BRIEF REPORT

Continuous Infusion of Amphotericin B Deoxycholate in the Treatment of Cryptococcal Meningoencephalitis: Analysis of Safety and Fungicidal Activity
Diego Rodrigues Falci, Luciano Weber Lunardi, Carina Guedes Ramos, Mo nica Baumgardt Bay, Vale rio Rodrigues Aquino, and Luciano Zubaran Goldani Infectious Diseases Unit, Hospital de Clinicas de Porto Alegre, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil

We measured fungicidal activity of continuous infusion of amphotericin B deoxycholate plus 5 ucytosine using quantitative cultures of cerebrospinal uid (CSF) obtained from lumbar punctures of human immunodeciency virus (HIV) infected patients with neurocryptococcosis during 14 days of treatment. Glomerular renal function was preserved in all patients. Mycological efcacy with progressive reduction in CSF cryptococcal colony-forming units was comparable to standard 4-h infusion of amphotericin B. The mainstay of treatment of cryptococcosis in human immunodeciency virus (HIV)infected patients remains amphotericin B deoxycholate, according to current antifungal treatment guidelines. The administration of intravenous amphotericin B is associated with nephrotoxicity and electrolyte abnormalities, along with other adverse effects, such as anemia [1, 2]. An innovative form of administration of amphotericin B deoxycholate (namely, continuous infusion) was proposed to reduce the toxicity of the drug while maintaining good antifungal efcacy. Continuous infusion of amphotericin B deoxycholate has been studied for the empirical treatment of neutropenic fever. However, all such trials have included patients with probable or unproven fungal infection, and concerns re-

Received 17 September 2009; accepted 10 November 2009; electronically published 2 February 2010. Reprints or correspondence: Dr Luciano Goldani, Section of Infectious Diseases, Hospital de Clinicas de Porto Alegre, Ramiro Barcelos 2350, Porto Alegre, RS Brazil 90640-002 (lgoldani@ufrgs.br). Clinical Infectious Diseases 2010; 50:e26e29 2010 by the Infectious Diseases Society of America. All rights reserved. 1058-4838/2010/5005-00E1$15.00 DOI: 10.1086/650489

garding efcacy have therefore remained [36]. Here, we report our experience in a pilot, noncomparative clinical trial with continuous infusion of amphotericin B deoxycholate plus oral 5 ucytosine in the treatment of HIV-associated cryptococcal meningoencephalitis. Patients and methods. The study was approved by our local ethics committee and conducted from September 2007 through May 2009 at a 600-bed university hospital in southern Brazil. We considered as eligible for the study all consecutive patients with positive India ink in direct examination of cerebrospinal uid (CSF) or positive CSF cryptococcal antigen test results, plus serological test results positive for HIV. Inclusion of the patient was further conrmed by grown of Cryptococcus neoformans in media culture of CSF samples. Exclusion criteria were baseline serum creatinine concentration 1300 mmol/L, an absolute neutrophil count ! 500 10 6 cells/L, a platelet count ! 50,000 10 6 cells/L, aminotransferases levels 15 times the upper limit of normal values (1200 IU/L), pregnancy or breast feeding, previous severe reactions to amphotericin or 5 ucytosine use, or any amphotericin B use during the previous 30 days. All patients in this pilot single-arm study received amphotericin B deoxycholate and 5 ucytosine. After 2 weeks of therapy, patients were switched to uconazole. Subjects received continuous (24-h) infusion of amphotericin B deoxycholate (0.7 mg/kg per day), given in 500-mL doses with 5% glucose. Patients also received oral 5 ucytosine (25 mg/kg 4 times per day). Both drugs were administered during a 14-day induction phase. Unless contraindicated, patients received 1 L of normal saline (0.9% solution) daily. Potassium and magnesium supplementation were provided as necessary, and the use of premedication was permitted by the protocol. Lumbar punctures were performed at baseline and on days 3, 7, and 14 of treatment. Patients with CSF opening pressure 135 cm H2O or symptoms attributable to elevated intracranial pressure underwent additional lumbar punctures. The use of antiretroviral therapy was allowed if indicated by the assistant physician. We prospectively evaluated patients for adverse effects and clinical and microbiological efcacy during a 14-day period. Renal function, blood cell count, and electrolytes were checked at baseline and every 3 days. Fungicidal activity was measured using serial quantitative cultures of CSF obtained from lumbar punctures. Quantitative fungal culture was performed as described elsewhere, and the number of cryptococcal colonyforming units (CFUs) was obtained at each treatment [7]. Clearance of Cryptococcus was calculated with the slope of linear

Downloaded from http://cid.oxfordjournals.org/ by guest on November 27, 2012

e26 CID 2010:50 (1 March) BRIEF REPORT

Table 1. Baseline Characteristics and Clinical Outcome of Patients with Cryptococcal Meningoencephalitis Treated with Continuous Amphotericin B Deoxycholate Infusion
Variable Age, mean years (range) Female sex, no.(%) of patients Baseline laboratory parameters, median value (range) Absolute neutrophil count, neutrophils/mm3 Hemoglobin level, g/dL Platelet count, 106 platelets/L Creatinine concentration, mg/dL Potassium level, mEq/L Sodium level, mEq/L Magnesium level, mEq/L CD4 cell count, cells/mL Death, no. (%) of patients By week 2 By week 12 Patients 37.5 (2052) 1 (10) 3095 (14205580) 11.35 (8.714.4) 173,500 (97,000216,000) 0.61 (0.401.16) 3.75 (3.204.60) 134 (129144) 1.80 (1.52.3) 41 (10255)

Downloaded from http://cid.oxfordjournals.org/ by guest on November 27, 2012

1 (10) 1 (10)

regression of log CFU counts against time for each patient, giving the rate of decrease in log CFU/mL of CSF per day. Negative culture results were assigned a value of 1 CFU/mL (log 0) [7]. We also performed polyssaccharide capsular cryptococcal antigen testing in serum and CSF samples. Azotemia was dened as an increase in serum creatinine concentration to 2.5 mg/dL; in this case, the switch of therapy to uconazole or another formulation of amphotericin B could be done by the assistant physician, and the patient was considered to have experienced treatment failure as a result of toxicity. We dened hypokalemia as a value !3.0 mEq/L and hypomagnesemia as a value of !1.5 mEq/L. We also evaluated mortality and measured subject survival at 14 days and at 12 weeks after enrollment. Results were presented as mean values (range) or as otherwise specied. A linear regression model was made for each patient to calculate decrease in CFU count. Continuous variables between before and after treatment were compared with use of the Students t test. Statistical analysis was performed with SPSS for Windows, version 16.0 (SPSS). Results. During the study period, 12 consecutive patients with HIV disease and cryptococcal meningoencephalitis were considered to be eligible for the study. Two patients did not give their consent. Therefore, we enrolled 10 patients in the study. One patient died during the study. This patient presented with untreatable intracranial hypertension and severe cerebral edema, which was the probable cause of death. The patient received only 3 days of amphotericin B plus 5 ucytosine treatment and had not experienced any adverse events associated with these drugs that could be linked to the fatal outcome. An autopsy was not performed. The mean age of this patient sample was 37 years. There

were 9 male patients and 1 female patient in this sample. All of patients were severely immunocompromised and had advanced HIV disease. Baseline characteristics and clinical outcomes of the patients are described in Table 1. At baseline, a high CSF fungal burden was present, expressed in the elevated cryptococcal CFU counts. Four patients had baseline CSF counts 1100,000 CFU/mL. The range of baseline counts was 15,00080,000 CFU/mL for other patients. In all patients who entered the mycological analysis, we observed a progressive reduction in CSF cryptococcal CFUs, with sterilization of CSF at 14 days of treatment. An exponential reduction in CFU counts was observed, as seen in Figure 1. Early fungicidal activity (EFA) or fungal clearance was calculated for each patient, as described above. The mean EFA ( standard deviation [SD]) was 0.37 0.05 log CFU/mL of CSF per day. Cryptococcal latex agglutination test results were positive in all patients, but titers did not correlate with reduction in CFU counts (data not shown). Mortality rate during the study was 10% (1 of 10 patients died) at 2 weeks; it was also 10% (1 of 10 patients died) at 12 weeks. As shown in Table 2, glomerular function was well-preserved in all patients, with creatinine serum levels below 1.5 mg/dL at the end of 14 days of amphotericin B deoxycholate therapy. The mean serum creatinine level increased from 0.70 mg/dL (range, 0.401.16 mg/dL) to 0.86 mg/dL (range, 0.491.12 mg/ dL). However, we observed a mean increase of 50% in creatinine concentration during treatment. Indeed, 3 patients had increases 150% at the end of treatment, and 1 patient had a maximum increase of 150%, which was partially reversed after cessation of amphotericin B therapy. Comparison between baseline values and values after 14 days of treatment demonBRIEF REPORT CID 2010:50 (1 March) e27

Downloaded from http://cid.oxfordjournals.org/ by guest on November 27, 2012

Figure 1. Cryptococcal colony-forming units (CFUs) in cerebrospinal uid (CFU), at baseline and days (d) 3, 7, and 14 of treatment with continuous infusion of amphotericin B deoxycholate plus 5 ucytosine. Patient 10 was excluded from mycological analysis. Early fungicidal activity was 0.37 0.05 log CFU/mL of CSF per day. Pt, patient.

strated that the change observed in mean creatinine concentration was not statistically signicant. No patient required dialysis. Every patient but 1 (the patient who died) received 14 days of amphotericin B deoxycholate plus 5 ucytosine as planned in the protocol. No severe infusion-related adverse effects were observed. The treatment was not changed to uconazole or another formulation of amphotericin B during the rst 14 days of the induction phase. Hypokalemia was observed

in 2 patients. Mean serum potassium concentrations decreased from 3.88 to 3.63 mEq/L. We observed no signs of cardiac arrhythmia. Hypomagnesemia was observed in 2 patients. The mean magnesium serum level decreased from 1.93 to 1.7 mEq/ L. Differences in potassium and magnesium concentrations before and after treatment were not statistically signicant. Anemia was a common observation. Mean hemoglobin levels decreased from 11.49 to 7.92 g/dL; 5 patients had decreases of at

Table 2. Laboratory Findings for Patients with Cryptococcal Meningoencephalitis Treated with Continuous Infusion of Amphotericin B Deoxycholate plus 5 ucytosine
Variable Creatinine concentration, mg/dL Potassium level, mEq/L Magnesium level, mEq/L Hemoglobin level, g/dL Baseline 0.70 0.09 3.88 0.14 1.94 0.12 11.49 0.6 Day 3 0.92 0.11 3.66 0.26 1.9 0.09 10.29 0.62 Day 6 0.99 0.11 4.14 0.36 1.9 0.11 8.69 0.49 Day 9 0.98 0.10 3.72 0.30 1.86 0.11 Day 12 0.87 0.07 3.82 0.25 1.64 0.09 Day 15 0.86 0.08 3.63 0.22 1.7 0.06 7.92 0.41

NOTE. Data are mean standard error.

e28 CID 2010:50 (1 March) BRIEF REPORT

least 3 g/dL. In 1 patient, blood transfusion was required. Comparison of mean hemoglobin concentrations before and after 14 days of treatment demonstrated a statistically signicant decrease (P p .001). Discussion. Our study conrms the fungicidal activity of amphotericin B deoxycholate plus 5 ucytosine when administered via continuous infusion. The mycological response observed was remarkable. We have demonstrated microbiological efcacy with this regimen, using an established method to assess the rate of reduction in the fungal burden of disease [7]. All patients achieved sterilization of CSF with 14 days of treatment, as expected with administration of a regimen of conventional amphotericin B plus 5 ucytosine [8]. The clearance of infection, measured as EFA, was similar to that found in other studies that have used the same technique [9]. This end point is suitable for evaluation of new treatment regimens for this disease, as recently reported, and our results suggest that this new form of administration can be an effective alternative for clinicians in settings where lipid formulations of amphotericin B are unavailable or unaffordable [10]. Decreased renal blood ow and increased tubular membrane permeability are believed to be the causes of amphotericin B deoxycholate nephrotoxicity, with incidence rates that may be as high as 50% [11]. Dose-dependent alterations in protein binding and distribution of amphotericin B deoxycholate may explain why nephrotoxicity has been less frequently observed among patients who received amphotericin B by continuous infusion [4, 6]. In our study, the rates of nephrotoxicity were comparable to those reported by previous studies with continuous infusion of amphotericin B deoxcyholate and were also similar to those reported in lipid formulation trials [11, 12]. Hypokalemia was frequently observed, as we expected. Continuous infusion can protect glomerular function but seems to have little effect with respect to tubular damage. Anemia was a very common phenomenon and was attributed to suppression of erythropoietin production by amphotericin B. The safety of amphotericin B deoxycholate when administered via continuous infusion provides a new and promising strategy for the treatment of systemic mycoses. Our results suggest that continuous infusion could reduce nephrotoxicity

while retaining fungicidal activity against C. neoformans. Larger and comparative trials are necessary to further evaluate this treatment regimen.
Acknowledgments
We thank the Brazilian Council of Research and Research Fund from Hospital de Cl nicas de Porto Alegre for partial support of this study. Potential conicts of interests. All authors: no conicts.

References
1. 2. Jarvis JN, Harrison TS. HIV-associated cryptococcal meningitis. AIDS 2007; 21:21192129. Pappalardo MC, Melhem MS. Cryptococcosis: a review of the Brazilian experience for the disease. Rev Inst Med Trop Sao Paulo 2003; 45: 299305. Peleg AY, Woods ML. Continuous and 4 h infusion of amphotericin B: a comparative study involving high-risk haematology patients. J Antimicrob Chemother 2004; 54:803808. Eriksson U, Seifert B, Schaffner A. Comparison of effects of amphotericin B deoxycholate infused over 4 or 24 hours: randomised controlled trial. BMJ 2001; 322: 579582. Imhof A, Walter RB, Schaffner A. Continuous infusion of escalated doses of amphotericin B deoxycholate: an open-label observational study. Clin Infect Dis 2003; 36:943951. Lewis RE, Wiederhold NP. The solubility ceiling: a rationale for continuous infusion amphotericin B therapy? Clin Infect Dis 2003; 37: 871872. Brouwer AE, Rajanuwong A, Chierakul W, et al. Combination antifungal therapies for HIV-associated cryptococcal meningitis: a randomised trial. Lancet 2004; 363:17641767. van der Horst CM, Saag MS, Cloud GA, et al. Treatment of cryptococcal meningitis associated with the acquired immunodeciency syndrome. National Institute of Allergy and Infectious Diseases Mycoses Study Group and AIDS Clinical Trials Group. N Engl J Med 1997; 337:1521. Brouwer AE, Teparrukkul P, Pinpraphaporn S, et al. Baseline correlation and comparative kinetics of cerebrospinal uid colony-forming unit counts and antigen titers in cryptococcal meningitis. J Infect Dis 2005; 192:681684. Bicanic T, Muzoora C, Brouwer AE, et al. Independent association between rate of clearance of infection and clinical outcome of HIVassociated cryptococcal meningitis: analysis of a combined cohort of 262 patients. Clin Infect Dis 2009; 49:702709. Walsh TJ, Finberg RW, Arndt C, et al. Liposomal amphotericin B for empirical therapy in patients with persistent fever and neutropenia. National Institute of Allergy and Infectious Diseases Mycoses Study Group. N Engl J Med 1999; 340:764771. Deray G. Amphotericin B nephrotoxicity. J Antimicrob Chemother 2002; 49(Suppl 1):3741.

3.

Downloaded from http://cid.oxfordjournals.org/ by guest on November 27, 2012

4.

5.

6.

7.

8.

9.

10.

11.

12.

BRIEF REPORT CID 2010:50 (1 March) e29