Candida Empyema Thoracis at a Hospital in Taiwan

Abstract

Background:

We conducted a study of the clinical manifestations and outcomes of empyema thoracis caused by Candida spp. in hospitalized patients.

Methods:

We identified patients who from January 2010 to June 2012 had signs of inflammation, such as fever and leukocytosis, and concomitant positive cultures of Candida spp. from specimens of pleural fluid. We reviewed retrospectively the medical records of all such patients.

Results:

Eight of the patients in whom we identified fever and leukocytosis with concomitant positive pleural fluid cultures of Candida spp. were found to have Candida empyema thoracis. Candida albicans was the most common species causing empyema (n=4), followed by C. glabrata (n=3) and C. tropicalis (n=1). Among the eight patients with Candida empyema thoracis, malignant disease was the most common underlying disease. Seven of the eight patients' episodes of Candida empyema thoracis (87.5%) were classified as health care-associated infections. One patient had a mixed infection with Candida spp. and Pseudomonas aeruginosa. None of the patients had concomitant candidemia. Fluconazole was the antifungal agent used most commonly in treating the patients, and all of the patients had drainage of pleural effusions, including tube thoracostomy in five patients and drainage through a pigtail catheter in three patients. One patient underwent video-assisted thoracic surgery for the management of empyema thoracis, and the patients' overall in-hospital mortality was 62.5%.

Conclusion:

Although Candida empyema thoracis is encountered rarely, it can develop in immunocompromised patients and can be associated with a high mortality rate.

Candida spp. have played a major role in health care-associated infection (HAI), especially in critically ill patients treated with an invasive procedure or device, those treated with broad-spectrum antimicrobial agents, and those receiving immunosuppressive agents or parenteral nutrition [1 –3]. Candida albicans remains the predominant cause of invasive candidiasis and accounts for more than 50% of all cases of this condition. However, the incidence of invasive candidiasis caused by species of Candida other than C. albicans, such as C. tropicalis, C. glabrata, and C. parapsilosis, is increasing [4,5]. The clinical manifestations of invasive candidiasis can vary according to the degree of a patient's immunosuppression and the type of candidal infection. Blood stream infection and urinary tract infection are the most common types of invasive candidiasis [6]. Recently, another rare clinical entity, Candida spp.-associated empyema thoracis, has been emerging [7 –11]. However, knowledge about this clinical entity is limited. We therefore conducted a study of the clinical manifestations and outcomes of patients with empyema thoracis caused by Candida spp.

Patients and Methods

Hospital setting and patient selection

This study was conducted at the Lioyung branch of the Chi Mei Medical Center, a 900-bed hospital in southern Taiwan. Using the computerized data base of the hospital's mycology laboratory, we identified patients who from January 2010 to June 2012 had signs of inflammation, such as fever (body temperature >38.3°C) and leukocytosis (white blood cell count [WBC] >10,000/mm³), and concomitant positive cultures of Candida spp. from pleural fluid [7]. The clinical charts of all patients with empyema caused by Candida spp. were reviewed retrospectively, and information was collected about each patient's age, gender, underlying immunocompromising conditions including a history of treatment with immunosuppressant drugs, diabetes mellitus, cirrhosis, end-stage renal disease, and malignant disease. Findings in the patients' laboratory examinations were collected, including the peripheral WBC and differential cell counts; serum lactate dehydrogenase (LDH) and C-reactive protein concentrations; characteristics of pleural effusions including cell counts, differential cell counts, total protein concentration, glucose concentration, and strains of pathogens isolated from culture of the effusions. The patients' modalities of treatment, including regimens of antifungal therapy, methods of drainage, and surgical management, were recorded.

Definitions

Empyema thoracis caused by Candida spp. was defined as the isolation of Candida spp. from exudative pleural effusions according to the criteria of Light et al. [12], together with signs of infection, such as fever and leukocytosis. Empyema was classified as a health care-associated infection (HAI) acquired during the course of treatment for other conditions in a health care setting, or as a community-acquired infection [13]. In-hospital mortality was defined as death due to any cause during hospitalization.

Candida isolates

Pleural fluid specimens were collected through thoracentesis under aseptic conditions or during tube thoracostomy, and were sent for culture. The specimens were inoculated onto Sabouraud dextrose agar plates (BBL Microbiology Systems, Cockeysville MD) [7]. The identifications of Candida spp. were confirmed with the API 32C system (bioMerieux, Marcy-l'Étoile, France) [14]. However, antifungal susceptibility testing was not done in the study.

Results

During the study period, 167 patients had culture-confirmed empyema thoracis. In 10 (6.0%) of these patients the condition was caused by fungi, including eight (4.8%) cases in which it was caused by Candida spp. and two (1.2%) cases in which it was caused by Aspergillus spp. Among the eight patients with Candida empyema thoracis, C. albicans was the most common species isolated (n=4), followed by C. glabrata (n=3) and C. tropicalis (n=1).

The clinical characteristics of the eight patients with Candida empyema thoracis are summarized in Table 1. Five of the eight patients were male (62.5%), and ages ranged from 47 to 82 years. Malignancy, and especially esophageal cancer, was the most common underlying disease in the patients, followed by diabetes mellitus. Fever was present in all of the patients, and dyspnea was noted in five patients. Four patients also had a cough and chest pain. The diagnosis of Candida empyema thoracis for all eight of the patients was made by a positive culture result for Candida spp. The period between the onset of symptoms and diagnosis ranged from 7–20 d (median, 16 d). In seven of eight cases (87.5%) the condition was classified as a hospital-acquired infection (HAI). One patient had empyema associated with both Candida spp. and Pseudomonas aeruginosa. None of the eight patients with Candida empyema thoracis had concomitant candidemia.

Table 1.

Clinical Manifestations of Candida Empyema Thoracis in Study Population

Case (year)	Age	Sex	Underlying condition	Candida species	Community-acquired infections	Antifungal agent	Drainage	Surgery	Outcome
1 (2010)	80	F	DM, CKD, CAD	C. albicans	Yes	Fluconazole	Pigtail catheter	No	Improvement
2 (2011)	49	M	Esophageal cancer, receiving chemotherapy	C. albicans	No	Fluconazole	Chest tube	VATS	Improvement
3 (2011)	77	M	Acute myelocytic leukemia, receiving chemotherapy, DM	C. glabrata	No	Fluconazole→caspofungin	Chest tube	No	Died
4 (2011)	53	M	Esophageal cancer, receiving chemotherapy	C. glabrata	No	Fluconazole→caspofungin	Chest tube	No	Died
5 (2011)	47	M	Lung cancer	C. tropicalis	No	Fluconazole	Chest tube	No	Died
6 (2012)	77	F	MUO, receiving chemotherapy	C. glabrata	No	Nil	Pigtail catheter	No	Died
7 (2012)	82	F	Esophageal rupture with mediastinitis, DM	C. albicans	No	Fluconazole→micafungin	Pigtail catheter	No	Died
8 (2012)	57	M	Esophageal cancer, receiving chemotherapy	C. albicans	No	Fluconazole	Chest tube	No	Improvement

CAD=coronary artery disease; CKD=chronic kidney disease; DM=diabetes mellitus; MUO=malignant disease of unknown origin; VATS=video-assisted thoracic surgery.

Analysis of pleural effusion fluid showed elevated concentrations of lactate dehydrogenase (LDH) in six patients, frank pus in four patients, and a low blood glucose concentration (<40 mg/dL) in three patients. Radiographic studies showed that most (n=6) of the patients' pleural effusions developed on the right side. Neither chest radiography nor computed tomography showed a cavity or mass lesion in any of the patients. All of the patients had elevated C-reactive protein concentrations (>6 mg/L). The most commonly used antifungal agent in the patients' treatment was fluconazole, and all eight of the patients had drainage of their pleural effusions. Five of the patients had chest tube thoracostomies and three patients had drainage with a pigtail catheter. One patient had video-assisted thoracic surgery to manage empyema thoracis. Overall, the patients' hospital stays ranged from 23–56 d (median, 44 d), and the in-hospital mortality was 62.5% (five patients).

Discussion

The present study focused on Candida spp.-associated empyema thoracis, and made several major findings. First, although Candida spp. cause empyema thoracis uncommonly, about 5% of the cases of empyema in our review of cases for the present study were caused by Candida spp. A recent study of 58 patients with complicated parapneumonic pleural effusion and empyema in a medical intensive care unit (ICU) identified two cases of Candida spp.-associated empyema, and an overall prevalence of this condition of 2.6% [8]. Additionally, Brozek et al. reported finding Candida spp. in 9.3% (five patients) of 54 patients with microbiology-confirmed pneumonia in an ICU [15]. All of these findings indicate that Candida spp. should be considered as possible pathogens in empyema thoracis.

Furthermore, among the eight cases of Candida empyema thoracis in our study, C. albicans was the most common pathogen, followed by C. glabrata and C. tropicalis. This epidemiologic finding was similar to that in previous studies [8,9]. In one study of 73 clinical fungal isolates from pleural effusions, C. albicans was the most frequent isolate (n=28, 38%), followed by C. glabrata (n=13, 18%) and C. tropicalis (n=13, 18%) [7]. Another study, of five patients with Candida empyema thoracis, found that four cases were caused by C. albicans and one by C. glabrata [9]. In the present work, we found one patient (Case 2) with a concomitant P. aeruginosa infection. This finding is consistent with that in a study by Ko et al. in which P. aeruginosa was the most common bacterial isolate in cases of fungal empyema, and in which other bacterial pathogens, such as Enterococcus species and Staphylococcus species, were reported as co-pathogens [7].

Beyond this, we found that all of the patients with Candida spp.-associated empyema in our study had various underlying conditions, including active cancer or diabetes mellitus, or were undergoing chemotherapy. These states of immune compromise may predispose to fungal infections [16]. Additionally, one patient with esophageal rupture and mediastinitis developed C. albicans empyema, which is consistent with the finding by Ishiguro et al. of Candida spp. as the cause of five of seven cases of empyema in patients with esophago- or gastropleural fistulas [9]. Our findings therefore suggest that Candida empyema thoracis may develop in immunocompromised patients or patients with gastrointestinal pathology.

Moreover, although most of the cases of Candida empyema thoracis in the present work were classified as HAIs, as in a previous study [8] in which 84% of cases of fungal empyema thoracis were HAIs, one patient in our study, who had diabetes mellitus (Case 1), had a community-acquired empyema caused by C. albicans, and Chuang et al. reported one fatal case of community-acquired pyopneumothorax caused by C. albicans in a 62-year-old male patient with a history of mitral and aerotic stenosis following valve replacements [11]. This suggests that thoracic infections caused by Candida spp. can occur as both HAIs and community-acquired infections.

Lastly, as in a previous study which reported that the mortality in Candida empyema thoracis may exceed 70% [7], the overall in-hospital mortality in our study was 63%. Furthermore, all three of the patients with C. glabrata empyema in our study died despite adequate drainage. This may have been attributable to the initial use of fluconazole as an inappropriate antifungal agent for C. glabrata, which is the species of Candida with the least susceptibility to this drug [3]. However, more evidence is needed to confirm this. Another issue of possible concern is the association between the method of drainage and outcome of Candida empyema thoracis. However, only one patient in the present study (Case 2) underwent video-assisted thoracic surgery (VATS) for drainage, and had subsequent clinical improvement. Most of the patients were treated with relatively non-invasive tube thoracostomy or pigtail catheterization for drainage and had variable outcomes. Although our experience with it is limited, we suggest that VATS might be considered the treatment modality of choice for managing Candida empyema thoracis, although a further, large-scale study is warranted before solidifying this conclusion.

There were several limitations to our study. First, it was conducted in a regional hospital in southern Taiwan, and conditions in that setting may not apply to locations other than southern Taiwan. Second, we did not conduct antifungal susceptibility testing to assess the drug resistance of the pathogens isolated in our study. Although in vitro susceptibility testing cannot reflect the in vivo activity of an antimicrobial agent, antifungal susceptibility tests are nevertheless necessary in the choice of such an agent and provide useful information about its efficacy. Third, we did not assess the mortality attributable to Candida empyema thoracis. However, we suggest that four of the deaths in our study (Cases 4, 5, 6, and 7) should be strongly associated with “difficult to control” Candida empyema, and that the death of the patient in Case 3 be attributed to bacterial sepsis. As a further limitation, the sample size in our study was too limited to permit solid conclusions to be drawn about its findings. Additional, large-scale study of Candida empyema thoracis is needed to confirm our findings.

In conclusion, our study provides a clinical picture of Candida empyema thoracis. Most cases of the condition in our study developed in immunocompromised patients, and especially cancer patients. The most important aspect of our study is its corroboration of a continuing high mortality in Candida empyema thoracis.

Footnotes

Author Disclosure Statement

The authors declare no conflicts of interest.

References

Almirante

, Rodríguez

, Park

, et al. Epidemiology and predictors of mortality in cases of Candida bloodstream infection: Results from population-based surveillance, Barcelona, Spain, from 2002 to 2003. J Clin Microbiol, 2005; 43:1829–1835.

Asmundsdóttir

, Erlendsdóttir

, Gottfredsson

. Increasing incidence of candidemia: Results from a 20-year nationwide study in Iceland. J Clin Microbiol, 2002; 40:3489–3492.

Lai

, Tan

, Huang

, et al. Current challenges in the management of invasive fungal infections. J Infect Chemother, 2008; 14:77–85.

Ruan

, Lee

, Jerng

, et al. Candida glabrata fungaemia in intensive care units. Clin Microbiol Infect, 2008; 14:136–140.

Ruan

, Hsueh

. Invasive candidiasis: An overview from Taiwan. J Formos Med Assoc, 2009; 108:443–451.

Hung

, Chen

, Chang

, et al. Nosocomial candidemia in a university hospital in Taiwan. J Formos Med Assoc, 1996; 95:19–28.

, Cheng

, Hsueh

, et al. Fungal empyema thoracis: An emerging clinical entity. Chest, 2000; 117:1672–1678.

, Hsu

, Hsia

, et al. The changing pathogens of complicated parapneumonic effusion or empyema in a medical intensive care unit. Intensive Care Med, 2006; 32:570–576.

Ishiguro

, Takayanagi

, Ikeya

, et al. Isolation of Candida species is a important clue for suspecting gastrointestinal tract perforation as a cause of empyema. Intern Med, 2010; 49:1957–1964.

10.

Baradkar

, Mathur

, Kulkarni

, et al. Thoracic empyema due to Candida albicans. Indian J Pathol Microbiol, 2008; 51:286–288.

11.

Chuang

, Yeh

, Ko

, et al. Fatal case of community-acquired empyema thoracis and candidemia caused by Candida albicans. Diagn Microbiol Infect Dis, 2011; 71:156–158.

12.

Light

, Macgregor

, Luchsinger

, et al. Pleural effusions: The diagnostic separation of transudates and exudates. Ann Intern Med, 1972; 77:507–513.

13.

Horan

, Andrus

, Dudec

. CDC/NHSN surveillance definition of health care-associated infection in the acute care setting. Am J Infect Control, 2008; 36:309–332.

14.

Merz

, Roberts

. Detection and recovery of fungi from clinical specimens. In: Murray

, Baron

, Pfaller

, et al, eds. Mannual of clinical microbiology, 6th ed. Washington, DC. American Society for Microbiology. 1995:709–722.

15.

Brozek

, McDonald

, Clarke

, et al. Pneumonia observational incidence and treatment: A multidisciplinary process improvement study. Am J Crit Care, 2007; 16:213–219.

16.

Davis

. Fungal pneumonia. Med Clin North Am, 1994; 78:1049–1065.