clinical features of the patients
A total of 90 respiratory specimens from 66 patients were identified as culture-proven Nocardia. Seven patients were diagnosed with airway nocardial colonization because they had minimal or no lung-field lesions and the absence of respiratory symptoms. These lesions remained stable after at least 1 year of follow-up, during which time no treatment was given. In total, 59 patients were diagnosed with pulmonary nocardiosis (43 male patients and 16 female patients). One
162
respiratory investigation 52 (2014) 160 –166
patient had disseminated nocardiosis of the central nervous system. The characteristics and underlying diseases of patients with airway nocardial colonization and pulmonary nocardiosis are detailed in Table 1. The mean age was 66 years (range, 15–88 years). A total of 55 (93%) of 59 patients had at least one underlying pulmonary disease. Pre-existing structural abnormalities of the lung were common. Notably, most patients did not appear to be immunocompromised. Only one patient was being treated with corticosteroids (for systemic vasculitis); no other patient was receiving immunosuppressive agents. No patient had undergone transplantation, and no patient had human immunodeficiency virus infection or were alcoholic. All patients with airway nocardial colonization were male patients, and their mean age was 71 years.
3.2. Symptoms, laboratory data, and radiographic patterns At the time when Nocardia infection was identified, 45 patients with pulmonary nocardiosis (76%) had new-onset cough and dyspnea, which was associated with a fever higher than 37.5 1C in 36 patients (61%). In one case, dissemination to the central nervous system was detected, and that patient died 55 days after diagnosis. Blood examination confirmed
Table 1 – Characteristics of airway nocardial colonization and pulmonary nocardiosis.
Male/female Age, years, mean7SD Never smoker/current or former smoker Underlying condition Non-pulmonary underlying condition Cardiovascular diseases Diabetes mellitus Connective tissue diseases Hepatic diseases Renal failure Pulmonary underlying condition Chronic obstructive pulmonary disease Pulmonary malignancy Pneumoconiosis Healed or active pulmonary tuberculosis Non-tuberculous mycobacterial infection Pulmonary aspergillosis Bronchiectasis Othersa
Airway nocardial colonization (n ¼7)
Pulmonary nocardiosis (n ¼59)
7/0 71717 1/9
43/16 66713 39/20
4 (57) 1 (14)
55 (93) 12 (20)
1 (14) 0 0
5 (8.5) 4 (6.8) 2 (3.4)
0 0 3 (43)
1 (1.7) 1 (1.7) 52 (88)
1 (14)
13 (22)
1 (14) 0 0
12 (20) 11 (19) 11 (19)
0
9 (15)
0 1 (14) 0
6 (10) 6 (10) 3 (5.1)
Data are presented as mean7SD or n (%). a Includes bronchial asthma, idiopathic pulmonary fibrosis, and chronic hypersensitivity pneumonitis.
leukocytosis in 20 patients (34%) and elevated C-reactive protein levels in 43 patients (73%). Radiographic findings in the present study are given in Table 2. The predominant abnormality on chest CT was airspace consolidation (Fig. 1), which may have been associated with cavitation. The second predominant abnormality was the presence of nodules (Fig. 2). Because a majority (88%) of the patients had at least one baseline pulmonary disease, radiographic patterns were indeterminate in 12 patients (20%). For example, mixed active infection with other organisms (such as Aspergillus species or Mycobacterium species) resulted in a radiographic appearance indistinguishable from that of pulmonary nocardiosis. All patients with airway nocardial colonization had tiny nodules in the lung fields, and these nodules persisted for at least 1 year of follow-up in the absence of treatment.
3.3.
Microbiological identification
In the present study, microbiological notation was rendered in its traditional form. Because Nocardia taxonomy was under continuous revision during the study period, we were not able to re-identify all of the Nocardia species using 16S ribosomal RNA sequence analysis. The species isolated from the respiratory samples in the present study were the following: Nocardia asteroides (40 patients [68%]), Nocardia farcinica (six patients [10%]), Nocardia nova (six patients [10%]), Nocardia otitidiscaviarum (four patients [6.8%]), and Nocardia brasiliensis (three patients [5.1%]). Infection was polymicrobial in 17 cases (30%), with the following organisms: Aspergillus fumigatus (four patients [6.8%]), Aspergillus niger (two patients [3.4%]), Mycobacterium tuberculosis (two patients [3.4%]), Mycobacterium avium complex (six patients [10%]), and Mycobacterium kansasii (three patients [5.1%]). Thirty-six (61%) patients were treated with TMP-SMZ (Table 3). Sixteen patients (27%) had contracted TMP-SMZ-resistant Nocardia infection: specifically, those with N. farcinica (5 of 6), N. otitidiscaviarum (4 of 4), and N. asteroides infection (7 of 40). Most patients infected with TMP-SMZ-resistant pathogens were treated with imipenem/cilastatin (Table 1).
Table 2 – Radiographic (computed tomography) presentations among patients with airway nocardial colonization and pulmonary nocardiosis.
Air-space consolidation Nodules Consolidation with cavity Cavity Pleural effusion Indeterminatea Total a
Airway nocardial colonization (n ¼7)
Pulmonary nocardiosis (n ¼ 59)
0
28 (47)
7 (100) 0
13 (22) 3 (5.1)
0 0 0 7 (100)
2 1 12 59
(3.4) (1.7) (20) (100)
Underlying pulmonary condition indistinguishable from pulmonary nocardiosis.
respiratory investigation 52 (2014) 160 –166
163
Fig. 1 – A 66-year-old man with pulmonary nocardiosis and idiopathic pulmonary fibrosis. Left: initial computed tomography (CT) scan shows bilateral subpleural reticulation and cystic lesions. Right: CT scan, obtained when the patient developed highgrade fever and cough, shows consolidation with air bronchogram in the left upper lobe.
Fig. 2 – A 77-year-old man with pulmonary nocardiosis and silicosis. Left: initial computed tomography (CT) scan shows bilateral small nodules consistent with silicosis. Right: CT scan, obtained when the patient developed cough and yellow sputum, shows nodules in the right lower lobe. Table 3 – Trimethoprim–sulfamethoxazole susceptibility and treatment among patients with airway nocardial colonization and pulmonary nocardiosis. Airway nocardial colonization (n ¼7) TMP/SMZ susceptibility Susceptible 7 (100) Resistant 0 Treatment TMP/SMZ Imipenem/ cilastatin Meropenem Minocycline Surgery Othersa
– – – – – –
Data are presented as n (%). a Includes ceftriaxone, linezolid, and SMZ ¼trimethoprim/sulfamethoxazole.
3.4.
Pulmonary nocardiosis (n¼ 59)
43 (73) 16 (27)
36 (61) 13 (22) 2 2 2 3
Fig. 3 – Kaplan–Meier estimates of overall survival in patients with pulmonary nocardiosis. Survival was calculated from the day of first diagnosis of the infection. Data were censored at the end of the follow-up period or at the time of death.
(3.4) (3.4) (3.4) (5.1)
amikacin.
TMP/
Risk factors
Survival rates from the date of diagnosis of pulmonary nocardiosis were 97% at 30 days, 83% at 90 days, and 74% at 180 days (Fig. 3). To determine the risk factors associated with mortality, we analyzed the data using Cox proportional hazard regression modeling. After univariate analysis, five significant risk factors
(po0.05) for mortality were identified: a history of heavy smoking (Brinkman index 4428), age 468 years, pulmonary malignancy, pulmonary aspergillosis, and TMP-SMZ resistance (Table 4). These significant risk factors identified by univariate analysis were then entered into multivariate analysis using a Cox proportional hazard regression model. The risk factors independently associated with overall mortality were age 468 years (hazard ratio [HR], 4.7; 95% confidence interval [CI], 1.6–14; p¼0.05), pulmonary aspergillosis (HR, 8.8; 95% CI, 2.4–33; p¼ 0.01), and TMP-SMZ resistance (HR, 4.3; 95% CI, 1.6–11; p¼ 0.04) (Table 5).
164
respiratory investigation 52 (2014) 160 –166
Table 4 – Univariate analysis of risk factors for overall mortality among patients with pulmonary nocardiosis.
Brinkman index 4428 Age 468 years Pneumoconiosis Healed or active pulmonary tuberculosis Pulmonary aspergillosis TMP/SMZ resistance
Hazard ratio (95% CI)
P value
3.0 3.6 0.43 0.66 3.6 2.8
0.020n 0.0057n 0.21 0.49 0.048n 0.029n
(1.2–8.4) (1.4–10.2) (0.068–1.5) (0.15–2.0) (1.0–11) (1.1–6.5)
CI ¼confidence interval; TMP/SMZ¼ trimethoprim/sulfamethoxazole. n Statistically significant value (po0.05).
Table 5 – Multivariate analysis of risk factors for overall mortality among patients with pulmonary nocardiosis.a
Brinkman index 4428 Age 468 years Pulmonary aspergillosis TMP/SMZ resistance
Hazard ratio (95% CI)
P value
2.2 4.7 8.8 4.3
0.11 0.005† 0.001† 0.004†
(0.8–6.6) (1.6–14) (2.4–33) (1.6–11)
CI ¼confidence interval; TMP/SMZ¼ trimethoprim/sulfamethoxazole. † Statistically significant value (po0.05). a Adjusted for pulmonary malignancy.
4.
Discussion
The present study describes the characteristics of patients with pulmonary nocardiosis and the factors associated with a poor prognosis. Although there have been some studies of pulmonary nocardiosis, they have generally dealt with small numbers of cases [3–8]. To our knowledge, ours is the largest study of pulmonary nocardiosis that includes clinical characteristics and prognostic factors. Our findings that older age, pulmonary aspergillosis, and TMP-SMZ resistance are independent and significant risk factors for overall mortality have not previously been reported in the published literature. Given the high prevalence of underlying pulmonary diseases and the potential impact of pulmonary nocardiosis, these findings have the potential to greatly influence daily clinical practice. Nocardia species are gram-positive aerobic bacilli that can cause a wide variety of infections [1]. Their taxonomy is under continuous revision; the final identification for each species may require confirmation by molecular techniques such as 16S ribosomal RNA sequencing and polymerase chain reaction analysis, which may change the initial biochemical identification. N. asteroides, formerly considered the most common species associated with human disease, has been redefined as a complex that includes N. asteroides sensu stricto, N. farcinica, N. nova, and Nocardia transvalensis complex. Most patients in the present study were male patients, which is consistent with previously published reports [8,9]. Smoking status tended to be associated with higher mortality, but this was not significant after multivariate analysis. Smoking is a risk factor for a number of pulmonary infections, probably because of its adverse effects on respiratory defenses [10]. Patients older than 68 years also had a higher risk of mortality, consistent with the increased risk
associated with pulmonary infection in elderly patients in general [11]. The potential for immunosuppression in elderly patients may contribute to the poorer outcomes in this group. Generally, the use of immunosuppressants (such as those used after solid organ transplantation and hematopoietic stem cell transplantation) and corticosteroid therapies are the main risk factors for nocardial infection [12]. However, a US study found no evidence of underlying illness or immunosuppressive treatment in 15% of patients with nocardiosis [13]. It is clear, therefore, that pulmonary nocardiosis may occur in non-immunocompromised patients. Although most patients in the present study did not have any apparent immunosuppression, 88% of patients had at least one underlying pulmonary disease; this was consistent with previous studies of pulmonary nocardiosis [5–7]. In a previous study, 19% of patients with pulmonary nocardiosis had preexisting structural abnormalities of the lung [14], and in the present study, both pulmonary malignancy and aspergillosis were associated with a higher risk of mortality. Although it is conceivable that a poor cancer prognosis contributed to the poor outcome of pulmonary nocardiosis, lung damage or changes in the lung structure due to pulmonary malignancy or aspergillosis may play an important role in nocardial infection of the human airway. These pulmonary structural abnormalities may be major risk factors for pulmonary nocardiosis in non-immunocompromised patients. Other pulmonary diseases such as chronic obstructive pulmonary disease, pneumoconiosis, mycobacterial infection, and bronchiectasis may also be risk factors for pulmonary nocardiosis, as previously reported [4,7,9,15–18], but these were not significant factors in the present study. Microbiological findings showed that 27% of cultures were resistant to TMP-SMZ, including most N. farcinica and N. otitidiscaviarum isolates. Generally, N. farcinica is associated with antibiotic resistance and poor outcomes [19,20]. However,
respiratory investigation 52 (2014) 160 –166
antimicrobial susceptibility patterns for this pathogen have varied from one report to another [21,22]; it was recently reported that problems exist in the inter-laboratory reproducibility of TMP-SMZ susceptibility testing [23]. Hence, the choice to not administer TMP-SMZ based on inaccurate susceptibility results may be a real risk factor for overall mortality, rather than TMP-SMZ resistance per se. In the present study, the most common CT manifestations of pulmonary nocardiosis were air-space consolidation and nodules. Although there is no specific radiologic pattern for pulmonary nocardiosis, nodules and cavities have commonly been reported and may help when establishing the diagnosis [5,24–28]. However, only two patients had cavitary lesions in this study, possibly because most of our patients did not have apparent immunosuppression (cavitation is especially common among immunosuppressed patients, such as those infected with human immunodeficiency virus) [28,29]. Moreover, in the present study, all patients with the cavitary form of pulmonary nocardiosis had mixed infection with Aspergillus or Mycobacterium species. Thirteen patients were deemed to have an indeterminate pattern, because the effects of the co-infecting organisms were indistinguishable from those of nocardiosis. Finally, most of the patients with pulmonary nocardiosis had underlying pulmonary diseases, decreasing the likely utility of radiographic findings in establishing the diagnosis for pulmonary nocardiosis. This study has some limitations that should be considered. First, this was a retrospective study, and as a result, there are many potential biases which we were not able to control. For example, our hospital was designed to manage patients with complicated pulmonary diseases. Therefore, most patients tended to have underlying pulmonary conditions; this may not be the case in other hospitals. In our population, no patient had apparent immunosuppression, and only one patient had disseminated nocardiosis. A second important limitation of the study is that molecular technology was not available, meaning that we could not reidentify Nocardia species that were initially identified more than 10 years ago. Because the agreement between molecular techniques and conventional methods ranges between 70% and 90% [30], refining the classification of Nocardia species would not have substantially changed the main outcomes of the present study. Therefore, the traditional microbiological notation system was used in this study. Third, due to underlying pulmonary conditions, there were some difficulties in distinguishing nocardial airway colonization from pulmonary nocardiosis. We excluded patients who had no symptoms of respiratory infection despite the isolation of Nocardia species. Fourth, 12 patients with pulmonary malignancy were included in the present study; the majority of them died from their underlying malignancy. Therefore, we believe that the adjustment for malignancy in the multivariate analysis is appropriate, and will minimize the effect of pulmonary malignancy on the prognosis of pulmonary nocardiosis. Despite these limitations, our study included the largest sample size of any study on pulmonary nocardiosis published to date, and included detailed clinical data. This study may help provide new information on the risk factors for mortality in patients with pulmonary nocardiosis.
5.
165
Conclusions
Pulmonary nocardiosis is not restricted to immunosuppressed patients. The disease can occur in immunocompetent patients who have an underlying pulmonary disease, with the most common being mycobacterial infection, chronic obstructive pulmonary disease, and malignancy. The predominant abnormality on chest CT in pulmonary nocardiosis was airspace consolidation, which may have been associated with cavitation. In patients with pulmonary nocardiosis, older age, pulmonary aspergillosis, and TMPSMZ resistance were associated with higher mortality risk. Clinicians should therefore be aware that pulmonary nocardiosis can occur even in non-immunosuppressed patients.
Conflict of interest The authors have no conflicts of interest.
Acknowledgment The investigators would like to thank Mr. Motohisa Tomita, Department of Clinical Laboratory, National Hospital Organization Kinki-Chuo Chest Medical Center, for microbiological analysis.
r e f e r e nc e s
[1] Saubolle MA, Sussland D. Nocardiosis: review of clinical and laboratory experience. J Clin Microbiol 2003;41:4497–501. [2] Brown-Elliott BA, Brown JM, Conville PS, et al. Clinical and laboratory features of the Nocardia spp. based on current molecular taxonomy. Clin Microbiol Rev 2006;19:259–82. [3] Menéndez R, Cordero PJ, Santos M, et al. Pulmonary infection with Nocardia species: a report of 10 cases and review. Eur Respir J 1997;10:1542–6. [4] Garcia-Bellmunt L, Sibila O, Solanes I, et al. Pulmonary nocardiosis in patients with COPD: characteristics and prognostic factors. Arch Bronconeumol 2012;48:280–5. [5] Hui CH, Au VW, Rowland K, et al. Pulmonary nocardiosis revisited: experience of 35 patients at diagnosis. Respir Med 2003;97:709–17. [6] Chen YC, Lee CH, Chien CC, et al. Pulmonary nocardiosis in southern Taiwan. J Microbiol Immunol Infect 2012. http://dx. doi.org/10.1016/j.jmii.2012.07.017. [7] Martínez R, Menéndez R, Reyes S, et al. Pulmonary nocardiosis: risk factors and outcomes. Respirology 2007;12:394–400. [8] Pintado V, Gómez-Mampaso E, Fortún J, et al. Infection with Nocardia species: Clinical spectrum of disease and species distribution in Madrid, Spain, 1978–2001. Infection 2002;30:338–40. [9] Mari B, Montón C, Mariscal D, et al. Pulmonary nocardiosis: clinical experience in ten cases. Respiration 2001;68:382–8. [10] McCusker K. Mechanisms of respiratory tissue injury from cigarette smoking. Am J Med 1992;93:18S–21S. [11] Marrie TJ. Community-acquired pneumonia in the elderly. Clin Infect Dis 2000;31:1066–78. [12] Minero MV, Marín M, Cercenado E, et al. Nocardiosis at the turn of the century. Medicine (Baltimore) 2009;88:250–61.
166
respiratory investigation 52 (2014) 160 –166
[13] Beaman BL, Burnside J, Edwards B, et al. Nocardial infections in the United States, 1972–1974. J Infect Dis 1976;134:286–9. [14] Blackmon KN, Ravenel JG, Gomez JM, et al. Pulmonary nocardiosis: Computed tomography features at diagnosis. J Thorac Imaging 2011;26:224–9. [15] Rivière F, Billhot M, Soler C, et al. Pulmonary nocardiosis in immunocompetent patients: can COPD be the only risk factor? Eur Respir Rev 2011;20:210–2. [16] Nakamura S, Mihara T, Hitotsumatsu T, et al. Case report of disseminated nocardiosis complicated in an elderly person with pneumoconiosis. Kansenshogaku Zasshi 2006;80:721–5. [17] Singh M, Sandhu RS, Randhawa HS, et al. Prevalence of pulmonary nocardiosis in a tuberculosis hospital in Amritsar, Punjab. Indian J Chest Dis Allied Sci 2000;42:325–39. [18] Cremades MJ, Menéndez R, Santos M, et al. Repeated pulmonary infection by Nocardia asteroides complex in a patient with bronchiectasis. Respiration 1998;65:211–3. [19] Wallace Jr RJ, Tsukamura M, Brown BA, et al. Cefotaximeresistant Nocardia asteroides strains are isolates of the controversial species Nocardia farcinica. J Clin Microbiol 1990;28:2726–32. [20] Husain S, McCurry K, Dauber J, et al. Nocardia infection in lung transplant recipients. J Heart Lung Transplant 2002;21:354–9. [21] Brown-Elliott BA, Biehle J, Conville PS, et al. Sulfonamide resistance in isolates of Nocardia spp. from a US multicenter survey. J Clin Microbiol 2012;50:670–2.
[22] Uhde KB, Pathak S, McCullum Jr I, et al. Antimicrobialresistant nocardia isolates, United States, 1995–2004. Clin Infect Dis 2010;51:1445–8. [23] Conville PS, Brown-Elliott BA, Wallace Jr RJ, et al. Multisite reproducibility of the broth microdilution method for susceptibility testing of Nocardia species. J Clin Microbiol 2012;50:1270–80. [24] Martínez R, Reyes S, Menéndez R. Pulmonary nocardiosis: risk factors, clinical features, diagnosis and prognosis. Curr Opin Pulm Med 2008;14:219–27. [25] Kanne JP, Yandow DR, Mohammed TL, et al. CT findings of pulmonary nocardiosis. Am J Roentgenol 2011;197:W266–72. [26] Oszoyoglu AA, Kirsch J, Mohammed TL. Pulmonary nocardiosis after lung transplantation: CT findings in 7 patients and review of the literature. J Thorac Imaging 2007;22:143–8. [27] Yoon HK, Im JG, Ahn JM, et al. Pulmonary nocardiosis: CT findings. J Comput Assist Tomogr 1995;19:52–5. [28] Kramer MR, Uttamchandani RB. The radiographic appearance of pulmonary nocardiosis associated with AIDS. Chest 1990;98:382–5. [29] Biscione F, Cecchini D, Ambrosioni J, et al. Nocardiosis in patients with human immunodeficiency virus infection. Enferm Infecc Microbiol Clin 2005;23:419–23. [30] Cloud JL, Conville PS, Croft A, et al. Evaluation of partial 16S ribosomal DNA sequencing for identification of Nocardia species by using the MicroSeq 500system with an expanded database. J Clin Microbiol 2004;42:578–84.
Contents lists available at ScienceDirect
Respiratory Investigation journal homepage: www.elsevier.com/locate/resinv
Original article
Pulmonary nocardiosis: A clinical analysis of 59 cases Yu Kuraharaa, Kazunobu Tachibanaa,b, Kazunari Tsuyuguchib, Masanori Akirac, Katsuhiro Suzukia,n, Seiji Hayashia a
Department of Internal Medicine, National Hospital Organization Kinki-Chuo Chest Medical Center, Osaka, Japan Clinical Research Center, National Hospital Organization Kinki-Chuo Chest Medical Center, Osaka, Japan c Department of Radiology, National Hospital Organization Kinki-Chuo Chest Medical Center, Osaka, Japan b
ar t ic l e in f o
abs tra ct
Article history:
Background: Pulmonary nocardiosis is a rare but severe infection caused by Nocardia
Received 15 August 2013
species. This study aimed at describing the clinical characteristics and prognosis of
Received in revised form
pulmonary nocardiosis.
5 September 2013
Methods: An observational, retrospective study was undertaken of patients diagnosed with
Accepted 26 September 2013
pulmonary nocardiosis over a 13-year period at the Kinki-Chuo Chest Medical Center,
Available online 7 November 2013
Osaka, Japan.
Keywords: Clinical characteristics Nocardia Pulmonary nocardiosis Radiographic findings Survival
Results: Seven patients with airway nocardial colonization and 59 patients with pulmonary nocardiosis were identified, one of whom had disseminated nocardiosis. Patients with pulmonary nocardiosis were predominantly male patients (73%), with a mean age of 66 (range, 15–88) years. New-onset cough and dyspnea were the most common manifestations (76%). Although 52 (88%) patients had at least one underlying pulmonary disease, most patients did not appear to be systemically immunocompromised. The predominant abnormality on chest computed tomography in pulmonary nocardiosis was airspace consolidation (52%), sometimes associated with cavitation. Multivariate Cox proportional-hazards analysis revealed the following significant and independent risk factors for overall mortality: age 468 years (hazard ratio [HR], 4.7; 95% confidence interval [CI], 1.6–14; p¼0.05), pulmonary aspergillosis (HR, 8.8; 95% CI, 2.4–33; p¼ 0.01), and trimethoprim/sulfamethoxazole (TMP-SMZ) resistance (HR, 4.3; 95% CI, 1.6– 11; p¼ 0.04). Conclusions: Clinicians should be aware that pulmonary nocardiosis can occur even in immunocompetent patients, especially those with an underlying pulmonary disease. In pulmonary nocardiosis, older age, pulmonary aspergillosis, and TMP-SMZ resistance are associated with increased risk of mortality. & 2013 The Japanese Respiratory Society. Published by Elsevier B.V. All rights reserved.
Abbreviations: CI, confidence interval; CT, computed tomography; HR, hazard ratio; TMP-SMZ, trimethoprim/sulfamethoxazole n Corresponding author. Tel.: þ81 72 252 3021; fax: þ81 72 251 1372. E-mail addresses: [email protected] (Y. Kurahara), [email protected] (K. Tachibana), [email protected] (K. Tsuyuguchi), [email protected] (M. Akira), [email protected], [email protected] (K. Suzuki), [email protected] (S. Hayashi). 2212-5345/$ - see front matter & 2013 The Japanese Respiratory Society. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.resinv.2013.09.004
respiratory investigation 52 (2014) 160 –166
1.
Introduction
The Nocardia genus of aerobic gram-positive bacteria causes a range of infectious diseases, including isolated pulmonary and skin infections and disseminated disease [1]. Pulmonary nocardiosis is a rare but severe infection that commonly presents as a subacute or chronic disease, mimicking lung cancer or other pulmonary infections such as mycosis or bacterial pneumonia. There are more than 90 Nocardia species, at least 33 of which cause disease in humans [2]. Because of changes in taxonomy and the difficulty inherent in the routine identification of Nocardia strains at the species level, there have been few published studies concerning pulmonary nocardiosis [3–8]. Of those that have been conducted, most have investigated only patients with immunosuppression and have included only a small number of patients. Moreover, the details of the clinical characteristics and prognosis of pulmonary nocardiosis remain unclear. The objective of the present study is to clarify the characteristics of pulmonary nocardial patients and to elucidate the factors that determine its etiology and prognosis.
tests, growth characteristics, and antimicrobial susceptibility patterns. Hence, the microbiological notation was rendered in its traditional form in this study. The susceptibility of the isolates was determined by the disc diffusion method. A total of 90 specimens from 66 patients were identified as cultureproven Nocardia.
2.3.
Materials and methods
2.1.
Study Subjects
We aimed at studying the clinical pattern of pulmonary nocardiosis in the Kinki-Chuo Chest Medical Center, Osaka, Japan, based on a retrospective review of cases from January 1999 to October 2012. The hospital is a 450-bed respiratory disease center designed to manage patients with various respiratory diseases. In the present study, patients with no evidence of pulmonary disease or symptoms of respiratory infection despite the isolation of Nocardia species were defined as having airway nocardial colonization. Pulmonary nocardiosis was defined as the presence of symptoms of respiratory infection consistent with pulmonary nocardiosis, with Nocardia species isolated from respiratory sample cultures on at least one occasion. Disseminated nocardiosis was defined as an infection in two noncontiguous organs or in the central nervous system. This study was approved by the Institutional Review Board at the Kinki-Chuo Chest Medical Center (Approval date: November 14, 2011; Approval number: 388).
2.2.
Radiographic assessment
Three pulmonary infectious disease experts (Y.K., K. Tachibana., and K.S.) and one chest radiologist (M.A.) retrospectively reviewed de-identified computed tomography (CT) films. The reviewers classified each case based on the presence of the following features: (1) airspace consolidation; (2) nodules (defined as those with diameters o3 cm); (3) cavity; (4) consolidation with cavity; (5) pleural effusion; and (6) indeterminate (we anticipated the possibility that some cases had an underlying pulmonary condition indistinguishable from pulmonary nocardiosis because of the complicated structure of lung fields). Any inconsistencies were resolved by consensus.
2.4.
2.
161
Study variables
The following data were collected: patient demographics, comorbidities, immunological state, clinical manifestations, radiographic findings, bacteriological reports, susceptibility to trimethoprim/sulfamethoxazole (TMP-SMZ), treatment course, and survival time from diagnosis.
2.5.
Statistical analysis
Statistical analysis was performed with SPSS version 15.0J software (SPSS Inc., Chicago, Illinois), and the Kaplan–Meier method was used for survival analysis. If the patient died or was lost to follow-up during the study period, this was recorded as censored data. A univariate assessment of selected risk factors was performed using a Cox proportional hazard model. To eliminate confounding factors in predicting mortality risk, variables with P values o0.05 on univariate analysis were then entered into a multivariate assessment. We used a Cox proportional hazard model with a step-up procedure, utilizing the likelihood ratio as the criterion for adding significant variables. Results are expressed as hazard ratios with corresponding 95% confidence intervals (CI). A P value of o0.05 was regarded as significant.
Microbiological identification
In order to diagnose pulmonary nocardiosis, samples were examined microscopically after Gram staining in order to determine the presence of microorganisms and establish specimen quality. The presumptive identification of Nocardia species was based on the microscopic characteristics of the isolates and the macroscopic morphology of the colonies. Although 16S ribosomal RNA sequence-based identification of Nocardia species has recently become the gold standard, this technology was not available in our laboratories during the study period. Therefore, Nocardia isolates were identified using a combination of traditional standard biochemical
3.
Results
3.1.
Characteristics and clinical features of the patients
A total of 90 respiratory specimens from 66 patients were identified as culture-proven Nocardia. Seven patients were diagnosed with airway nocardial colonization because they had minimal or no lung-field lesions and the absence of respiratory symptoms. These lesions remained stable after at least 1 year of follow-up, during which time no treatment was given. In total, 59 patients were diagnosed with pulmonary nocardiosis (43 male patients and 16 female patients). One
162
respiratory investigation 52 (2014) 160 –166
patient had disseminated nocardiosis of the central nervous system. The characteristics and underlying diseases of patients with airway nocardial colonization and pulmonary nocardiosis are detailed in Table 1. The mean age was 66 years (range, 15–88 years). A total of 55 (93%) of 59 patients had at least one underlying pulmonary disease. Pre-existing structural abnormalities of the lung were common. Notably, most patients did not appear to be immunocompromised. Only one patient was being treated with corticosteroids (for systemic vasculitis); no other patient was receiving immunosuppressive agents. No patient had undergone transplantation, and no patient had human immunodeficiency virus infection or were alcoholic. All patients with airway nocardial colonization were male patients, and their mean age was 71 years.
3.2. Symptoms, laboratory data, and radiographic patterns At the time when Nocardia infection was identified, 45 patients with pulmonary nocardiosis (76%) had new-onset cough and dyspnea, which was associated with a fever higher than 37.5 1C in 36 patients (61%). In one case, dissemination to the central nervous system was detected, and that patient died 55 days after diagnosis. Blood examination confirmed
Table 1 – Characteristics of airway nocardial colonization and pulmonary nocardiosis.
Male/female Age, years, mean7SD Never smoker/current or former smoker Underlying condition Non-pulmonary underlying condition Cardiovascular diseases Diabetes mellitus Connective tissue diseases Hepatic diseases Renal failure Pulmonary underlying condition Chronic obstructive pulmonary disease Pulmonary malignancy Pneumoconiosis Healed or active pulmonary tuberculosis Non-tuberculous mycobacterial infection Pulmonary aspergillosis Bronchiectasis Othersa
Airway nocardial colonization (n ¼7)
Pulmonary nocardiosis (n ¼59)
7/0 71717 1/9
43/16 66713 39/20
4 (57) 1 (14)
55 (93) 12 (20)
1 (14) 0 0
5 (8.5) 4 (6.8) 2 (3.4)
0 0 3 (43)
1 (1.7) 1 (1.7) 52 (88)
1 (14)
13 (22)
1 (14) 0 0
12 (20) 11 (19) 11 (19)
0
9 (15)
0 1 (14) 0
6 (10) 6 (10) 3 (5.1)
Data are presented as mean7SD or n (%). a Includes bronchial asthma, idiopathic pulmonary fibrosis, and chronic hypersensitivity pneumonitis.
leukocytosis in 20 patients (34%) and elevated C-reactive protein levels in 43 patients (73%). Radiographic findings in the present study are given in Table 2. The predominant abnormality on chest CT was airspace consolidation (Fig. 1), which may have been associated with cavitation. The second predominant abnormality was the presence of nodules (Fig. 2). Because a majority (88%) of the patients had at least one baseline pulmonary disease, radiographic patterns were indeterminate in 12 patients (20%). For example, mixed active infection with other organisms (such as Aspergillus species or Mycobacterium species) resulted in a radiographic appearance indistinguishable from that of pulmonary nocardiosis. All patients with airway nocardial colonization had tiny nodules in the lung fields, and these nodules persisted for at least 1 year of follow-up in the absence of treatment.
3.3.
Microbiological identification
In the present study, microbiological notation was rendered in its traditional form. Because Nocardia taxonomy was under continuous revision during the study period, we were not able to re-identify all of the Nocardia species using 16S ribosomal RNA sequence analysis. The species isolated from the respiratory samples in the present study were the following: Nocardia asteroides (40 patients [68%]), Nocardia farcinica (six patients [10%]), Nocardia nova (six patients [10%]), Nocardia otitidiscaviarum (four patients [6.8%]), and Nocardia brasiliensis (three patients [5.1%]). Infection was polymicrobial in 17 cases (30%), with the following organisms: Aspergillus fumigatus (four patients [6.8%]), Aspergillus niger (two patients [3.4%]), Mycobacterium tuberculosis (two patients [3.4%]), Mycobacterium avium complex (six patients [10%]), and Mycobacterium kansasii (three patients [5.1%]). Thirty-six (61%) patients were treated with TMP-SMZ (Table 3). Sixteen patients (27%) had contracted TMP-SMZ-resistant Nocardia infection: specifically, those with N. farcinica (5 of 6), N. otitidiscaviarum (4 of 4), and N. asteroides infection (7 of 40). Most patients infected with TMP-SMZ-resistant pathogens were treated with imipenem/cilastatin (Table 1).
Table 2 – Radiographic (computed tomography) presentations among patients with airway nocardial colonization and pulmonary nocardiosis.
Air-space consolidation Nodules Consolidation with cavity Cavity Pleural effusion Indeterminatea Total a
Airway nocardial colonization (n ¼7)
Pulmonary nocardiosis (n ¼ 59)
0
28 (47)
7 (100) 0
13 (22) 3 (5.1)
0 0 0 7 (100)
2 1 12 59
(3.4) (1.7) (20) (100)
Underlying pulmonary condition indistinguishable from pulmonary nocardiosis.
respiratory investigation 52 (2014) 160 –166
163
Fig. 1 – A 66-year-old man with pulmonary nocardiosis and idiopathic pulmonary fibrosis. Left: initial computed tomography (CT) scan shows bilateral subpleural reticulation and cystic lesions. Right: CT scan, obtained when the patient developed highgrade fever and cough, shows consolidation with air bronchogram in the left upper lobe.
Fig. 2 – A 77-year-old man with pulmonary nocardiosis and silicosis. Left: initial computed tomography (CT) scan shows bilateral small nodules consistent with silicosis. Right: CT scan, obtained when the patient developed cough and yellow sputum, shows nodules in the right lower lobe. Table 3 – Trimethoprim–sulfamethoxazole susceptibility and treatment among patients with airway nocardial colonization and pulmonary nocardiosis. Airway nocardial colonization (n ¼7) TMP/SMZ susceptibility Susceptible 7 (100) Resistant 0 Treatment TMP/SMZ Imipenem/ cilastatin Meropenem Minocycline Surgery Othersa
– – – – – –
Data are presented as n (%). a Includes ceftriaxone, linezolid, and SMZ ¼trimethoprim/sulfamethoxazole.
3.4.
Pulmonary nocardiosis (n¼ 59)
43 (73) 16 (27)
36 (61) 13 (22) 2 2 2 3
Fig. 3 – Kaplan–Meier estimates of overall survival in patients with pulmonary nocardiosis. Survival was calculated from the day of first diagnosis of the infection. Data were censored at the end of the follow-up period or at the time of death.
(3.4) (3.4) (3.4) (5.1)
amikacin.
TMP/
Risk factors
Survival rates from the date of diagnosis of pulmonary nocardiosis were 97% at 30 days, 83% at 90 days, and 74% at 180 days (Fig. 3). To determine the risk factors associated with mortality, we analyzed the data using Cox proportional hazard regression modeling. After univariate analysis, five significant risk factors
(po0.05) for mortality were identified: a history of heavy smoking (Brinkman index 4428), age 468 years, pulmonary malignancy, pulmonary aspergillosis, and TMP-SMZ resistance (Table 4). These significant risk factors identified by univariate analysis were then entered into multivariate analysis using a Cox proportional hazard regression model. The risk factors independently associated with overall mortality were age 468 years (hazard ratio [HR], 4.7; 95% confidence interval [CI], 1.6–14; p¼0.05), pulmonary aspergillosis (HR, 8.8; 95% CI, 2.4–33; p¼ 0.01), and TMP-SMZ resistance (HR, 4.3; 95% CI, 1.6–11; p¼ 0.04) (Table 5).
164
respiratory investigation 52 (2014) 160 –166
Table 4 – Univariate analysis of risk factors for overall mortality among patients with pulmonary nocardiosis.
Brinkman index 4428 Age 468 years Pneumoconiosis Healed or active pulmonary tuberculosis Pulmonary aspergillosis TMP/SMZ resistance
Hazard ratio (95% CI)
P value
3.0 3.6 0.43 0.66 3.6 2.8
0.020n 0.0057n 0.21 0.49 0.048n 0.029n
(1.2–8.4) (1.4–10.2) (0.068–1.5) (0.15–2.0) (1.0–11) (1.1–6.5)
CI ¼confidence interval; TMP/SMZ¼ trimethoprim/sulfamethoxazole. n Statistically significant value (po0.05).
Table 5 – Multivariate analysis of risk factors for overall mortality among patients with pulmonary nocardiosis.a
Brinkman index 4428 Age 468 years Pulmonary aspergillosis TMP/SMZ resistance
Hazard ratio (95% CI)
P value
2.2 4.7 8.8 4.3
0.11 0.005† 0.001† 0.004†
(0.8–6.6) (1.6–14) (2.4–33) (1.6–11)
CI ¼confidence interval; TMP/SMZ¼ trimethoprim/sulfamethoxazole. † Statistically significant value (po0.05). a Adjusted for pulmonary malignancy.
4.
Discussion
The present study describes the characteristics of patients with pulmonary nocardiosis and the factors associated with a poor prognosis. Although there have been some studies of pulmonary nocardiosis, they have generally dealt with small numbers of cases [3–8]. To our knowledge, ours is the largest study of pulmonary nocardiosis that includes clinical characteristics and prognostic factors. Our findings that older age, pulmonary aspergillosis, and TMP-SMZ resistance are independent and significant risk factors for overall mortality have not previously been reported in the published literature. Given the high prevalence of underlying pulmonary diseases and the potential impact of pulmonary nocardiosis, these findings have the potential to greatly influence daily clinical practice. Nocardia species are gram-positive aerobic bacilli that can cause a wide variety of infections [1]. Their taxonomy is under continuous revision; the final identification for each species may require confirmation by molecular techniques such as 16S ribosomal RNA sequencing and polymerase chain reaction analysis, which may change the initial biochemical identification. N. asteroides, formerly considered the most common species associated with human disease, has been redefined as a complex that includes N. asteroides sensu stricto, N. farcinica, N. nova, and Nocardia transvalensis complex. Most patients in the present study were male patients, which is consistent with previously published reports [8,9]. Smoking status tended to be associated with higher mortality, but this was not significant after multivariate analysis. Smoking is a risk factor for a number of pulmonary infections, probably because of its adverse effects on respiratory defenses [10]. Patients older than 68 years also had a higher risk of mortality, consistent with the increased risk
associated with pulmonary infection in elderly patients in general [11]. The potential for immunosuppression in elderly patients may contribute to the poorer outcomes in this group. Generally, the use of immunosuppressants (such as those used after solid organ transplantation and hematopoietic stem cell transplantation) and corticosteroid therapies are the main risk factors for nocardial infection [12]. However, a US study found no evidence of underlying illness or immunosuppressive treatment in 15% of patients with nocardiosis [13]. It is clear, therefore, that pulmonary nocardiosis may occur in non-immunocompromised patients. Although most patients in the present study did not have any apparent immunosuppression, 88% of patients had at least one underlying pulmonary disease; this was consistent with previous studies of pulmonary nocardiosis [5–7]. In a previous study, 19% of patients with pulmonary nocardiosis had preexisting structural abnormalities of the lung [14], and in the present study, both pulmonary malignancy and aspergillosis were associated with a higher risk of mortality. Although it is conceivable that a poor cancer prognosis contributed to the poor outcome of pulmonary nocardiosis, lung damage or changes in the lung structure due to pulmonary malignancy or aspergillosis may play an important role in nocardial infection of the human airway. These pulmonary structural abnormalities may be major risk factors for pulmonary nocardiosis in non-immunocompromised patients. Other pulmonary diseases such as chronic obstructive pulmonary disease, pneumoconiosis, mycobacterial infection, and bronchiectasis may also be risk factors for pulmonary nocardiosis, as previously reported [4,7,9,15–18], but these were not significant factors in the present study. Microbiological findings showed that 27% of cultures were resistant to TMP-SMZ, including most N. farcinica and N. otitidiscaviarum isolates. Generally, N. farcinica is associated with antibiotic resistance and poor outcomes [19,20]. However,
respiratory investigation 52 (2014) 160 –166
antimicrobial susceptibility patterns for this pathogen have varied from one report to another [21,22]; it was recently reported that problems exist in the inter-laboratory reproducibility of TMP-SMZ susceptibility testing [23]. Hence, the choice to not administer TMP-SMZ based on inaccurate susceptibility results may be a real risk factor for overall mortality, rather than TMP-SMZ resistance per se. In the present study, the most common CT manifestations of pulmonary nocardiosis were air-space consolidation and nodules. Although there is no specific radiologic pattern for pulmonary nocardiosis, nodules and cavities have commonly been reported and may help when establishing the diagnosis [5,24–28]. However, only two patients had cavitary lesions in this study, possibly because most of our patients did not have apparent immunosuppression (cavitation is especially common among immunosuppressed patients, such as those infected with human immunodeficiency virus) [28,29]. Moreover, in the present study, all patients with the cavitary form of pulmonary nocardiosis had mixed infection with Aspergillus or Mycobacterium species. Thirteen patients were deemed to have an indeterminate pattern, because the effects of the co-infecting organisms were indistinguishable from those of nocardiosis. Finally, most of the patients with pulmonary nocardiosis had underlying pulmonary diseases, decreasing the likely utility of radiographic findings in establishing the diagnosis for pulmonary nocardiosis. This study has some limitations that should be considered. First, this was a retrospective study, and as a result, there are many potential biases which we were not able to control. For example, our hospital was designed to manage patients with complicated pulmonary diseases. Therefore, most patients tended to have underlying pulmonary conditions; this may not be the case in other hospitals. In our population, no patient had apparent immunosuppression, and only one patient had disseminated nocardiosis. A second important limitation of the study is that molecular technology was not available, meaning that we could not reidentify Nocardia species that were initially identified more than 10 years ago. Because the agreement between molecular techniques and conventional methods ranges between 70% and 90% [30], refining the classification of Nocardia species would not have substantially changed the main outcomes of the present study. Therefore, the traditional microbiological notation system was used in this study. Third, due to underlying pulmonary conditions, there were some difficulties in distinguishing nocardial airway colonization from pulmonary nocardiosis. We excluded patients who had no symptoms of respiratory infection despite the isolation of Nocardia species. Fourth, 12 patients with pulmonary malignancy were included in the present study; the majority of them died from their underlying malignancy. Therefore, we believe that the adjustment for malignancy in the multivariate analysis is appropriate, and will minimize the effect of pulmonary malignancy on the prognosis of pulmonary nocardiosis. Despite these limitations, our study included the largest sample size of any study on pulmonary nocardiosis published to date, and included detailed clinical data. This study may help provide new information on the risk factors for mortality in patients with pulmonary nocardiosis.
5.
165
Conclusions
Pulmonary nocardiosis is not restricted to immunosuppressed patients. The disease can occur in immunocompetent patients who have an underlying pulmonary disease, with the most common being mycobacterial infection, chronic obstructive pulmonary disease, and malignancy. The predominant abnormality on chest CT in pulmonary nocardiosis was airspace consolidation, which may have been associated with cavitation. In patients with pulmonary nocardiosis, older age, pulmonary aspergillosis, and TMPSMZ resistance were associated with higher mortality risk. Clinicians should therefore be aware that pulmonary nocardiosis can occur even in non-immunosuppressed patients.
Conflict of interest The authors have no conflicts of interest.
Acknowledgment The investigators would like to thank Mr. Motohisa Tomita, Department of Clinical Laboratory, National Hospital Organization Kinki-Chuo Chest Medical Center, for microbiological analysis.
r e f e r e nc e s
[1] Saubolle MA, Sussland D. Nocardiosis: review of clinical and laboratory experience. J Clin Microbiol 2003;41:4497–501. [2] Brown-Elliott BA, Brown JM, Conville PS, et al. Clinical and laboratory features of the Nocardia spp. based on current molecular taxonomy. Clin Microbiol Rev 2006;19:259–82. [3] Menéndez R, Cordero PJ, Santos M, et al. Pulmonary infection with Nocardia species: a report of 10 cases and review. Eur Respir J 1997;10:1542–6. [4] Garcia-Bellmunt L, Sibila O, Solanes I, et al. Pulmonary nocardiosis in patients with COPD: characteristics and prognostic factors. Arch Bronconeumol 2012;48:280–5. [5] Hui CH, Au VW, Rowland K, et al. Pulmonary nocardiosis revisited: experience of 35 patients at diagnosis. Respir Med 2003;97:709–17. [6] Chen YC, Lee CH, Chien CC, et al. Pulmonary nocardiosis in southern Taiwan. J Microbiol Immunol Infect 2012. http://dx. doi.org/10.1016/j.jmii.2012.07.017. [7] Martínez R, Menéndez R, Reyes S, et al. Pulmonary nocardiosis: risk factors and outcomes. Respirology 2007;12:394–400. [8] Pintado V, Gómez-Mampaso E, Fortún J, et al. Infection with Nocardia species: Clinical spectrum of disease and species distribution in Madrid, Spain, 1978–2001. Infection 2002;30:338–40. [9] Mari B, Montón C, Mariscal D, et al. Pulmonary nocardiosis: clinical experience in ten cases. Respiration 2001;68:382–8. [10] McCusker K. Mechanisms of respiratory tissue injury from cigarette smoking. Am J Med 1992;93:18S–21S. [11] Marrie TJ. Community-acquired pneumonia in the elderly. Clin Infect Dis 2000;31:1066–78. [12] Minero MV, Marín M, Cercenado E, et al. Nocardiosis at the turn of the century. Medicine (Baltimore) 2009;88:250–61.
166
respiratory investigation 52 (2014) 160 –166
[13] Beaman BL, Burnside J, Edwards B, et al. Nocardial infections in the United States, 1972–1974. J Infect Dis 1976;134:286–9. [14] Blackmon KN, Ravenel JG, Gomez JM, et al. Pulmonary nocardiosis: Computed tomography features at diagnosis. J Thorac Imaging 2011;26:224–9. [15] Rivière F, Billhot M, Soler C, et al. Pulmonary nocardiosis in immunocompetent patients: can COPD be the only risk factor? Eur Respir Rev 2011;20:210–2. [16] Nakamura S, Mihara T, Hitotsumatsu T, et al. Case report of disseminated nocardiosis complicated in an elderly person with pneumoconiosis. Kansenshogaku Zasshi 2006;80:721–5. [17] Singh M, Sandhu RS, Randhawa HS, et al. Prevalence of pulmonary nocardiosis in a tuberculosis hospital in Amritsar, Punjab. Indian J Chest Dis Allied Sci 2000;42:325–39. [18] Cremades MJ, Menéndez R, Santos M, et al. Repeated pulmonary infection by Nocardia asteroides complex in a patient with bronchiectasis. Respiration 1998;65:211–3. [19] Wallace Jr RJ, Tsukamura M, Brown BA, et al. Cefotaximeresistant Nocardia asteroides strains are isolates of the controversial species Nocardia farcinica. J Clin Microbiol 1990;28:2726–32. [20] Husain S, McCurry K, Dauber J, et al. Nocardia infection in lung transplant recipients. J Heart Lung Transplant 2002;21:354–9. [21] Brown-Elliott BA, Biehle J, Conville PS, et al. Sulfonamide resistance in isolates of Nocardia spp. from a US multicenter survey. J Clin Microbiol 2012;50:670–2.
[22] Uhde KB, Pathak S, McCullum Jr I, et al. Antimicrobialresistant nocardia isolates, United States, 1995–2004. Clin Infect Dis 2010;51:1445–8. [23] Conville PS, Brown-Elliott BA, Wallace Jr RJ, et al. Multisite reproducibility of the broth microdilution method for susceptibility testing of Nocardia species. J Clin Microbiol 2012;50:1270–80. [24] Martínez R, Reyes S, Menéndez R. Pulmonary nocardiosis: risk factors, clinical features, diagnosis and prognosis. Curr Opin Pulm Med 2008;14:219–27. [25] Kanne JP, Yandow DR, Mohammed TL, et al. CT findings of pulmonary nocardiosis. Am J Roentgenol 2011;197:W266–72. [26] Oszoyoglu AA, Kirsch J, Mohammed TL. Pulmonary nocardiosis after lung transplantation: CT findings in 7 patients and review of the literature. J Thorac Imaging 2007;22:143–8. [27] Yoon HK, Im JG, Ahn JM, et al. Pulmonary nocardiosis: CT findings. J Comput Assist Tomogr 1995;19:52–5. [28] Kramer MR, Uttamchandani RB. The radiographic appearance of pulmonary nocardiosis associated with AIDS. Chest 1990;98:382–5. [29] Biscione F, Cecchini D, Ambrosioni J, et al. Nocardiosis in patients with human immunodeficiency virus infection. Enferm Infecc Microbiol Clin 2005;23:419–23. [30] Cloud JL, Conville PS, Croft A, et al. Evaluation of partial 16S ribosomal DNA sequencing for identification of Nocardia species by using the MicroSeq 500system with an expanded database. J Clin Microbiol 2004;42:578–84.
