Fluconazole Compared with Ketoconazole for the Treatment of Candida Esophagitis in AIDS A Randomized Trial Loren Laine, MD; Robin H. Dretler, MD; Chris N. Conteas, MD; Carmelita Tuazon, MD; Frederick M. Koster, MD; Fred Sattler, MD; Kathleen Squires, MD; and Muhammed Z. Islam, PhD

• Objective: To determine the clinical and endoscopic response of Candida esophagitis to antifungal therapy and to compare the two oral antifungal agents, fluconazole and ketoconazole. • Design: Multicenter, randomized, double-blind trial. • Setting: Fifteen U.S. centers including university, private practice, and county hospital settings. • Patients: A total of 169 patients with the acquired immunodeficiency syndrome (AIDS); odynophagia, dysphagia, or retrosternal pain; white esophageal plaques at endoscopy; and pseudohyphae on esophageal brushings or biopsies. • Intervention: Patients were randomly assigned to fluconazole (100 mg/d) or ketoconazole (200 mg/d). Doses were doubled at week 1 or 2 if no symptomatic improvement had occurred during the preceding week. Therapy was continued for 2 weeks after resolution of symptoms or for a maximum of 8 weeks. • Measurements: Patients were clinically evaluated weekly, and laboratory tests were done every 2 weeks. Endoscopy was repeated within 5 days after the end of therapy. • Results: A total of 143 patients were clinically evalu a t e (assessed within 7 days after therapy), and 129 patients were endoscopically evaluable (endoscopy repeated after therapy). Endoscopic cure occurred in 9 1 % of patients treated with fluconazole and in 52% of those given ketoconazole for a difference of 39% (95% CI, 24% to 52%; P< 0.001). Esophageal symptoms resolved in 85% of fluconazole-treated patients and in 65% of ketoconazole-treated patients for a difference of 20% (CI, 6% to 34%; P = 0.006). Intention-to-treat analyses also yielded statistically significant differences for the comparisons listed above. Side effects were minimal and comparable in the two groups; only one patient in each group had therapy discontinued for adverse effects that were possibly related to the study medications. • Conclusions: Fluconazole is associated with significantly greater rates of endoscopic and clinical cure than ketoconazole in patients with AIDS and Candida esophagitis. Both drugs appear to be safe and well tolerated.

C a n d i d a esophagitis is one of the most common opportunistic infections encountered in patients with the acquired immunodeficiency syndrome (AIDS) (1, 2) and also frequently complicates the course of other immunocompromised patients (3, 4). A number of agents have been used to treat Candida esophagitis (5-10), including topical medications such as nystatin, clotrimazole, and miconazole; oral compounds such as ketoconazole and 5-flucytosine; and intravenous agents such as amphotericin B. Yet, surprisingly little information is available regarding the response of Candida esophagitis to antifungal therapy. On the basis of limited data previously available, ketoconazole, an orally active imidazole compound, generally has been considered to be the treatment of choice for Candida esophagitis (1, 8). Fluconazole, a new, orally active, triazole antifungal agent, has greater in-vivo activity against Candida albicans than does ketoconazole and has more consistent absorption (11, 12). In addition, successful treatment of Candida esophagitis with fluconazole has been reported (13). We therefore did a randomized comparison of fluconazole and ketoconazole for the treatment of Candida esophagitis in patients with AIDS. Our multicenter study is the first large-scale trial that provides data on the clinical and endoscopic response of Candida esophagitis to treatment. Methods Patients with AIDS and symptoms of odynophagia, dysphagia, or retrosternal pain were eligible for this study if upper gastrointestinal endoscopy revealed white plaques or exudate consistent with Candida and if pseudohyphae were identified on brushings or biopsy of the esophageal mucosa. Patients were excluded if they were less than 13 years of age; if they were pregnant or lactating; if they were women of child-bearing age not using an effective method of contraception; if they had a history of allergy to imidazoles; if they were receiving concomitant therapy with immunostimulants, lymphocyte replacement, interferon, rifampin, ansamycin, warfarin, cyclosporin A, or medications that raise gastric pH (antacids, H2-receptor antagonists, or omeprazole); if they were using illicit drugs; if they had received antifungal therapy in the last 3 days; if they had evidence of noncandidal systemic fungal infection; if they had a life expectancy of less than 2 months; if their hemoglo-

Annals of Internal Medicine. 1992;117:655-660. From the Multicenter Candida Esophagitis Study Group. For current author addresses and a list of additional participants in the Study Group, see end of text.

Generic Name fluconazole ketoconazole

Drugs Brand Name Diflucan Nizoral

© 1992 American College of Physicians Downloaded from https://annals.org by Tulane University user on 12/10/2018

655

bin was less than 100 g/L, leukocyte count was less than 1.0 x 109/L, or platelet count was less than 60 x 109/L; if their alanine aminotransferase, aspartate aminotransferase, or alkaline phosphatase level was more than three times the upper limit of normal, bilirubin level was greater than 51.3 /xmol/L, or prothrombin time was more than 5 seconds over control; if their serum creatinine concentration was greater than 176.8 /xmol/L or they were on dialysis; or if they were unable or unwilling to give informed consent. Patients were randomly assigned at each center in a 1:1 ratio according to a computer-generated randomization schedule. All participating clinical centers received their study medication (bottles of identical capsules of fluconazole, 100 mg [Diflucan; Roerig-Pfizer, New York, New York] or ketoconazole, 200 mg [Nizoral; Janssen Pharmaceutica Inc., Piscataway, New Jersey]) and randomization schedules from a single central pharmacy. Patients were randomly assigned to receive either fluconazole, 100 mg/d, or ketoconazole, 200 mg/d. The dose of the drug was doubled at the end of either week 1 or week 2 if clinical symptoms had not improved in the preceding week. No patient received more than two capsules a day (200 mg of fluconazole or 400 mg of ketoconazole). Patients received a new supply of study medication weekly. Treatment continued for 2 weeks after resolution of symptoms (minimum of 3 weeks) or for a maximum of 8 weeks. Patients returned weekly for a clinical evaluation and pill count. Esophageal symptoms were graded by severity as absent, mild, moderate, or severe; and clinical responses were divided into " c u r e " (resolution of symptoms), "improvement" (decreased severity of symptoms as compared with baseline), and "failure" (lack of response). Patients had laboratory testing done every 2 weeks (complete blood count, blood urea nitrogen, creatinine, liver tests, electrolytes, glucose, and urinalysis). A repeat endoscopy was done within 5 days after the completion of therapy. The gross appearance of the esophageal mucosa was scored on a scale of 0 to 3 (0, normal esophageal mucosa; 1, individual raised plaques, each < 2 mm in size; 2, individual raised white plaques > 2 mm in size; 3, confluent plaques or plaques combined with ulceration). Patients were considered evaluable if they received at least 5 days of therapy. Patients were clinically evaluable if they had their last clinical evaluation within 7 days of discontinuing therapy and endoscopically evaluable if they had a repeat endoscopy after therapy. Intention-to-treat analysis of all patients enrolled was also done using the last assessment available for each patient. If no follow-up information was available after entry, the patient was considered to have no change from baseline for the intention-to-treat analysis. Table 1. Patient Characterssties> at Baseline in the Two Treatment Groups Characteristic

Male, % Mean age, v White, % Black, % Hispanic, % Treatment with zidovudine, % Antifungal therapy within 3 months, % Antifungal therapy within 1 week, % Odynophagia, % Dysphagia, % Retrosternal pain, % Oropharyngeal candidiasis, % Grade 1 esophageal candidiasis, % Grade 2 esophageal candidiasis*, % Grade 3 esophageal candidiasis*, %

Group Fluconazole Ketoconazole (n = 87) (": = 82) 95 35 59 29 13 66 49

94 36 57 27 16 63 48

38

30

83 87 75 49 17 34 48

91 78 80 54 12 59 28

* P < 0.01 for comparison of flu'conaizole and ketoconazo le groups. 656

15 October 1992 • Annals of Internal Medicine

Table 2. Summary of Patients Discontinuing Prematurely Reason for Discontinuation

Group Fluconazole (n = 87)

Ketoconazole (n = 82) n

Side effects possibly related to study medication Laboratory abnormalities possibly related to study medication Laboratory abnormalities unrelated to study medication Lack of efficacy Lost to follow-up Intercurrent illness Death due to intercurrent illness Did not meet selection criteria Other Total discontinuing therapy

0

1

1

0

0

2

1 8 3 1 3 2 19

8 6 5 1 1 3 27

Proportional data were compared using the chi-square test for 2 x 2 tables. When an ordinal variable with more than two categories of response was used as a classification variable, a Cochran-Mantel-Haenszel test was applied. Confidence limits for between-group differences were computed using the normal approximation to the binomial distribution. Because the baseline endoscopic grades were dissimilar in the two treatment groups, analyses were done for the clinical and endoscopic responses after adjusting for the baseline imbalance of endoscopic severity. Baseline endoscopy results were classified as severe (grade 3) or not severe (< grade 3) for this purpose, and the Cochran-Mantel-Haenszel test (14) was done. The probability of clinical cure by time was done using the Gray test for comparing treatment groups for cumulative incidence of competing risks (15); the 95% confidence intervals (CI) for the probabilities of cure were computed based on normal approximation of the estimates of probabilities and their variances. The treatment-by-center interactions in clinical and endoscopic responses were tested by applying the Breslow-Day test for homogeneity of the odds ratio. Statistical comparisons of response adjusted for the center were performed using CochranMantel-Haenszel statistics (14). All tests were done as twosided tests, and P values less than 0.05 were considered significant. This study was approved by the individual institutional review board of each participating center, and all patients gave written, informed consent. Results O n e h u n d r e d sixty-nine patients at 15 c e n t e r s (range, 1 to 47 patients per center) entered the trial. Patient characteristics are s h o w n in Table 1. T h e patients had a m e a n age of 36 y e a r s and w e r e p r e d o m i n a n t l y m a l e . O n e h u n d r e d nine patients (64%) w e r e taking zidovudine, and 82 patients (49%) had received antifungal therapy at s o m e time during t h e previous 3 m o n t h s . S e v e r e (grade 3) esophageal candidiasis w a s significantly m o r e frequent at the initial e n d o s c o p y in patients assigned to fluconazole than in t h o s e receiving k e t o c o n a z o l e . Forty-six patients discontinued t h e r a p y p r e m a t u r e l y (Table 2). O n e h u n d r e d forty-three patients w e r e clinically evaluable, and 129 w e r e endoscopically evaluable. Specific identification of the fungal p a t h o g e n b y culture w a s available in 130 c a s e s . Candida albicans was p r e s e n t in 121 of t h e s e c a s e s : Candida tropicalis in t w o

• Volume 117 • Number 8

Downloaded from https://annals.org by Tulane University user on 12/10/2018

Therapy

cases (both in the fluconazole group), unspecified Candida species in four cases (two in each treatment group), and Candida glabrata in three patients (two treated with fluconazole and one with ketoconazole). Pill counts revealed that 78% of evaluable patients took all of their study medication during the period of evaluation (fluconazole, 81%; ketoconazole, 76%) and that only 3% missed more than 10% of their medication (fluconazole, 1%; ketoconazole, 6%). The dose of study medication was doubled in nine (10%) patients treated with fluconazole and in 17 (21%) patients given ketoconazole (P = 0.06). Results of treatment are shown in Tables 3 and 4. Endoscopic cure of esophageal candidiasis (grade 0 at post-treatment endoscopy) occurred in 58 of 64 (91%) patients treated with fluconazole and in 34 of 65 (52%) given ketoconazole for a difference of 39% (CI, 24% to 52%; P < 0.001). Resolution of symptoms occurred in 61 of 72 (85%) patients treated with fluconazole and in 46 of 71 (65%) patients receiving ketoconazole for a difference of 20% (CI, 6% to 34%; P = 0.006). After adjustment for baseline dissimilarities in endoscopic grade, the P value for endoscopic cure was still < 0.001, and for clinical cure it was 0.005. Intention-totreat analysis showed similar significant differences in endoscopic (62 of 87 [71%] compared with 35 of 82 [43%] for a difference of 28%; CI, 14% to 43%; P < 0.001) and clinical (70 of 87 [80%] compared with 47 of 82 [57%] for a difference of 23%; CI, 10% to 37%; P = 0.001) rates of cure. Comparison of clinical and endoscopic responses showed that 53 of 55 (96%) fluconazole-treated patients with a clinical cure also had an endoscopic cure, whereas only 25 of 44 (57%) ketoconazole-treated patients with resolution of symptoms also had an endoscopic cure (P < 0.001). In addition, 12% of patients with an endoscopic cure had continued esophageal symptoms (fluconazole, 7%; ketoconazole, 21%; P = 0.045). The probabilities of clinical cure by time for the two treatment groups are shown in Figure 1. After 2 weeks of therapy, the probability of resolution of symptoms in fluconazole-treated patients was 53% compared with 39% in the ketoconazole group for a difference of 14% (CI, - 2 % to 31%; P = 0.09). At 4 weeks of treatment, the probability of clinical cure was 76% in the fluconazole group compared with 55% in the ketoconazole

Table 3. Endoscopic Assessment of Esophageal Candidiasis after Treatment in the Two Study Groups Grade at Post-Treatment Endoscopy

Group Fluconazole* Ketoconazole (n = 64) (n = 65) n_m

Grade Grade Grade Grade

0 (endoscopic resolution) 1 2 3

58 (91) 2 (3) 3 (5) 1 (2)

34 (52) 16 (25) 10 (15) 5 (8)

* P < 0.001 (Cochran-Mantel-Haenszel) when compared with ketoconazole.

Table 4. Clinical Response to Treatment in the Two Study Groups Clinical Response

Group Fluconazole* (n = 72)

Ketoconazole (n = 71)

n_m Cure Improvement Failure

61 (85) 7 (10) 4 (6)

46 (65) 15 (21) 10 (14)

* P = 0.006 (Cochran-Mantel-Haenszel) when compared with ketoconazole.

group for a difference of 21% (CI, 5% to 37%; P = 0.009). The antifungal medications used by evaluable patients in the 3 months before entry into the trial were clotrimazole (40% of antifungal agents used), nystatin (32%), ketoconazole (26%), miconazole vaginal cream (1%), and amphotericin B (1%). The two treatment groups did not differ significantly in the use of these agents. Patients who had received previous antifungal therapy had significantly lower rates of endoscopic and clinical cure in the ketoconazole group but not in the fluconazole group than did patients who had not used antifungal agents. Ketoconazole-treated patients who had received antifungal therapy in the previous 3 months had a 38% rate of endoscopic cure (26% if antifungal treatment had been given in the previous week) and a 49% rate of clinical cure (46% with therapy in the preceding week). The endoscopic and clinical cure rates among patients in the fluconazole group who had received antifungal agents in the previous 3 months were 86% and 76%, respectively—significantly greater than the rates in ketoconazole-treated patients who previously had been treated with antifungal therapy. When patients who had not received previous antifungal treatment were analyzed, fluconazole was associated with a significantly greater rate of endoscopic cure than was ketoconazole (33 of 35 [94%] compared with 23 of 36 [64%], P = 0.002) but not a significantly higher rate of clinical cure (35 of 38 [92%] compared with 29 of 36 [81%]). Zidovudine treatment did not significantly influence endoscopic or clinical response in either the fluconazole or ketoconazole group. Side effects were minimal and comparable in the two treatment groups. The most common side effects reported were nausea (fluconazole, 3%; ketoconazole, 7%), diarrhea (fluconazole, 5%; ketoconazole, 6%), and rash (fluconazole, 2%; ketoconazole, 7%). Only one patient was withdrawn because of side effects. This patient was taking several medications and developed pruritus and nausea after 4 weeks of ketoconazole therapy; he had achieved a clinical and endoscopic resolution at the time the study medication was stopped. Laboratory abnormalities were also relatively mild and similar in the two treatment groups. Threefold increases in transaminases and twofold elevations in alkaline phosphatase were seen in 9% of patients receiving fluconazole and in 6% of those given ketoconazole; no patient had a twofold or greater rise in bilirubin.

15 October 1992 • Annals of Internal Medicine • Volume 117 • Number 8 Downloaded from https://annals.org by Tulane University user on 12/10/2018

657

Figure 1. Probability of clinical cure by time for evaluable patients. The probability of cure at day 14 of treatment was 53% for fluconazole (O) and 39% for ketoconazole (O) for a difference of 14% (95% CI, - 2 % to 31%; P = 0.09), and at day 28 it was 76% for fluconazole and 55% for ketoconazole for a difference of 21% (CI, 5% to 37%; P = 0.009).

Only one patient was withdrawn because of laboratory abnormalities possibly related to the study drug. This patient had a two- to fourfold increase in transaminases during 4 weeks of treatment with fluconazole without improvement in esophageal symptoms; he also had been taking isoniazid for 10 weeks at the time of withdrawal from the study. Repeat endoscopy showed no evidence of candidiasis or other esophageal abnormality, and transaminases remained elevated at 1 month after withdrawal of fluconazole. Comparison of response of all evaluable patients after adjusting for centers using Cochran-Mantel-Haenszel statistics showed a P value of 0.008 for clinical response and < 0.001 for endoscopic response, indicating that fluconazole was better than ketoconazole in at least one center. The test of interaction of center and drug was not significant for either the clinical or endoscopic response. Discussion Symptoms of esophageal disease (odynophagia and dysphagia) are common in patients with AIDS and are reported to occur in approximately 40% of these patients (16). Because most cases of esophagitis in patients with AIDS are caused by Candida species (1, 2), it is important to determine the efficacy of antifungal therapy for Candida esophagitis. However, no large, prospective trial has examined the clinical and endoscopic response of Candida esophagitis to antifungal treatment. Ketoconazole generally has been considered to be the drug of choice for Candida esophagitis (1, 8), although data regarding its efficacy are relatively limited. A prospective study using 200 mg of ketoconazole a day in 12 patients with Candida esophagitis (none with AIDS) documented symptomatic and endoscopic resolution in 658

the 11 patients who completed a 2-week course of therapy (8). Connolly and colleagues (1), in a series of patients with AIDS and esophageal symptoms, reported that all 26 patients with only esophageal candidiasis had symptomatic resolution within 3 days after starting ketoconazole, 200 mg twice a day; however, only four patients had endoscopic evaluation after therapy. In a subsequent randomized, controlled trial of antifungal therapy for oral candidiasis, these investigators reported that the subgroup of patients with esophageal candidiasis assigned to receive ketoconazole (n < 12) had 100% clinical and endoscopic response at 4 weeks (17). Finally, in the only other randomized, controlled trial of therapy for Candida esophagitis, ketoconazole produced a symptomatic and endoscopic cure in four of six patients, whereas oral miconazole gel (an uncommonly used agent not available in the United States) led to a resolution of symptoms and gross esophageal lesions in all six patients treated (7). Our trial provides a large-scale, systematic evaluation of the clinical and endoscopic response of Candida esophagitis to fluconazole, a new oral triazole antifungal agent, compared with ketoconazole. At the end of treatment, approximately 90% of patients with AIDS who received fluconazole had endoscopically confirmed clearance of their candidiasis compared with approximately 50% of those patients receiving ketoconazole. Resolution of symptoms was also significantly more frequent in patients treated with fluconazole—20% higher than in those given ketoconazole. In addition, symptoms resolved more rapidly in patients treated with fluconazole. The greater difference in endoscopic compared with clinical cure rate between the two study groups is explained by the fact that over 40% of ketoconazoletreated patients had resolution of symptoms without an

15 October 1992 • Annals of Internal Medicine • Volume 117 • Number 8

Downloaded from https://annals.org by Tulane University user on 12/10/2018

endoscopic cure, whereas this occurred in only 4% of patients treated with fluconazole. Thus, symptoms of Candida esophagitis may resolve despite continued gross involvement of the esophageal mucosa. In addition, over 10% of patients continued to have esophageal symptoms after treatment despite a grossly normal esophagus at endoscopy. Other investigators also have noted that approximately 10% to 20% of patients seropositive for human immunodeficiency virus (HIV) who have esophageal symptoms have no evidence of esophageal abnormalities (1, 2). What are the potential explanations for the difference in therapeutic response to fluconazole and ketoconazole? The substitution of a triazole ring (fluconazole) for an imidazole ring (ketoconazole) results in greater polarity and water solubility for fluconazole as well as markedly reduced protein binding (11, 12). Oral absorption of fluconazole seems to be more consistent than absorption of ketoconazole (11, 12), and the half-life of fluconazole is significantly longer than that of ketoconazole (22 hours compared with 8 hours) (11, 12, 18). In addition, fluconazole has greater in-vivo activity against Candida species than does ketoconazole (11). Although the greater activity of fluconazole against Candida (11) may be the primary reason for its higher rate of cure in our study, the effect of gastric acid secretion also must be considered. Lake-Bakaar and colleagues (19, 20) have reported that most patients with AIDS have gastric acid hyposecretion and that absorption of ketoconazole in patients with AIDS is impaired as a result of reduced gastric acid secretion. Absorption of fluconazole, on the other hand, is not pH-dependent (21). Thus, in our study population of patients with AIDS, impaired absorption and reduced serum concentrations of ketoconazole might explain at least some of the difference in response rates to fluconazole and ketoconazole. The use of antifungal therapy (primarily clotrimazole, nystatin, and ketoconazole) in the 3 months before entry into the trial was associated with statistically significantly decreased rates of endoscopic and clinical cure in patients treated with ketoconazole but not in patients who received fluconazole. The hypothesis that previous antifungal therapy may increase resistance to ketoconazole was not evaluated in this study. Both fluconazole and ketoconazole were safe and well tolerated in the treatment of Candida esophagitis. Determining whether adverse effects are related to study medications can be difficult in patients with AIDS who receive multiple medications and often have concurrent illnesses. Side effects and laboratory abnormalities were generally mild, however, and did not differ significantly between the two treatment groups. Only one patient in each treatment group had medication stopped for side effects or laboratory abnormalities considered possibly related to the study drug. In summary, fluconazole is associated with significantly higher rates of endoscopic and clinical cure than ketoconazole in patients with AIDS and Candida esophagitis. Both fluconazole and ketoconazole appear to be safe and well tolerated.

Additional Participants in the Multicenter Candida Esophagitis Study Group Paul J. Cimoch, MD; William M. Reiter, MD; Joseph C. Meerof, MD, PhD; Carl S. Citron, MD: Center for Special Immunology, Irvine, California; Coleman Rotstein, MD: Division of Infectious Diseases, Roswell Park Memorial Institute, Buffalo, New York; Vincent G. Pons, MD: Division of Infectious Diseases, University of California, San Francisco, California; Eugene Speck, MD, PhD; Jerome Hruska, MD, PhD: Infectious Disease Consultants, Las Vegas, Nevada; William Lang, MD: ViRx, Inc., San Francisco, California; Richard Pollard, MD: Division of Infectious Diseases, University of Texas, Galveston, Texas; Jerrold Kuenn, MD: Medical Center Clinic. Pensacola, Florida; Andrea Katz, RN; David Parenti, MD; Gary Simon, MD; Michael Albert, MD: George Washington University Medical Center, Washington, DC; Francisco Garcia, LVN; Maurizio Bonacini, MD: University of Southern California School of Medicine, Los Angeles, California; Lawrence Cone, MD: Eisenhower Medical Center, Rancho Mirage, California; Michael Sands, MD: Bay State Medical Center, Springfield, Massachusetts. Grant Support: By Pfizer Central Research. Requests for Reprints: Loren Laine, MD, GI Division, Department of Medicine, U.S.C. School of Medicine, 2025 Zonal Avenue, Los Angeles, CA 90033. Current Author Addresses: Dr. Laine: GI Division, Department of Medicine, U.S.C. School of Medicine, 2025 Zonal Avenue, Los Angeles, CA 90033. Dr. Dretler: Infectious Diseases Specialists of Atlanta, PC, 2801 North Decatur Road, Suite 350, Decatur, GA 30033. Dr. Conteas: Southern California Permanente Medical Group, 1505 North Edgemont Street, Los Angeles, CA 90027. Dr. Tuazon: Infectious Diseases Division, George Washington University School of Medicine, 2150 Pennsylvania Avenue, NW, Washington, DC 20037. Dr. Koster: Infectious Diseases Division, University of New Mexico School of Medicine, Albuquerque, NM 87131. Dr. Sattler: Department of Medicine, L.A. County-U.S.C. Medical Center, U.S.C. School of Medicine, OPD 5P77, 1175 North Cummings Street, Los Angeles, CA 90033. Dr. Squires: Infectious Diseases Division, Cornell University Medical College, 1300 York Avenue, New York, NY 10021. Dr. Islam: Department of Clinical Research, Pfizer Central Research, Eastern Point Road, Groton, CT 06340.

References 1. Connolly GM, Hawkins D, Harcourt-Webster JN, Parsons PA, Husain OA, Gazzard BG. Oesophageal symptoms, their causes, treatment, and prognosis in patients with the acquired immunodeficiency syndrome. Gut. 1989;30:1033-9. 2. Bonacini M, Young T, Laine L. The causes of esophageal symptoms in human immunodeficiency virus infection. A prospective study of 110 patients. Arch Intern Med. 1991;151:1567-72. 3. Wheeler RR, Peacock JE Jr, Cruz JM, Richter JE. Esophagitis in the immunocompromised host: role of esophagoscopy in diagnosis. Rev Infect Dis. 1987;9:88-96. 4. Alexander JA, Brouillette DE, Chien MC, Yoo YK, Tarter RE, Gavaler JS, et al. Infectious esophagitis following liver and renal transplantation. Dig Dis Sci. 1988;33:1121-6. 5. Kantrowitz PA, Fleischli DJ, Butler WT. Successful treatment of chronic esophageal moniliasis with a viscous suspension of nystatin. Gastroenterology. 1969;57:424-30. 6. Lalor E, Rabeneck L. Esophageal candidiasis in AIDS Successful therapy with clotrimazole vaginal tablets taken by mouth. Dig Dis Sci. 1991;36:279-81. 7. Deschamps MM, Pape JW, Verdier RI, DeHovitz J, Thomas F, Johnson WD Jr. Treatment of Candida esophagitis in AIDS patients. Am J Gastroenterol. 1988; 83:20-1. 8. Fazio RA, Wickremesinghe PC, Arsura EL. Ketoconazole treatment of Candida esophagitis—a prospective study of 12 cases. Am J Gastroenterol. 1983;78:261-4.

15 October 1992 • Annals of Internal Medicine • Volume 117 • Number 8 Downloaded from https://annals.org by Tulane University user on 12/10/2018

659

9. Medoff G, Dismukes WE, Meade RH 3d, Moses JM. A new therapeutic approach to Candida infections. A preliminary report. Arch Intern Med. 1972;130:241-5. 10. Walsh TJ, Hamilton SR, Belitsos N. Esophageal candidiasis. Managing an increasingly prevalent infection. Postgrad Med. 1988;84: 193-205. 11. Saag MS, Dismukes WE. Azole antifungal agents: emphasis on new triazoles. Antimicrob Agents Chemother. 1988;32:1-8. 12. Galgiani JN. Fluconazole, a new antifungal agent. Ann Intern Med. 1990;113:177-9. 13. Hendel L, Svejgaard E, Walsoe I, Kieffer M, Stenderup A. Esophageal candidosis in progressive systemic sclerosis: occurrence, significance, and treatment with fluconazole. Scand J Gastroenterol. 1988;23:1182-6. 14. Landis RJ, Hayman ER, Koch GG. Average partial association in three-way contingency tables: a review and discussion of alternative tests. Int Stat Rev. 1978;46:237-54. 15. Gray RJ. A class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat. 1988;16:1141-54.

16. Gelb A, Miller S. AIDS and gastroenterology. Am J Gastroenterol. 1986;81:619-22. 17. Smith DE, Midgley J, Allan M, Connolly GM, Gazzard BG. Itraconazole versus ketaconazole in the treatment of oral and oesophageal candidosis in patients infected with HIV. AIDS. 1991;5:1367-71. 18. Dismukes WE. Azole antifungal drugs: old and new. Ann Intern Med. 1989;109:177-9. 19. Lake-Bakaar G, Tom W, Lake-Bakaar D, Gupta N, Beidas S, Elsakr M, et al. Gastropathy and ketoconazole malabsorption in the acquired immunodeficiency syndrome (AIDS). Ann Intern Med. 1988; 109:471-3. 20. Lake-Bakaar G, Quadros E, Beidas S, Elaskr M, Tom W, Wilson DE, et al. Gastric secretory failure in patients with the acquired immunodeficiency syndrome (AIDS). Ann Intern Med. 1988; 109: 502-4. 21. Blum RA, D'Andrea DT, Florentino BM, Wilton JH, Hilligoss DM, Gardner MJ, et al. Increased gastric pH and the bioavailability of fluconazole and ketoconazole. Ann Intern Med. 1991;114:755-7.

In 1736 I lost one of my sons, a fine boy of 4 years old, by the smallpox taken in the common way. I bitterly regretted that I had not given it to him by inoculation. This I mention for the sake of parents who omit that operation, on the supposition that they should never forgive themselves if a child died under it. My example shows that the regret may be the same either way, and that therefore the safer should be chosen. Benjamin Franklin Quoted in: William M. Schmidt Health and welfare of colonial American children. Am J Dis Child. 1976;130:694-701. Submitted by: Mary E. Wilson, MD Mount Auburn Hospital Cambridge, MA 02238

Submissions from readers are welcomed. If the quotation is published, the sender's name will be acknowledged. Please include a complete citation, as done for any reference.—The Editors

660

15 October 1992 • Annals of Internal Medicine • Volume 117 • Number 8

Downloaded from https://annals.org by Tulane University user on 12/10/2018

Fluconazole compared with ketoconazole for the treatment of Candida esophagitis in AIDS. A randomized trial.

To determine the clinical and endoscopic response of candida esophagitis to antifungal therapy and to compare the two oral antifungal agents, fluconaz...
1MB Sizes 0 Downloads 0 Views