Comparison of Three Regimens for Treatment of Mild to Moderate Pneumocystis carinii Pneumonia in Patients with AIDS: A Double-Blind, Randomized Trial of Oral Trimethoprim-Sulfamethoxazole, Dapsone-Trimethoprim, and Clindamycin-Primaquine

  1. Sharon Safrin, MD, MPH;
  2. Dianne M. Finkelstein, PhD;
  3. Judith Feinberg, MD;
  4. Peter Frame, MD;
  5. Gail Simpson, MD;
  6. Albert Wu, MD, MPH;
  7. Tony Cheung, MD;
  8. Ruy Soiero, MD;
  9. Peter Hojczyk, BS; and
  10. John R. Black, MD
  1. From University of California, San Francisco, San Francisco, California; Harvard School of Public Health, Boston, Massachusetts; Johns Hopkins University School of Medicine, Baltimore, Maryland; University of Cincinnati College of Medicine, Cincinnati, Ohio; Harbor-UCLA Medical Center, Torrance, California; Mount Sinai Medical Center and Albert Einstein College of Medicine, New York, New York; Frontier Science Technology and Research Foundation, Amherst, New York; and Methodist Hospital of Indiana, Indianapolis, Indiana. Acknowledgments: The authors thank Drs. John Mills, Fred Sattler, and Sam Bozzette for consultation in the design and performance of this trial. They also thank the patients who participated in the trial. Grant Support: By grants from the AIDS Clinical Trials Group of the National Institute of Allergy and Infectious Diseases and the General Clinical Research Units of the National Center for Research Resources and Agency for Health Care Policy and Research grant HS07824. Requests for Reprints: Sharon Safrin, MD, San Francisco General Hospital, Building 80, Ward 84, San Francisco, CA 94110. Current Author Addresses: Dr. Safrin: San Francisco General Hospital, Building 80, Ward 84, San Francisco, CA 94110.
     
    Next Section

    Abstract

    Objective: To compare the tolerability and efficacy of three oral regimens for the treatment of patients with the acquired immunodeficiency syndrome (AIDS) and Pneumocystis carinii pneumonia.

    Design: A randomized, double-blind study.

    Setting: 24 U.S. academic medical centers.

    Patients: 181 patients with morphologically confirmed P. carinii pneumonia and alveolar-arterial oxygen differences (PAO2-PaO2) of 45 mm Hg or less.

    Intervention: Patients were randomly assigned to receive trimethoprim-sulfamethoxazole, dapsone-trimethoprim, or clindamycin-primaquine for 21 days. Patients with a PAO (2-PaO)2 of 35 to 45 mm Hg at study entry also received prednisone.

    Measurements: Serial clinical and laboratory evaluations for therapeutic response and toxicity. Therapeutic failure at day 21 was defined by any of the following: increase in PAO2-PaO2 of greater than 20 mm Hg; no remission of baseline signs and symptoms; and change in antipneumocystis therapy for reasons other than toxicity, intubation, or death. Dose-limiting toxicity was defined as discontinuation of therapy by the primary physician because of one or more adverse reactions.

    Results: No statistically significant differences were seen among treatment groups in the proportions of patients who had dose-limiting toxicity (P = 0.2), therapeutic failure (P > 0.2), or a complete course of therapy (P > 0.2). Survival during therapy or for 2 months thereafter did not differ among the three groups (P > 0.2). However, elevation of serum aminotransferase levels to more than five times the baseline levels was more frequent in the trimethoprim-sulfamethoxazole group (P = 0.003), and one or more serious hematologic toxicities (neutropenia, anemia, thrombocytopenia, or methemoglobinemia) occurred more frequently in the clindamycin-primaquine group (P = 0.01).

    Conclusions: The rates of dose-limiting toxicity, therapeutic failure, and survival did not differ among patients with AIDS who were receiving oral trimethoprim-sulfamethoxazole, dapsone-trimethoprim, or clindamycin-primaquine for mild to moderate P. carinii pneumonia. However, the limited sample size prevents the unequivocal demonstration of the equality of these three regimens. Differences in expected categories of toxicities associated with each regimen should guide the clinician in choosing first-line therapy, particularly for patients with baseline hepatic insufficiency or myelosuppression.

    *For additional members of the ACTG 108 study group, see the Appendix.

    In 1994, 15 440 cases of Pneumocystis carinii pneumonia occurring in the United States were reported to the Centers for Disease Control and Prevention [1]. Thus, despite the advent of prophylactic agents to prevent this infection, the need for effective and nontoxic therapeutic regimens remains. Increased physician and patient awareness, along with improved methods of diagnosis, have made earlier institution of ambulatory therapy with oral medications a feasible alternative to hospitalization for inpatient treatment in many instances.

    Previous studies [2-8] suggest that the efficacy of trimethoprim-sulfamethoxazole, available since 1968, is equivalent or superior to that of all alternative therapies for P. carinii pneumonia. However, rates of treatment-limiting toxicity ranging from 20% to 57% in patients with the acquired immunodeficiency syndrome (AIDS) who receive this regimen [2, 3, 5, 7, 8] have necessitated a continued search for better-tolerated regimens. In one study [9], the combination of dapsone and trimethoprim was successfully used to treat 15 patients with a first episode of P. carinii pneumonia. In a subsequent randomized trial [5], this combination was compared with trimethoprim-sulfamethoxazole in 60 patients with arterial oxygen pressures of 60 mm Hg or greater. In this latter study, the efficacy of dapsone-trimethoprim was similar to that of trimethoprim-sulfamethoxazole (93% compared with 90%), but dapsone-trimethoprim was associated with a lower frequency of major toxicities (30% compared with 57%).

    The combination of clindamycin with primaquine has shown excellent activity against P. carinii in in vitro studies and in an experimental rat model [10]. Successful use of this regimen in the treatment of P. carinii pneumonia, generally with intravenous administration of clindamycin for all or part of therapy, has been described since 1989 [11-15]. In one study of 60 patients with an alveolar-arterial oxygen difference (PAO2-PaO2) of 40 mm Hg or less [15], the administration of intravenous or oral clindamycin and oral primaquine was associated with therapeutic success in 92% of patients and with doselimiting toxicity in 15% of patients. A randomized trial [16] compared intravenous clindamycin and oral primaquine with intravenous or oral trimethoprim-sulfamethoxazole in 49 patients with a first episode of P. carinii pneumonia and an arterial oxygen pressure of 50 mm Hg or greater; 90% of patients in each group were classified as having successful therapy, and dose-limiting toxicity occurred in 18% and 20% of patients, respectively.

    Thus, although dapsone-trimethoprim and clindamycin-primaquine have gained widespread use in the treatment of P. carinii pneumonia, their relative efficacies have not yet been validated in a large controlled trial, and their toxicity profiles have not been directly compared. To guide the clinician in selecting the optimal oral therapy for patients with AIDS and mild to moderate P. carinii pneumonia, we compared the toxicities and efficacies of trimethoprim-sulfamethoxazole, dapsone-trimethoprim, and clindamycin-primaquine in a randomized, doubleblind multicenter trial.

    Previous SectionNext Section

    Methods

    Beginning in May 1991, patients were enrolled at 24 centers participating in the AIDS Clinical Trials Group (ACTG) of the National Institutes of Allergy and Infectious Diseases (NIAID). Each site's institutional review board approved the study (ACTG trial 108), and all participants gave informed consent before the study drug was administered.

    Patients

    Eligible patients had human immunodeficiency virus (HIV) infection, were older than 13 years of age, weighed 35 to 100 kg, and had symptoms or signs of P. carinii pneumonia, such as cough, shortness of breath, or an abnormal chest radiograph. Enrollment was limited to patients whose room air PAO2-PaO2 was 45 mm Hg or greater. Morphologic confirmation of the diagnosis by visualization of P. carinii in induced sputum, bronchoscopic lavage, or transbronchial biopsy specimens was required within 10 days of study entry. Treatment of P. carinii pneumonia lasting no more than 24 hours was permitted before randomization. Exclusion criteria were concurrent pulmonary pathologic conditions that could obscure the evaluation of response to therapy; the third trimester of pregnancy; receipt of systemic corticosteroids within 7 days of study entry; deficiency of glucose-6-phosphate-dehydrogenase (G6PD) or nicotinamide adenine dinucleotide methemoglobin reductase; hemoglobin M abnormality; previous enrollment in the study; inability to receive oral therapy; and serum creatinine level greater than 152.5 µmol/L, hemoglobin level less than 80 g/L, absolute neutrophil count less than 0.75 × 109/L, platelet count less than 50 × 109/L, or alanine aminotransferase levels greater than 7.5 times the upper limit of normal.

    Randomization and Dosing

    Patients were assigned to treatment on the basis of a permuted block randomization. Randomization was stratified by treatment center and by the use of antipneumocystis prophylaxis within 30 days and was accomplished by computerized linkage to a central data management center. Active study drug and placebo assignments were implemented by each site's pharmacist, who labeled the bottles in a blinded manner. The Burroughs Wellcome Company (Research Triangle Park, North Carolina), the Jacobus Pharmaceutical Company (Princeton, New Jersey), the Upjohn Company (Kalamazoo, Michigan), and Sterling-Winthrop Pharmaceuticals (New York, New York) provided the study drug.

    The dosages of the study drugs were as follows: dapsone, 100 mg daily; clindamycin, 600 mg three times daily; and primaquine base, 30 mg daily. The dosages of trimethoprim and sulfamethoxazole were based on patient weight: Patients weighing 51 to 80 kg received two double-strength trimethoprim-sulfamethoxazole tablets (320:1600 mg) three times daily or trimethoprim (300 mg) three times daily with dapsone once daily. Patients weighing 36 to 50 kg received 240:1200 mg of trimethoprim-sulfamethoxazole (1.5 double-strength tablets) three times daily or 200 mg of trimethoprim three times daily with dapsone once daily. Patients weighing 81 to 99 kg received 400:2000 mg of trimethoprim-sulfamethoxazole (2.5 double-strength tablets) three times daily or 400 mg of trimethoprim three times daily with dapsone once daily. To maintain a doubleblind status, all patients received one active regimen and one placebo regimen. Patients with a PAO2-PaO2 of 35 to 45 mm Hg received adjunctive prednisone, 40 mg twice daily for 5 days, then 40 mg daily for 5 days, then 20 mg daily until antipneumocystis therapy was discontinued [17]. Patients with a history of intolerance to trimethoprim-sulfamethoxazole were enrolled beginning in September 1992 and were randomly assigned to one of the other treatment arms.

    Therapy was administered for 21 ± 1 days. For patients with dose-limiting toxicity, the protocol specified either double-blind crossover to an alternative regimen (according to a second randomized list) or the substitution of intravenous pentamidine (3 to 4 mg/kg of body weight daily). Antipneumocystis therapy could be terminated if the patient had received therapy for at least 14 days and if clinical signs and symptoms had remitted. Patients meeting criteria for therapeutic failure (see below) were to receive intravenous pentamidine to complete therapy. We did not permit concurrent therapy with zidovudine, ganciclovir, colony-stimulating factors, rifampin, rifabutin, folinic acid, investigational agents other than triazole antifungal agents, and other medications potentially effective against P. carinii (such as pyrimethamine and sulfadiazine).

    Clinical and Laboratory Assessments

    At baseline, physical examination, venipuncture (for complete blood count with differential; reticulocyte count; and determination of creatinine, aminotransferase, and lactic acid dehydrogenase levels), measurement of room air arterial blood gas, and chest radiography were done. Physical examination and venipuncture were repeated on days 0, 3, 7, 10, 14, and 21 of therapy; arterial blood gas determination was repeated on days 7 and 21; and chest radiography was repeated on day 7. Serum methemoglobin levels were measured on days 3, 7, and 10 of therapy. Physical examination, venipuncture, and chest radiograph were repeated 2 weeks after therapy was completed. Survival status and recurrence of P. carinii pneumonia were determined 60 days after completion of therapy. Secondary antipneumocystis prophylaxis was advised for all patients who completed the study, and each patient's primary physician chose the medication.

    We used a battery of instruments to assess the effect of treatment on patient-reported health status. Physical function was measured using the Duke Activity Status Index [18], a 12-item index weighted on the basis of known metabolic costs of each activity. Energy, pain, and general health perceptions were measured using scales from the Medical Outcomes Study [19], supplemented by four additional general health items [20]. Disability was measured by the number of days spent in bed or the decrease in the number of usual activities the patient could perform [21]. Severity of pulmonary (cough, dyspnea, and chest tightness) and other symptoms (fever, pain, nausea, rash, and dizziness) was assessed using a questionnaire that required approximately 5 minutes to complete and was available in English and Spanish [22].

    Definitions of End Points

    Therapeutic failure at day 7 was defined by one of the following: 1) increase in PAO2-PaO2 of greater than 20 mm Hg over baseline without remission of baseline signs and symptoms; 2) change in antipneumocystis therapy for reasons other than toxicity; 3) intubation; and 4) death. Therapeutic failure at day 21 was defined by any of the above variables or by therapeutic failure at day 7. We used neither persistence of fever (because of its multifactorial nature) nor lack of improvement seen on chest radiograph (because of its typical lag behind clinical improvement and the potential for interobserver variation) to exclusively assess therapeutic failure. Previous studies [8, 15] have shown no correlation between a failure to improve in these measures and overall clinical success. Toxicities were graded according to the NIAID standardized system (range, I to IV). Dose-limiting toxicity was defined as the discontinuation of therapy by the primary physician because of one or more adverse effects, regardless of grade. Survival was defined as the patient being alive 60 days after the discontinuation of therapy.

    Statistical Analysis

    The original accrual goals of our study were larger; however, we reconfigured the sample size in February 1993 because of low rates of both therapeutic failure and accrual. After redesign, we ultimately sought 195 patients with P. carinii pneumonia so that our study would have 80% power to detect differences in the rates of dose-limiting toxicity of 25% or greater (for example, 55% compared with 30%). We set the last date of study entry as 30 June 1993 according to our expectation of enrolling 195 patients by this date.

    We compared baseline variables using the chisquare, Fisher exact, Student t, or analysis of variance tests; all P values are two-sided. We calculated rates of therapeutic failure and dose-limiting toxicity from Kaplan-Meier estimates and compared these outcomes using a log-rank test. For infrequent events such as death and therapeutic failure, we used odds ratios to approximate the relative risk; the odds ratios are followed by exact 95% CIs (StatXact, Cytel Software Corp., Cambridge, Massachusetts). For more common outcomes (such as completion of therapy and dose-limiting toxicity), we provide relative risks and estimates of the 95% CIs (SAS, SAS Institute, Inc., Cary, North Carolina). Therapeutic failures were censored only by loss to follow-up, whereas dose-limiting toxicity was censored by therapeutic failure and loss to follow-up. We compared graded toxicities that occurred during receipt of the initially assigned regimen using the Kruskal-Wallis exact test. Unless otherwise specified, all analyses were done according to intention to treat. We analyzed change in health status scores using the Kruskal-Wallis and Wilcoxon rank-sum tests at days 7 and 21. For scales in which at least 50% of items were completed, mean scores were substituted for the missing responses. To provide a summary statistic for health status, we used a random-effects model with initial group assignment and dimension of health status as fixed effects and person as a random effect [23].

    Previous SectionNext Section

    Results

    Two hundred fifty-six patients were enrolled in the study on the basis of suspected P. carinii pneumonia; we report on the 181 patients in whom the diagnosis was morphologically confirmed. Most patients were male (89%), had not previously received antiretroviral therapy (62%), had no history of P. carinii pneumonia (90%), and had never received antipneumocystis prophylaxis (67%) (Table 1). Forty-four percent of patients were from racial and ethnic minority groups. Of 60 patients receiving antipneumocystis prophylaxis at study entry, 48% were receiving systemic agents and 52% were receiving aerosolized pentamidine. Ninety-four patients had received treatment for P. carinii pneumonia for as long as 24 hours before randomization: trimethoprim-sulfamethoxazole (84 patients), dapsone-trimethoprim (2 patients), or intravenous pentamidine (8 patients). No statistically significant differences were seen among treatment groups in the distribution of demographic or laboratory characteristics at the time of study entry (Table 1). The mean room air PAO2-PaO2 was 27 mm Hg; in 30% of patients, the PAO2-PaO2 was greater than 35 mm Hg. Ninety-six percent of patients had cough or shortness of breath, and 95% had abnormalities seen on chest radiograph. In all seven patients who did not have cough or shortness of breath at study entry, interstitial infiltrates were seen on chest radiograph. Three percent of enrolled patients withdrew from therapy prematurely, were lost to follow-up, or never began therapy with the study drug.

    View this table:
    Table 1. Characteristics of Patients with Pneumocystis carinii Pneumonia at Study Entry according to Treatment Assignment*

    Response to Therapy

    Fifty-four percent of patients completed a full course of therapy for P. carinii pneumonia while receiving the initially assigned treatment regimen (50% of trimethoprim-sulfamethoxazole recipients, 59% of dapsone-trimethoprim recipients, and 52% of clindamycin-primaquine recipients; P > 0.2) (Table 2). Pairwise comparisons showed no statistical differences. For trimethoprim-sulfamethoxazole compared with dapsone-trimethoprim, the relative risk was 0.8 (CI, 0.6 to 1.2); for trimethoprim-sulfamethoxazole compared with clindamycin-primaquine, the relative risk was 1.0 (CI, 0.7 to 1.4); and for dapsonetrimethoprim compared with clindamycin-primaquine, the relative risk was 1.1 (CI, 0.8 to 1.6).

    View this table:
    Table 2. Outcome of Therapy according to Treatment Assignment

    At the day 7 evaluation, 11 patients (6.1%) met the criteria for therapeutic failure; no statistically significant differences were seen among the three treatment groups (8% of trimethoprim-sulfamethoxazole recipients, 5% of dapsone-trimethoprim recipients, and 5% of clindamycin-primaquine recipients; P > 0.2) (Table 2). The odds ratio estimates for day 7 failure were the following: trimethoprim-sulfamethoxazole compared with dapsone-trimethoprim, 0.6 (CI, 0.09 to 3.4); trimethoprim-sulfamethoxazole compared with clindamycin-primaquine, 0.6 (CI, 0.1 to 3.5); and dapsone-trimethoprim compared with clindamycinprimaquine, 1.0 (CI, 0.1 to 3.5). By day 21, 17 patients (9%) were considered to have therapeutic failure (9% of trimethoprim-sulfamethoxazole recipients, 12% of dapsone-trimethoprim recipients, and 7% of clindamycin-primaquine recipients; P > 0.2). The odds ratios were the following: trimethoprim-sulfamethoxazole compared with dapsone-trimethoprim, 1.3 (CI, 0.4 to 5.0); trimethoprim-sulfamethoxazole compared with clindamycin-primaquine, 0.7 (CI, 0.1 to 3.2); and dapsone-trimethoprim compared with clindamycin-primaquine, 0.6 (CI, 0.1 to 2.3).

    The time to therapeutic failure was similar among the three groups whether the comparison was three-way (P > 0.2) or pairwise (trimethoprim-sulfamethoxazole compared with dapsone-trimethoprim or clindamycin-primaquine [P = 0.2 for both comparisons]; dapsone-trimethoprim compared with clindamycin-primaquine [P > 0.2]) (Figure 1). Most therapeutic failures occurred within the first week of therapy (70%); regardless of treatment group, only 12% of patients met the criteria for therapeutic failure within the second week of P. carinii pneumonia therapy, and only 18% of patients met the criteria within the third week (Figure 2).

    Figure 1. Log-rank value more than 0.2. * equals trimethoprim-sulfamethoxazole group (TS); † equals dapsone-trimethoprim group (DT); ‡ equals clindamycin-primaquine group (CP).
    View larger version:
    Figure 1. Log-rank value more than 0.2. * equals trimethoprim-sulfamethoxazole group (TS); † equals dapsone-trimethoprim group (DT); ‡ equals clindamycin-primaquine group (CP). Time until therapeutic failure, according to treatment regimen.P
    Figure 2. CP equals clindamycin-primaquine; DT equals dapsone-trimethoprim; TS equals trimethoprim-sulfamethoxazole.
    View larger version:
    Figure 2. CP equals clindamycin-primaquine; DT equals dapsone-trimethoprim; TS equals trimethoprim-sulfamethoxazole. Occurrence of dose-limiting toxicities and therapeutic failure according to day of antipneumocystis therapy and treatment regimen.

    We found no association between the receipt of antipneumocystis prophylaxis in the 30 days before study entry and subsequent failure to respond to treatment with that agent. Only one of six patients receiving trimethoprim-sulfamethoxazole for both prophylaxis and treatment was classified as having therapeutic failure (day 3); this patient successfully completed therapy with intravenous trimethoprim-sulfamethoxazole in the ensuing 3 weeks. None of the five patients who received dapsone as antipneumocystis prophylaxis and then received dapsonetrimethoprim therapy was classified as having therapeutic failure. We also detected no benefit of receipt of antipneumocystis therapy in the 24 hours before randomization: Ten of 94 patients who had such therapy (10.6%) were ultimately classified as having therapeutic failures compared with 7 of 87 (8.0%) patients who had not received therapy before randomization (P > 0.2). One of 8 patients receiving intravenous pentamidine before randomization had failure compared with 9 of 84 patients receiving trimethoprim-sulfamethoxazole (P = 1.0).

    Because the daily dosage of study drug varied according to body weight, we evaluated the association between the weighted dosage of drug and the occurrence of therapeutic failure. We found no statistically significant association for any of the study drugs (P > 0.2 for trimethoprim in the trimethoprim-sulfamethoxazole and dapsone-trimethoprim regimens; P > 0.2 for sulfamethoxazole; P = 0.1 for dapsone; P > 0.2 for clindamycin; and P > 0.2 for primaquine).

    Only eight deaths were reported during the 81-day study period; three of these were associated with unresolved P. carinii pneumonia, and one was associated with recurrence of P. carinii pneumonia. Four patients had each received trimethoprim-sulfamethoxazole for 21 days; two had received dapsone-trimethoprim for 3 and 21 days, respectively; and two had received clindamycin-primaquine for 3 and 21 days, respectively (P > 0.2). The odds ratios were as follows: trimethoprim-sulfamethoxazole compared with dapsone-trimethoprim, 0.5 (CI, 0.05 to 3.9); trimethoprim-sulfamethoxazole compared with clindamycin-primaquine, 0.5 (CI, 0.05 to 3.9); and dapsone-trimethoprim compared with clindamycin-primaquine, 1.0 (CI, 0.07 to 14.4).

    In seven patients, P. carinii pneumonia recurred 21 to 76 days after completion of therapy; all patients had received secondary antipneumocystis prophylaxis after successful completion of acute therapy. One patient had been assigned to receive trimethoprim-sulfamethoxazole, four had been assigned to receive dapsone-trimethoprim (two of the four had been switched to clindamycin-primaquine on days 3 and 11, respectively), and two had been assigned to receive clindamycin-primaquine (one of the two had been switched to dapsone-trimethoprim on day 10).

    Adverse Reactions

    Dose-limiting toxicity occurred in 56 patients (31%) (36% of trimethoprim-sulfamethoxazole recipients, 24% of dapsone-trimethoprim recipients, and 33% of clindamycin-primaquine recipients; P = 0.2) (Table 2). Seventy-one percent of these toxicities were classified as grade III or IV. Pairwise comparisons showed no differences in the rates of dose-limiting toxicity: trimethoprim-sulfamethoxazole compared with dapsone-trimethoprim, P = 0.2 (odds ratio, 0.6; CI, 0.2 to 1.3); trimethoprim-sulfamethoxazole compared with clindamycin-primaquine, P > 0.2 (odds ratio, 0.9; CI, 0.4 to 2.0); and dapsone-trimethoprim compared with clindamycin-primaquine, P = 0.2 (odds ratio, 1.6; CI, 0.6 to 3.9) (Figure 3). Although dose-limiting toxicity occurred most frequently in each treatment group during the second week of therapy (63%), we also noted such toxicities in the first week of therapy (21%) (Figure 2). In patients receiving trimethoprim-sulfamethoxazole, toxicities continued to occur during the third week of therapy; thus, the occurrence of dose-limiting toxicities was distributed throughout the course of therapy (Figure 2). Twenty-one study patients had a history of intolerance to trimethoprim-sulfamethoxazole. Three of the 11 (27%) patients randomly assigned to receive dapsone-trimethoprim compared with 12 of 48 (25%) tolerant patients receiving dapsone-trimethoprim had dose-limiting toxicities. Two of 10 (20%) patients with trimethoprim-sulfamethoxazole intolerance who received clindamycin-primaquine had dose-limiting toxicities compared with 17 of the 48 (35%) other patients assigned to that group. The mg/kg daily dosage of study drug was not statistically associated with the occurrence of dose-limiting toxicity (P > 0.2 for trimethoprim with sulfamethoxazole and P = 0.1 for trimethoprim with dapsone; P > 0.2 for sulfamethoxazole; P = 0.1 for dapsone; P > 0.2 for clindamycin; and P > 0.2 for primaquine).

    Figure 3. Log-rank value more than 0.2. * equals trimethoprim-sulfamethoxazole group (TS); † equals dapsone-trimethoprim group (DT); ‡ equals clindamycin-primaquine group (CP).
    View larger version:
    Figure 3. Log-rank value more than 0.2. * equals trimethoprim-sulfamethoxazole group (TS); † equals dapsone-trimethoprim group (DT); ‡ equals clindamycin-primaquine group (CP). Time until dose-limiting toxicity, according to treatment regimen.P

    Rash was the most frequent manifestation of dose-limiting toxicity in all three treatment groups (19% of trimethoprim-sulfamethoxazole recipients, 10% of dapsone-trimethoprim recipients, and 21% of clindamycin-primaquine recipients; P = 0.2) (Table 3). Although no patient had exfoliative dermatitis or mucosal involvement, grade III rashes (characterized by vesiculation, moist desquamation, or ulceration) occurred most frequently in patients receiving clindamycin-primaquine (6% of trimethoprim-sulfamethoxazole recipients, 2% of dapsone-trimethoprim recipients, and 16% of clindamycin-primaquine recipients; P = 0.07). The other most frequent dose-limiting toxicities were elevated serum aminotransferase levels and nausea or vomiting in the trimethoprim-sulfamethoxazole group, nausea or vomiting in the dapsone-trimethoprim group, and anemia in the clindamycin-primaquine group (Table 3). According to the NIAID grading system, the most frequent serious toxicities (that is, grade III or IV) in patients receiving trimethoprim-sulfamethoxazole were fever (temperature more than 39.5 °C) (19%) and elevation of serum alanine or aspartate aminotransferase levels greater than five times the upper limit of normal (19%). In patients receiving dapsone-trimethoprim, they were fever (19%), neutropenia (absolute neutrophil count less than 0.75 × 109/L) (5%), anemia (hemoglobin level less than 80 g/L) (5%) and nausea or vomiting (5%). In patients receiving clindamycin-primaquine, they were rash (16%), neutropenia (14%), and anemia (10%).

    View this table:
    Table 3. Categories of Dose-Limiting Toxicity by Treatment Assignment*

    Serious hematologic toxicities (that is, grade III or IV neutropenia or anemia, platelet count less than 50 × 109/L, or methemoglobin levels of 15% or greater) occurred most frequently in patients receiving clindamycin-primaquine (11% of trimethoprim-sulfamethoxazole recipients, 12% of dapsone-trimethoprim recipients, and 28% of clindamycin-primaquine recipients; P = 0.01). Neutropenia and anemia occurred more frequently in patients receiving clindamycin-primaquine (neutropenia developed in 6% of trimethoprim-sulfamethoxazole recipients, 5% of dapsone-trimethoprim recipients, and 14% of clindamycin-primaquine recipients [P = 0.1]; anemia developed in 5% of trimethoprim-sulfamethoxazole recipients, 5% of dapsone-trimethoprim recipients, and 10% of clindamycin-primaquine recipients [P = 0.2]). However, these conditions were dose limiting in only three patients (two of whom had concurrent neutropenia and anemia). Severe thrombocytopenia occurred in only one patient while he was receiving trimethoprim-sulfamethoxazole. No patients in the trimethoprim-sulfamethoxazole group had methemoglobin levels of 15% or greater compared with one patient (2%) in the dapsone-trimethoprim group and four patients (7%) in the clindamycin-primaquine group. Although three patients (one receiving dapsone-trimethoprim and two receiving clindamycin-primaquine) had methemoglobinemia that was classified as potentially life-threatening (that is, methemoglobin level more than 20%), only one (a clindamycin-primaquine recipient) had a case of methemoglobinemia considered to be dose limiting by the primary physician. This patient was treated with methylene blue as an antidote on day 8. One additional patient classified as having dose-limiting methemoglobinemia had methemoglobin levels of 10% to 15% while receiving clindamycin-primaquine.

    Elevations in serum alanine or aspartate aminotransferase levels to greater than five times the upper limit of normal occurred most frequently in the group receiving trimethoprim-sulfamethoxazole (19% of trimethoprim-sulfamethoxazole recipients, 3% of dapsone-trimethoprim recipients, and 7% of clindamycin-primaquine recipients; P = 0.003). Severe diarrhea (> 7 stools/d with or without dehydration) occurred in only one patient each in the dapsone-trimethoprim and clindamycin-primaquine groups; in the former patient, the diarrhea was considered dose limiting. In four other patients, Clostridium difficile toxin was found in the stool. Three of the four patients had received clindamycin-primaquine, but the association with the study drug seemed clear in only one patient on day 14 of therapy. In two patients, C. difficile toxin-associated diarrhea was diagnosed 21 and 27 days, respectively, after the completion of acute therapy; these patients had also received trimethoprim-sulfamethoxazole, cefotaxime, ampicillin, or pentamidine. In the fourth patient, C. difficile toxin was found in the stool 34 days after the patient had received therapy for P. carinii pneumonia with dapsone-trimethoprim, intravenous and oral trimethoprim-sulfamethoxazole, and intravenous pentamidine.

    Quality of Life Analysis

    In the three groups, rates of the completion of the health status questionnaire at baseline, day 7, and day 21 were 87%, 80%, and 74%, respectively. Completion rates were balanced across groups at baseline and day 7 and were highest for the clindamycin-primaquine group at day 21. Respondents and nonrespondents were similar at baseline and day 21. At day 7, respondents were less likely than nonrespondents to have had therapeutic failure (P = 0.09), dose-limiting toxicity (P > 0.2), or change from initial therapy (P = 0.007).

    At day 7, the median increase in patient-reported health status scores was greatest for patients assigned to receive clindamycin-primaquine (Table 4). This difference was particularly marked in comparison with patients receiving trimethoprim-sulfamethoxazole (for general health perception, P = 0.06; for pulmonary symptoms, P = 0.05; for Duke Activity Status Index, P = 0.03; and for disability, P = 0.01). Day 7 scores, averaged across all dimensions by multivariate analysis, were higher in clindamycin-primaquine recipients than in trimethoprim-sulfamethoxazole recipients (P = 0.007) or dapsone-trimethoprim recipients (P = 0.2). By day 21, health status scores in all patients had improved further, and differences among the groups were less evident (P > 0.05 for all pairwise and threeway comparisons).

    View this table:
    Table 4. Median Improvement in Health Status Scores during Therapy*
    Previous SectionNext Section

    Discussion

    We found no statistically significant differences in the proportion of patients in whom therapy was changed because of toxicity, an event that occurred in nearly one third of patients. Although our sample size afforded us 80% power to detect differences of 25% or greater in the pairwise comparison of the rate of dose-limiting toxicity (for example, 30% compared with 55%), the maximum difference observed was 12% (24% of patients receiving dapsone-trimethoprim compared with 36% of patients receiving trimethoprim-sulfamethoxazole). Similarly, we did not detect statistically significant differences in either the proportion of patients completing a full course of therapy for P. carinii pneumonia or the proportion of patients classified as having therapeutic failures. Although ours is the largest comparative trial of these regimens yet conducted, our sample size afforded suboptimal power for detecting differences in therapeutic efficacy. Thus, although we had 80% power to detect pairwise differences in the rate of therapeutic failure of 20% or greater (for example, 5% compared with 25%), we could not exclude potential clinically important differences, albeit smaller in magnitude, among these regimens.

    The overall rate of therapeutic failure in the study was 9%, which compares favorably with rates found in previous studies of patients with mild to moderate P. carinii pneumonia [5, 7, 13, 15, 24]. A propensity for therapeutic failure during the first week of therapy was seen in all three treatment regimens, as noted by previous investigators. Although patients receiving clindamycin-primaquine had the greatest improvement in measures of patient-reported health status scores by day 7, no differences among the groups were seen by day 21. The apparently accelerated improvement in patients receiving clindamycin-primaquine might reflect a slightly lower rate of dose-limiting toxicity by day 7 in this group, which subsequently dissipated (Figure 2), or to a more rapid recovery of functional capacity. Alternatively, this analysis may have been biased (in either direction) by the proportion of persons not responding to the questionnaire. Further investigation is necessary to clarify these possibilities.

    Our results highlight differences in the patterns of toxicity associated with the three regimens. As in previous studies, rash was the most common adverse reaction in all three treatment groups, occurring in 47% of study patients and making up 50% of all dose-limiting toxicities. Although more frequently classified as severe in patients receiving clindamycin-primaquine (6% of trimethoprim-sulfamethoxazole recipients, 2% of dapsone-trimethoprim recipients, and 16% of clindamycin-primaquine recipients; P = 0.07), rash was considered dose limiting as frequently as in patients receiving trimethoprim-sulfamethoxazole (21% compared with 19%). The potential for investigator bias in choosing to discontinue therapy on the basis of rash was probably minimized by the double-blinded nature of the trial. Elevations in serum aminotransferase levels of fivefold or greater were most frequent in the group receiving trimethoprim-sulfamethoxazole (19% of trimethoprim-sulfamethoxazole recipients, 3% of dapsone-trimethoprim recipients, and 7% of clindamycin-primaquine recipients; P = 0.003). In addition, toxicities were distributed throughout the course of therapy in this group; in the other two groups, toxicities occurred primarily during the second week. In the group receiving dapsone-trimethoprim, rates of dose-limiting toxicity (36% of trimethoprim-sulfamethoxazole recipients, 24% of dapsone-trimethoprim recipients, and 33% of clindamycin-primaquine recipients; P = 0.2) and dose-limiting rash (19% of trimethoprim-sulfamethoxazole recipients, 10% of dapsone-trimethoprim recipients, and 21% of clindamycin-primaquine recipients; P = 0.2) were approximately two thirds and one half, respectively, of those in other groups. Serious hematologic toxicities (anemia, neutropenia, or methemoglobinemia) occurred more than twice as often in patients receiving clindamycin-primaquine (11% of trimethoprim-sulfamethoxazole recipients, 12% of dapsone-trimethoprim recipients, and 28% of clindamycin-primaquine recipients; P = 0.01) and made up 12% of all dose-limiting toxicities in this group. It is conceivable that our use of a higher daily dosage of primaquine (30 mg/d base) than that used in many previous studies (15 mg/d base) [6, 11, 14, 16] may have predisposed patients to hematologic toxicities; we cannot use our current data to predict whether their rates would be reduced by decreasing the primaquine dose.

    Factors that could affect the outcome of P. carinii pneumonia include the severity of illness, the use of adjunctive corticosteroids, and a history of episodes of P. carinii pneumonia [25]. In our study, differences in room air PAO (2-PaO)2 were evenly distributed at entry, and adjunctive corticosteroids were prescribed for all patients with a PAO2-PaO2 of 35 to 45 mm Hg [17]. Clindamycin-primaquine recipients had the highest frequency of previous P. carinii pneumonia (14% compared with 10% of trimethoprim-sulfamethoxazole recipients and 6% of trimethoprim-dapsone recipients) and the lowest mean serum lactic acid dehydrogenase level (401 U/L compared with 491 U/L in trimethoprim-sulfamethoxazole recipients and 475 U/L in trimethoprim-dapsone recipients) (Table 1); these factors might conceivably decrease and increase, respectively, the likelihood of therapeutic success in this group [25]. We found no association between the receipt of antipneumocystis prophylaxis in the 30 days before study entry and therapeutic response to that agent.

    Several potential sources of bias should be considered when our findings are interpreted. First, our study design allowed patients to receive agents effective against P. carinii in the 24 hours before randomization; it is conceivable that patients receiving such treatment may have had an improved outcome. However, we found no benefit of the therapy given before randomization: Eleven percent of such patients were subsequently classified as having therapeutic failure compared with 8% of patients who did not receive therapy before randomization (P > 0.2). Second, to most closely reflect clinical reality, we randomly assigned patients who had a history of trimethoprim-sulfamethoxazole intolerance to receive one of the two other regimens; this decision may have caused an inflated rate of intolerance during study drug therapy. The rates of dose-limiting toxicities in this group were similar to or less than the rates in other patients in the study; however, we could not formally analyze cross-allergenicity because of the few patients involved. Third, we established the dosage of both trimethoprim and sulfamethoxazole according to weight; the trimethoprim dosage ranged from 12 to 20 mg/kg daily in the trimethoprim-sulfamethoxazole group and from 12 to 18 mg/kg daily in the dapsone-trimethoprim group. This weight-based dosage closely mimics clinical practice [26]. Although trimethoprim dosages of 20 mg/kg daily in combination with either sulfamethoxazole [2, 3, 5, 8] or dapsone [5, 9] have been shown in previous trials to successfully treat P. carinii pneumonia, accumulating data suggest that these dosages are higher than needed for efficacy and are associated with a high frequency of toxicity [4, 27-29]. Thus, we must consider whether patients receiving higher doses were more prone to toxicity and whether patients receiving lower doses had treatment failure more frequently. We could not detect either association. Given the use of fixed oral formulations of the drugs, we sought to maximize the generalizability of our results while accommodating the greatest range of body weights at study entry and maintaining the double-blinded nature of the trial.

    Our findings suggest that the choice of treatment regimen for a patient presenting with mild to moderate P. carinii pneumonia (that is, PAO2-PaO2 less than equals 45 mm Hg) may ultimately rest more on expected toxicities than on anticipated differences in efficacy. Evidence of hepatic insufficiency at the time of presentation should prompt clinicians to consider a regimen other than trimethoprim-sulfamethoxazole, and severe myelosuppression at baseline may suggest that a regimen other than clindamycin-primaquine be used. Although the above variables do not constitute absolute contraindications, the use of these regimens in the above circumstances should involve close monitoring.

    Previous SectionNext Section

    Appendix

    The following persons participated in the AIDS Clinical Trials Group (in order of enrolled evaluable patients): Harbor-UCLA Medical Center, University of California, Los Angeles: Mario Guerrero, Sally Kruger, and Gildon Beall; University of Cincinnati College of Medicine: Bonnie Jackson, Jill Leonard, and Michael Dohn; University of California, San Francisco: John Stansell, Rowena Mah, and Deborah Harriss; Mount Sinai School of Medicine: Alice Fox and Donna Mildvan; Albert Einstein College of Medicine: Elizabeth Jenny and Carol Harris; Indiana University: Beth Zwickl and Jean Craft; Tulane University: Newton E. Hyslop Jr.; Washington University School of Medicine: William G. Powderly, Mark Meyers, and Michael Royal; Case Western Reserve University: John Carey, Michael Chance, and Susan Birdsong; University of North Carolina: Charles N. van der Horst and David A. Ragan; State University of New York at Stony Brook: Roy T. Steigbigel, Christine Wallace, and Peter Mariuz; University of Rochester: Jane Reid, Donald C. Blair, and Sally P. Klemens; Hershey Medical Center: John Zurlo, W. Christopher Ehmann, and Margaret Kreher; St. Luke's/Roosevelt Hospital: George F. McKinley, Michael H. Grieco, and James O'Connor; Georgetown University: James Lavelle, Phillip F. Pierce, and Sheila Davis; State University of New York at Brooklyn: Keith Chirgwin, W. Alfredo Leon, and Mary Murphy; Northwestern University Medical School and Rush-Presbyterian-St. Luke's Medical Center: Robert Murphy, Constance Benson, and Harold Kessler; Massachusetts General Hospital: Howard M. Heller, Margaret White-Guthro, and Paul E. Sax; New York University: Samuel Yonren, Fred T. Valentine, and Margarita Vasquez; Johns Hopkins University: Charlie Raines, Vivian Rexroad, and Becky Becker; University of Southern California: Francoise Kramer, Claire Hughlett, and John M. Leedom; University of Washington, Seattle: Thomas Hooton, Ann C. Collier, and Lawrence Corey; University of Hawaii: Margo Health-Chiozzi, Debra Ogata-Arakaki, and Sandra Akina; and University of Massachusetts: Patrick G. Fairchild, Thomas C. Greenough, and Sarah H. Cheeseman. Necia Briggs, Richard Hafner, and Jeremy Gradon managed the protocol, and Peter Slasor and Lihua Yang provided statistical assistance.

    Dr. Finkelstein: Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115.

    Dr. Feinberg: University of Cincinatti, Holmes Hospital, Eden and Bethesda Avenues, Cincinatti, OH 45267.

    Dr. Frame: University of Cincinatti Medical Center, PO Box 670405, Cincinatti, OH 45267-0405.

    Dr. Simpson: 979 Denver Place, Ventura, CA 93004.

    Dr. Wu: Johns Hopkins School of Medicine, 624 North Broadway, Baltimore, MD 21205.

    Dr. Cheung: Mount Sinai Medical Center, Box 1042, One Gustav Levy Place, New York, NY 10029.

    Dr. Soeiro: Department of Medicine, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, NY 10461.

    Mr. Hojczyk: Frontier Science, 4033 Maple Road, Amherst, NY 14226.

    Dr. Black: 1633 North Capitol #700, Indianapolis, IN 46202.

    Previous Section
     

    References

    1. 1.↵
      National Center for Infectious Diseases. HIV/AIDS Surveillance Report. Atlanta: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Center for Infectious Diseases, Division of HIV/AIDS: CDC National AIDS Clearinghouse; 1994:1-36.
    2. 2.↵
      Wharton JM, Coleman DL, Wofsy CB, Luce JM, Blumenfeld W, Hadley WK, et al. Trimethoprim-sulfamethoxazole or pentamidine for Pneumocystis carinii pneumonia in the acquired immunodeficiency syndrome. A prospective randomized trial. Ann Intern Med. 1986; 105:37-44.
    3. 3.↵
      Klein NC, Duncanson FP, Lenox TH, Forszpaniak C, Sherer CB, Quentzel H, et al. Trimethoprim-sulfamethoxazole versus pentamidine for Pneumocystis carinii pneumonia in AIDS patients: results of a large prospective randomized treatment trial. AIDS. 1992; 6:301-5.
    4. 4.↵
      Sattler FR, Cowan R, Nielsen DM, Ruskin J. Trimethoprim-sulfamethoxazole compared with pentamidine for treatment of Pneumocystis carinii pneumonia in the acquired immunodeficiency syndrome. A prospective, noncrossover study. Ann Intern Med. 1988; 109:280-7.
    5. 5.↵
      Medina I, Mills J, Leoung G, Hopewell PC, Lee B, Modin G, et al. Oral therapy for Pneumocystis carinii pneumonia in the acquired immunodeficiency syndrome. A controlled trial of trimethoprim-sulfamethoxazole versus trimethoprim-dapsone. N Engl J Med. 1990; 323:776-82.
    6. 6.↵
      Toma E, Fournier S, Dumont M, Bolduc P, Deschamps H. Clindamycin/primaquine versus trimethoprim-sulfamethoxazole as primary therapy for Pneumocystis carinii pneumonia in AIDS: a randomized, double-blind pilot trial. Clin Infect Dis. 1993; 17:178-84.
    7. 7.↵
      Hughes W, Leoung G, Kramer F, Bozzette SA, Safrin S, Frame P, et al. Comparison of atovaquone (566C80) with trimethoprim-sulfamethoxazole to treat Pneumocystis carinii pneumonia in patients with AIDS. N Engl J Med. 1993; 328:1521-7.
    8. 8.↵
      Sattler FR, Frame P, Davis R, Nichols L, Shelton B, Akil B, et al. Trimetrexate with leucovorin versus trimethoprim-sulfamethoxazole for moderate to severe episodes of Pneumocystis carinii pneumonia in patients with AIDS: a prospective, controlled multicenter investigation of the AIDS Clinical Trials Group Protocol 029/031. J Infect Dis. 1994; 170:165-72.
    9. 9.↵
      Leoung GS, Mills J, Hopewell PC, Hughes W, Wofsy C. Dapsone-trimethoprim for Pneumocystis carinii pneumonia in the acquired immunodeficiency syndrome. Ann Intern Med. 1986; 105:45-8.
    10. 10.↵
      Queener SF, Bartlett MS, Richardson JD, Durkin MM, Jay MA, Smith JW. Activity of clindamycin with primaquine against Pneumocystis carinii in vitro and in vivo. Antimicrob Agents Chemother. 1988; 32:807-13.
    11. 11.↵
      Toma E, Poisson M, Fournier M, Morisset R, Phaneuf D, Vega C. Clindamycin with primaquine for Pneumocystis carinii pneumonia. Lancet. 1989; 1:1046-8.
    12. 12.↵
      Kay R, DuBois RE. Clindamycin/primaquine therapy and secondary prophylaxis against Pneumocystis carinii pneumonia in patients with AIDS. South Med J. 1990; 83:403-4.
    13. 13.↵
      Toma E. Clindamycin/primaquine for treatment of Pneumocystis carinii pneumonia in AIDS. Eur J Clin Microbiol Infect Dis. 1991; 10:210-3.
    14. 14.↵
      Ruf B, Rohde I, Pohle HD. Efficacy of clindamycin/primaquine versus trimethoprim/sulfamethoxazole in primary treatment of Pneumocystis carinii pneumonia. Eur J Clin Microbiol Infect Dis. 1991; 10:207-10.
    15. 15.↵
      Black JR, Feinberg J, Murphy RL, Fass RJ, Finkelstein D, Akil B, et al. Clindamycin and primaquine therapy for mild-to-moderate episodes of Pneumocystis carinii pneumonia in patients with AIDS: AIDS Clinical Trials Group 044. Clin Infect Dis. 1994; 18:905-13.
    16. 16.↵
      Toma E, Fournier S, Dumont M, Bolduc P, Deschamps H. Clindamycin/primaquine versus trimethoprim-sulfamethoxazole as primary therapy for Pneumocystis carinii pneumonia in AIDS: a randomized, double-blind trial. Clin Infect Dis. 1993; 17:178-84.
    17. 17.↵
      Consensus statement on the use of corticosteroids as adjunctive therapy for pneumocystis pneumonia in the acquired immunodeficiency syndrome. The National Institutes of Health-University of California Expert Panel for Corticosteroids as Adjunctive Therapy for Pneumocystis Pneumonia. N Engl J Med. 1990; 323:1500-4.
    18. 18.↵
      Hlatky MA, Boineau RE, Higgenbotham MB, Lee KL, Mark DB, Califf RM, et al. A brief self-administered questionnaire to determine functional capacity (the Duke Activity Status Index). Am J Cardiol. 1989; 64:651-4.
    19. 19.↵
      Wu AW, Rubin HR, Mathews WC, Ware JE Jr, Brysk LT, Hardy WD, et al. A health status questionnaire using 30 items from the Medical Outcomes Study. Preliminary validation in persons with early HIV infection. Med Care. 1991; 29:786-98.
    20. 20.↵
      Stewart AL, Hays RD, Ware JE. Health perceptions and energy/fatigue. In: Stewart AL, Ware JE, eds. Measuring Functioning and Well-Being: The Medical Outcomes Study Approach. Durham, NC: Duke Univ Pr; 1992.
    21. 21.↵
      Ries PW. Current estimates from the National Health Interview Survey, United States 1984. Data from the National Health Survey Series 10, No. 156. DHHS publication no. (PHS) 86-1584; Bethesda, MD: U.S. Department of Health and Human Services; 1986.
    22. 22.↵
      Rubin HR, Wu AW, Gutierrez M, Liriano O, Safrin S. Spanish translation of a functional status questionnaire for Pneumocystis carinii pneumonia. VIII International Conference on AIDS/III STD World Congress, Amsterdam, the Netherlands, 1992.
    23. 23.↵
      Laird NM, Ware JH. Random-effects models for longitudinal data. Biometrics. 1982; 38:963-74.
    24. 24.↵
      Safrin S, Lee BL, Sande MA. Adjunctive folinic acid with trimethoprim-sulfamethoxazole for Pneumocystis carinii pneumonia in AIDS patients is associated with an increased risk of therapeutic failure and death. J Infect Dis. 1994; 170:912-7.
    25. 25.↵
      Garay SM, Greene J. Prognostic indicators in the initial presentation of Pneumocystis carinii pneumonia. Chest. 1989; 95:769-72.
    26. 26.↵
      Safrin S. Pneumocystis carinii infection. In: Stein JH, ed. Internal Medicine. 4th ed. St. Louis: Mosby; 1994; 2270-4.
    27. 27.↵
      Stevens RC, Laizure SC, Williams CL, Stein DS. Pharmacokinetics and adverse effects of 20-mg/kg/day trimethoprim and 100-mg/kg/day sulfamethoxazole in healthy adult subjects. Antimicrob Agents Chemother. 1991; 35:1884-90.
    28. 28.↵
      Bowden FJ, Harman PJ, Lucas CR. Serum trimethoprim and sulphamethoxazole levels in AIDS [Letter]. Lancet. 1986; 1:853.
    29. 29.↵
      Lee BL, Medina I, Benowitz NL, Jacob P 3d, Wofsy CB, Mills J 5th. Dapsone, trimethoprim, and sulfamethoxazole plasma levels during treatment of Pneumocystis pneumonia in patients with the acquired immunodeficiency syndrome (AIDS). Evidence of drug interactions. Ann Intern Med. 1989; 110:606-11.
    « Previous | Next Article »Table of Contents

    This Article

    1. Ann Intern Med May 1, 1996 vol. 124 no. 9 792-802
    1. AbstractFREE
    2. Figures
    3. » Full Text
    4. Full Text (PDF)

    Rapid Responses

    1. Submit a response
    2. No responses published

    Navigate This Article

    Current Issue

    1. November 17, 2009, 151 (10)
    1. Alert me to current issues