Portal de Periódicos da CAPES

PERSISTENT CANDIDEMIA IN A PREMATURE INFANT TREATED WITH FLUCONAZOLE

1999; Lippincott Williams & Wilkins; Volume: 18; Issue: 8 Linguagem: Inglês

10.1097/00006454-199908000-00021

1532-0987

Lisa-Gaye Robinson, Lucky Jain, Athena P. Kourtis,

Pneumocystis jirovecii pneumonia detection and treatment

Systemic candidiasis in very low birth weight (VLBW) premature infants is a common and potentially serious infectious disease. Conventional therapy for this infection has included prompt and aggressive treatment with amphotericin B. We report on a case of persistent candidemia for 27 days in a VLBW infant that did not respond to a prolonged course of combination therapy with amphotericin B and flucytosine, but rapidly cleared after the addition of fluconazole. Case report. A 680-g male infant was delivered vaginally at 25 weeks of gestation. The baby was intubated at birth, blood cultures were obtained and the baby was given empirically ampicillin and gentamicin. On the fourth day of life he developed grossly bloody nasogastric aspirates, a distended abdomen and acidosis. An abdominal film revealed bowel perforation. He was treated with vancomycin, gentamicin and clindamycin and transferred to our institution for further management. The infant had an ileostomy on the fifth day of life, requiring a reoperation the following day for a persistent surgical site bleed. Hyperalimentation and intravenous lipids were begun on the second post-operative day. On the seventh postoperative day he showed signs of clinical sepsis, at which time amikacin was added empirically to his antibiotic regimen for a possible nosocomially acquired Gram-negative organism. The following day Klebsiella pneumoniae and Staphylococcus epidermidis grew from his blood. All of his catheters were changed at that time. On the ninth postoperative day Candida albicans grew from his blood and amphotericin B (0.5 mg/kg as the initial dose and 1 mg/kg daily thereafter) was added to his regimen. The infant, however, remained fungemic on successive daily cultures. A workup to identify the source of the candidemia included abdominal, heart and brain ultrasound scans which were unrevealing. The infant's intravascular catheters were changed again at that time (Day 7 of fungemia). On the tenth day of fungemia the infant's antifungal therapy was changed to amphotericin B lipid complex (Abelcet, 5 mg/kg/day), because of worsening renal function (serum creatinine increased from 1.0 mg/dl to 1.9 mg/dl). On the 11th day he developed multiple cutaneous abscesses that grew C. albicans. Flucytosine (50 mg/kg/day) was begun via orogastric tube (dose adjusted for the degree of renal function). The cutaneous abscesses were drained daily for 1 week. On the 15th day of fungemia the Abelcet dose was increased to 7.5 mg/kg/day. Three days later the flucytosine dose was also increased to 75 mg/kg/day, as the infant's renal function began to improve (creatinine decreased to 0.7 mg/dl). An echocardiogram demonstrated a possible thrombus or vegetation in the left atrium. Retinal and cerebrospinal fluid examinations as well as brain and abdominal ultrasounds showed no evidence of localized fungal disease. Because of the concern of antifungal resistance, susceptibility tests were performed by the E test (AB Biodisk, Solna, Sweden) at the Emory University Department of Clinical Microbiology. The C. albicans blood isolate from the tenth day of fungemia was susceptible to amphotericin (MIC 0.25, 58 g/ml), fluconazole (MIC 0.38, 58 g/ml), itraconazole (MIC 0.125, 58 g/ml) and flucytosine (MIC 0.023, 58 g/ml). Continued growth of C. albicans for 27 days, despite high dosages of liposomal amphotericin and flucytosine, prompted the start of intravenous fluconazole (6 mg/kg/day for the first dose, then 3 mg/kg/day as daily maintenance). The first negative culture was obtained the following day. A follow-up echocardiogram on Day 28 of therapy demonstrated resolution of the left atrial lesion. He subsequently received flucytosine for 2 more weeks and Abelcet and fluconazole for 4 additional weeks (Fig. 1). Repeated blood cultures were sterile and the infant steadily improved. The ileostomy closed completely at 3.5 months of life, and the infant is currently receiving full gastric feedings.Fig. 1: Summary of therapeutic interventions in a VLBW infant with fungemia. abs, abscess.Discussion. Disseminated systemic candidiasis has become a problem in the neonatal intensive care unit, particularly in the lower birth weight infants. It has been estimated that 3 to 5% of infants weighing <1500 g and up to 10% of infants <1000 g develop systemic fungal infections.1 Past observations have demonstrated these infections to be universally fatal if left untreated. Systemic therapy with amphotericin B has been the treatment of choice2, 3; if the infection is severe or associated with central nervous system disease, addition of flucytosine is recommended. Treatment with these agents has been successful but often associated with adverse effects. The introduction of liposomal amphotericin B, a drug formulation that offers reduced renal toxicity, has been reported to be effective and safe, although appropriate dosages for premature infants have not been established. Several reports from the literature have demonstrated the bis-triazole derivative fluconazole to be a very effective antifungal with an excellent pharmacokinetic profile. It was shown in a randomized prospective trial to be equally as effective as amphotericin for disseminated neonatal fungal infections, although with less side effects.4 A recent report demonstrated successful treatment of 25 episodes of neonatal fungemia with fluconazole monotherapy.5 In our patient amphotericin B therapy was begun immediately after isolating C. albicans from the blood. Persistent fungemia and development of cutaneous lesions, despite removal of all indwelling catheters, was very concerning. Oral flucytosine was added for the possible synergy with amphotericin B, and the infant's renal function was considered in the dosing decision. Serum concentrations of flucytosine in the VLBW neonatal population are extremely variable and therapeutic values have not been established.6 We did not obtain serum fluconazole concentrations for these reasons and closely monitored the renal and hematologic values for signs of toxicity. The left atrial lesion noted on the echocardiogram established a probable source for the continued fungemia that persisted despite the use of Abelcet and flucytosine. Fluconazole was added on the twenty-seventh day of fungemia. Synergy between fluconazole and flucytosine has been shown, however synergy with amphotericin is more controversial.7 There is evidence in the literature suggesting amphotericin B-fluconazole antagonism, in that fluconazole purportedly reduces the fungicidal activity of amphotericin B.8-11 Our patient's fungemia cleared 24 h after the fluconazole was added. It is possible that quantitative blood cultures (not obtained) might have shown decreasing concentrations of fungal organisms indicative of a gradual therapeutic response to our initial two drug combination. However, sterilization shortly after the addition of fluconazole is suggestive of its beneficial role. Amphotericin and flucytosine penetrate poorly into thombi; fluconazole may have more favorable penetration to such sites.12-14 Longman et al15 reported in a rabbit model superior eradication of candidal endocarditis with fluconazole alone compared with amphotericin and flucytosine singly or in combination. There was a case report of an infant with candidemia who developed brain abscesses that did not resolve until after the addition of fluconazole to several weeks of therapy with amphotericin B and flucytosine.16 This could be explained on the basis of excellent penetration of fluconazole into the central nervous system.17 The candida species isolated from our infant was susceptible in vitro to the three anti-fungals the baby received, arguing against resistance having a role in his persistent fungemia. However, the clinical value of in vitro antifungal susceptibility testing is limited.18 The E test method used here for antifungal susceptibility testing compares favorably with the National Committee for Clinical Laboratory Standards reference method, M27.19, 20 To our knowledge this is the first case of prolonged fungemia in a VLBW infant that rapidly responded to the addition of fluconazole after a prolonged course of amphotericin and flucytosine. Amphotericin B remains first line therapy for an ill VLBW infant with disseminated candidiasis. However, as our case may indicate, when fungemia persists despite use of amphotericin B, addition of fluconazole should be considered. Lisa-Gaye Robinson, M.D. Lucky Jain, M.D. Athena P. Kourtis, M.D., Ph.D. Divisions of Pediatric Infectious Diseases (LGR, APK) and Neonatology (LJ); Department of Pediatrics; Emory University School of Medicine; Atlanta, GA