Cryptococcemia According to Immune Status: An Analysis of 65 Critical Cases

C. neoformans was isolated from the blood culture of patients (January 2010 to December 2019) at the Xiangya Hospital of Central South University (Changsha, Hunan, China). The inclusion criteria was positive blood culture of C. neoformans. The hospital is a tertiary medical center with a capacity of 3500 beds that provides both primary and tertiary referral care, and includes a bone marrow and solid organ transplantation center. Before 2015, the isolated strains of C. neoformans were cultured and identified in our hospital using India ink stain. After 2015, the fungi strain was identified by mass spectroscopy. Serum cryptococcal antigen was not detected in our hospital. The automatic fungal BD BACTEC™ Instrumented Blood Culture System (Becton Dickinson Diagnostic Instrument Systems, Sparks, MD, USA) was applied to facilitate rapid reporting. All blood samples obtained from patients with suspected fungal infection were incubated routinely for 15 days before ‘no growth’ was reported. The list of patients among whom C. neoformans had been isolated from the blood was obtained from the computer database of the clinical microorganism isolation system. The study protocol was approved by the Ethics Committee of Xiangya Hospital (no. 2018091080). Data of the patients submitted were anonymized.

The medical records of the patients were obtained from the database of Xiangya Hospital and reviewed carefully. We also collected data regarding demographics and clinical manifestations, such as age, sex, duration of hospitalization, predisposing conditions, significant symptoms, physical and laboratory findings, antifungal treatment, and clinical outcomes.

The diagnosis of cryptococcemia was confirmed by two specialists based on the following criteria: (1) the patients demonstrated symptoms and signs of infection, and (2) their blood culture yielded positive results of infection at least once. Accordingly, the patients were divided into two subgroups (the immunosuppressed and immunocompetent groups). As per the definition of pneumonia (ventilator-associated pneumonia and non-ventilator-associated pneumonia) put out by the United States Centers for Disease Control, immunosuppressed patients represented those (1) with neutropenia, i.e., an absolute neutrophil count or total white blood cell count (WBC) < 500/mm³, (2) with leukemia, lymphoma, or acquired immune deficiency syndrome (AIDS) with CD4 count < 200 cells/mm³, (3) who underwent splenectomy, (4) with a history of solid organ or hematopoietic stem cell transplant, (5) who underwent cytotoxic chemotherapy, and (6) who took steroids (excluding inhaled steroids) daily for > 2 weeks. The immunocompetent group included subjects with no underlying diseases or who had not been properly diagnosed with the disease. Sepsis and septic shock were defined according to the criteria of The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) [7]. The clinical outcomes of the patients were considered after they were admitted to the hospital (day 1). Signs of improvement, deterioration, and cause of death were recorded until their 60th day of hospitalization. Improvement was assessed in terms of a favorable clinical response to therapy at the end of hospitalization. Antifungal treatment was analyzed using a one-dose-received approach (the initial dose and treatment after admission), and empirical treatment was evaluated based on the antifungal treatment administered before positive culture results were obtained (except for antifungal prophylaxis). According to European guidelines for primary antifungal prophylaxis in adult hematology patients, azoles are considered the first choice for primary antifungal prophylaxis for patients receiving intensive remission induction chemotherapy for AML or MDS. Patients with lymphoma tend to be at low risk of IFD. Therefore, no recommended antifungal protocol was mentioned in this guideline [8].

Differences in qualitative variables between the two groups were analyzed using the chi-squared test with continuity correction or Fisher’s exact test, as appropriate. Quantitative data were compared using the t test. Multivariate analysis was not performed as the patient pool was not large enough to yield reliable results. All statistical analyses were performed using SPSS v.19 (IBM, Armonk, NY, USA). Analysis items with p < 0.05 were considered statistically significant.

In total, 65 patients with cryptococcemia were diagnosed at Xiangya Hospital from January 2010 through December 2019. Eighty-two percent of the patients were reported to have underlying immunosuppressive conditions; the number of patients in this group was 4.5 times higher than that of those in the immunocompetent group. No significant differences in age, gender, or the duration of hospitalization were observed between the two groups (Table 1).

Twelve patients with cryptococcemia did not show evidence of underlying immune suppression. Among the remaining 53 immunosuppressed patients, treatment with immunosuppressive drugs (n  = 21), malignant diseases (n = 10), and AIDS (n = 8) were the three major predisposing conditions. The patients with chronic renal failure, nephritic syndrome, systemic lupus erythematosus, and chronic glomerulonephritis underwent treatment comprising immunosuppressive drugs, such as glucocorticoids and mycophenolate. Four patients had lymphoma and one was diagnosed with glioblastoma. Six subjects were receiving chemotherapy or radiotherapy prior to being diagnosed with cryptococcemia; however, none of the patients who underwent immunosuppressive therapy was a recipient of bone marrow or solid organ transplant in our study. Nevertheless, the distributions of other predisposing conditions, such as tuberculosis, liver cirrhosis, and diabetes, are shown in Table 2.

Fifty-six of the 65 patients admitted reported fever (axillary temperature ≥ 37 °C). Thirty-eight of the 56 patients presented with high fever (axillary temperature ≥ 39 °C). In the immunosuppressed group, 47 patients (88%, 47/53) had fever; however, only 20 patients (38%, 20/53) were critically ill. Sixteen immunosuppressed patients (29%, 16/53) and 8 immunocompetent patients (67%, 8/12) had cryptococcal meningitis, which was confirmed via cerebrospinal fluid culture.

Immunosuppressed patients were more likely to have septic shock at admission than immunocompetent patients (47% vs. 0%, p = 0.003). Several patients had abnormal laboratory results, such as those regarding WBC count, procalcitonin, alanine transaminase, and blood urea nitrogen levels. The clinical manifestations are summarized in Table 3.

In our study, most cases of cryptococcemia involved infections of the CNS (55%, 36/65) and lungs (23%, 15/65), and disseminated infection (45%). Positive results of cerebrospinal fluid culture or India ink stain suggested meningeal involvement. Pulmonary cryptococcosis was confirmed by the fungal culture of endotracheal aspirates, bronchoalveolar lavage, or pleural fluid. Cutaneous cryptococcosis was diagnosed based on skin tissue biopsy. One patient with pre-existing liver cirrhosis developed cryptococcal peritonitis proven on culture, whereas another female patient had C. neoformans isolated from a culture of endometrial tissue. Twenty-nine patients (45%) had disseminated infections and C. neoformans affected two or more organs (Table 4). There was a higher incidence of disseminated infection among immunosuppressed patients than among immunocompetent patients.

Two of four immunosuppressed patients (6%, 4/65) had AIDS, one had lymphoma, and the other had liver cirrhosis. These patients did not receive prior antifungal therapy and died before their blood cultures could be obtained. The remaining 61 patients were treated with early systemic antifungal therapy, which was initiated within 48 h of hospitalization. Thirty patients (46%, 30/65) received amphotericin B (AmB) alone or followed by sequential treatment with other antifungal drugs. Ten patients (15%, 10/65) were treated with AmB in combination with 5-fluorocytosine (5-FC) or fluconazole (FCZ). Three patients received oral FCZ and three were treated empirically with caspofungin alone, as the dosage could not be changed due to their severely impaired liver and renal functions (Table 5).

The clinical outcomes of all 65 patients were recorded on the 60th day after admission. In the immunosuppressed group, there were 28 cases of death (53%, 28/53), 3 cases of deterioration (unfortunately, these patients died soon after admission due to refractory hypotension and multiple organ failure), and 22 cases of clinical improvement (41%, 22/53). The rate of acute mortality (within 14 days after admission) was high among patients in the immunosuppressed group (51%, 27/53). The rates of mortality, improvement, and deterioration were 53%, 41%, and 6%, respectively. The only patient who died in the immunocompetent group was an elderly patient with disseminated cryptococcosis. That patient died soon after admission on account of multiple organ failure. There were significant differences in the rates of mortality and clinical improvement between the two groups of patients with different immune statuses (p < 0.05) (Table 6).