Pleural effusion as the initial clinical presentation in disseminated cryptococcosis and fungaemia: an unusual manifestation and a literature review

C.neoformans is a nonmycelial, budding, encapsulated, opportunistic, pathogenic, and yeast-like fungus. Inhalation of either the yeast form or basidiospores from an environmental source, such as bird droppings or soil, is believed to cause infection with this opportunistic fungus [8]. The organism may remain confined to the lungs but also frequently disseminates to the CNS, along with the bones, skin, and other parts of the body, depending on the host’s immune status [9]. The most commonly recognized form of human infection due to C. neoformans is meningitis, but the fungus typically establishes infection throughout the respiratory tract [10].

The clinical presentation of cryptococcal disease is highly variable and occasionally consists of atypical manifestations. Cryptococcal disease can present as a cough, sputum, shortness of breath, mild fever, headache, chest pain and other symptoms, whereas nearly one-third of pulmonary cryptococcosis cases were discovered by chest X-ray due to a lack of symptoms. Disseminated disease can occur in any organ, but the CNS is the most vulnerable. CNS disease specifically presents with headache, dizziness and other neurological symptoms.

Reeves first described pleural effusion as a manifestation of cryptococcosis in 1941 [11]. Newman et al. reported the first case of cryptococcal pleural effusion in association with acquired immune deficiency syndrome (AIDS) [3]. With a declining incidence of AIDS-related cryptococcosis in the highly active antiretroviral therapy (HAART) era and with increasing use of immunosuppressives worldwide, non-human immunodeficiency virus (HIV)-infected individuals (with or without predisposing factors) may become the predominant group affected by cryptococcosis [12]. In Rozenbaum’s research, only one patient manifested with pleural effusion among 11 patients who had cryptococcosis along with other diseases or the use of immunosuppressive drugs [13]. According to a review by Chao-Chuang Chou, pleural effusion presented in only one patient among 14 patients studied in Taiwan [14]. Here, we report the first case of pleural effusion as the initial clinical presentation in a renal transplant recipient who was diagnosed with fungaemia and disseminated cryptococcosis. By reviewing and summarizing the characteristics and oversights of the clinical course, we hope to contribute to understanding of the disease and to aid in reducing misdiagnoses. Early targeted treatment for cryptococcal disease and intensive care may ultimately improve poor prognoses.

Pleural effusion has rarely been reported as the initial clinical presentation in cryptococcosis. To our knowledge, only 4 cases have been reported in the English-language literature to date [4‐7]. Cryptococcosis is difficult to diagnose and may be overlooked, especially when pleural effusion is the only clinical manifestation.

In the case described here, pathology consultation was used to identify the aetiology of the pleural effusion. Additional analyses via PAS and methenamine silver staining revealed numerous Cryptococcus organisms in pleural sections, providing reliable evidence of cryptococcal pleuritis. Regrettably, after reviewing previous HE staining, we found Cryptococcus-like organisms (unstained capsules) that had been overlooked in the initial examination. In this case, the atypical manifestation of the pleural effusion, along with the macroscopic observation and pathologiacl study, caused the early-stage misdiagnosis. In reference to cases reported in the literature, multiple plaques and nodules have been found on the pleural surface in most cases, and pathological findings of cryptococcal pleuritis have shown multiple yeast-like fungal bodies in the inflammatory granuloma [15].

Pleural effusion cultures for Cryptococcus are usually negative, which make diagnosis even more difficult. Young et al. [7] reviewed 26 cases of cryptococcal pleural effusions based in English-language literature, and only 12 patients (42 %) had cultures positive for pleural effusion. The negative culture results may have been due to the small number of fungi in the pleural fluid. In Salyer’s opinion, Cryptococcus can survive in the pleura in the form of a granuloma that releases a large number of antigens into pleural effusion, which is consistent with his research [16]. Similar results in our case support this theory, as the pleural effusion cultures were negative, but the serum was positive for cryptococcal antigens. It has been postulated that the release of antigen, rather than organism growth, is responsible for pleural manifestations. In particular, various clinical manifestations (pleural effusions and others) have been described in HIV-infected patients due to acquired immune deficiency [17]. In the present case, pleural effusion was observed as an unusual initial clinical presentation in a transplant recipient with cryptococcosis. This manifestation is not typical of cryptococcal infection but probably occurred because our patient was on immunosuppressive therapy. His total CD4+ cell count was about 200/ul, less than HIV-infected patients whose CD4+ cell count were about 250/μl.

A cryptococcosis diagnosis is usually made by demonstration of the presence of the organism using histopathology, culture or detection of cryptococcal antigens in the blood and CSF, and by CrAg testing of the pleural fluid. Detection of the capsular antigen is a reliable diagnostic method, and identifying Cryptococcus via histopathological analysis is the gold standard for confirming cryptococcosis. Organisms, and especially yeasts, which are only weakly stained with HE stain, may be difficult to find in routine sections. Gomori-methenamine silver stains is a characteristic black colour, and PAS stains the yeast wall a red colour. According to the literature, the detection rates of C. neoformans with these two stains are 100 % [18]. India ink staining is the traditional method for identifying Cryptococcus by delineating the refractile mucinous capsule surrounding the organism, giving the organism a characteristic “starry night” appearance. However, the sensitivity and specificity of India ink staining can be highly variable and are often operator-dependent.

Approximately 8 % of invasive fungal infections in solid-organ transplant (SOT) recipients are due to cryptococcosis [19]. The overall incidence of cryptococcosis in SOT recipients is approximately 2.8 % and ranges from 0.3 to 5 %, with a mortality rate of 20 to 40 % [20, 21]. Once the CNS is involved, the mortality rate is as high as 50 %. Early diagnosis and treatment are crucial to improve the survival of patients with cryptococcosis, and especially disseminated cryptococcosis. In the case described here, the patient received amphotericin B liposome combined with 5-flucytosine and voriconazole for the first 11 days, then amphotericin B liposome combined with 5-flucytosine sustained to 8 weeks, after which this regimen was changed to fluconazole for maintenance. The patient was still on treatment and being followed up at the time that this report was written.

The current guidelines from the Infectious Diseases Society of America (IDSA) recommend treating SOT recipients with disseminated, CNS, or severe pulmonary cryptococcosis with at least 2 weeks of induction therapy with a combination of amphotericin B lipid complex (ABLC) and 5-flucytosine [22]. This regimen slightly differs from Chinese expert consensus on the strategies for treating cryptococcosis. The recommendation from Chinese experts for the management of CNS and pulmonary cryptococcosis in HIV-negative patients is induction therapy with ABLC combined with 5-flucytosine for a minimum of 8 weeks. This difference might be partly attributable to disparities between gradually increasing the dose up to effective dose (which takes time) and administrating lower maintaining doses of ABLC in Chinese [23].

Voriconazole, which is a synthetic derivative of fluconazole, is a relatively new broad-spectrum triazole with excellent in vitro activity against C. neoformans (voriconazole exhibits superior in vitro activity compared with fluconazole), good penetration of the blood–brain barrier, and extended activity against moulds and yeasts that were not susceptible to earlier azole-type drugs [24‐28]. In fact, several case studies have reported the successful treatment of cryptococcal meningitis with voriconazole [26‐28]. The IDSA prefers to recommend amphotericin B deoxycholate (AmBd) plus flucytosine for induction therapy, rather than triple therapy (amphotericin B, flucytosine, and fluconazole), based on data from a randomized trial by Brouwer et al. [22]. Chinese experts suggest that triple therapy (AmBd, flucytosine and voriconazole) could be considered in cryptococcal meningitis patients with AIDS [23]. We chose triple therapy (ABLC, flucytosine, and voriconazole) for the first 2 weeks in this patient. Because it usually cost nearly 2 weeks for gradually increasing the dose up to the effective dose of ABLC, voriconazole could be an important complementary drug during this period. But the patient’s liver function turned bad during triple therapy, we have to stop the voriconazole on the 12th day of antifungal therapy, triple therapy only sustained 11 days, then amphotericin B liposome combined with 5-flucytosine sustained to 8 weeks. As we know, cryptococcal meningitis is fatal, even with specific treatment, so helping patients to survive an acute attack and stabilizing their condition for further consolidation therapy are vital. In this case, we chose triple therapy (ABLC, flucytosine, and voriconazole) for the first stage of induction therapy, with a composite evaluation of safety and efficacy and achievement of a good curative effect. Reassuringly, the patient’s temperature returned to normal, his pleural effusion was absorbed, and his general condition was good. His cerebrospinal pressure decreased to normal, and Cryptococcus were no longer found in the CSF.

C. neoformans-associated immune reconstitution inflammatory syndrome (c-IRIS) was initially observed in HIV-infected patients receiving HAART along with rapid reversal of immune deficiency (rapid lymphocyte CD4+ cell count recovery) [29]. The biologic basis of c-IRIS in SOT recipients is believed to be reversal from a Th2 (anti-inflammatory) to a Th1 (pro-inflammatory) cytokine profile [30‐32]. The clinical presentation of c-IRIS may be similar to that of C. neoformans meningitis. However, c-IRIS has to be differentiated from progression of the cryptococcal disease, although both phenomena can appear simultaneously [22].

The rhG-CSF is known primarily as a haematopoietic growth factor that stimulates the production and release of neutrophilic granulocytes from the bone marrow to limit infection [33]. In our case, this immunosuppressed patient was only treated with rhG-CSF once on day 8. The patient’s condition deteriorated on day 32. rhG-CSF is used to restore neutrophil counts and to prime neutrophil functions, but not lymphocyte counts and functions, to protect vulnerable patients from infection in the clinic [34]. Only one researcher “Hartung” reported continuous rhG-CSF administration could increases monocyte and lymphocyte production, intermittent rhG-CSF treatment could not induce lymphocytosis in volunteers [35, 36]. In this case, followed up of Lymphocytes(/μl) count showed no increasing of Lymphocytes and CD4+ cell after usage of rhG-CSF. The feature of c-IRIS in patients with cryptococcosis is CSF negative fungal culture [1]. In this case, the finding of a positive fungal culture excluded c-IRIS and supported the diagnosis of progression of the cryptococcal disease. Therefore, c-IRIS should not be suspected because of using rhG-CSF in this case. The relationships between rhG-CSF therapy and c-IRIS will need to be further scrutinized.

Cryptococcal infection must be considered in patients with pleural effusion, and especially immunosuppressed patients, even if culture of the pleural fluid is negative. Detection of cryptococcal capsular antigen in the serum and the CSF will be useful in this context. Employing a variety of reliable diagnostic tools, such as a bronchoscopy with bronchoalveolar lavage, video-assisted thoracoscopy, transbronchial biopsy, and exploratory thoracotomy to acquire specimens in the early stages, are also helpful for a definitive diagnosis. Careful review of multiple stained biopsy sections and close communication between the pathologist and the clinician are both important as well.