|
|
| |||||||||||||||||||||||||||||||||||||||||
Background Although trimethoprim–sulfamethoxazole is the drug of choice for the prevention of Pneumocystis carinii pneumonia, many patients cannot tolerate it and must switch to an alternative agent.
Methods We conducted a multicenter, open-label, randomized trial comparing daily atovaquone (1500-mg suspension) with daily dapsone (100 mg) for the prevention of P. carinii pneumonia among patients infected with the human immunodeficiency virus who could not tolerate trimethoprim–sulfamethoxazole. The median follow-up period was 27 months.
Results Of 1057 patients enrolled, 298 had a history of P. carinii pneumonia. P. carinii pneumonia developed in 122 of 536 patients assigned to atovaquone (15.7 cases per 100 person-years), as compared with 135 of 521 in the dapsone group (18.4 cases per 100 person-years; relative risk for atovaquone vs. dapsone, 0.85; 95 percent confidence interval, 0.67 to 1.09; P=0.20). The relative risk of death was 1.07 (95 percent confidence interval, 0.89 to 1.30; P=0.45), and the relative risk of discontinuation of the assigned medication because of adverse events was 0.94 (95 percent confidence interval, 0.74 to 1.19; P=0.59). Among the 546 patients who were receiving dapsone at base line, the relative risk of discontinuation because of adverse events was 3.78 for atovaquone as compared with dapsone (95 percent confidence interval, 2.37 to 6.01; P<0.001); among those not receiving dapsone at base line, it was 0.42 (95 percent confidence interval, 0.30 to 0.58; P<0.001).
Conclusions Among patients who cannot tolerate trimethoprim–sulfamethoxazole, atovaquone and dapsone are similarly effective for the prevention of P. carinii pneumonia. Our results support the continuation of dapsone prophylaxis among patients who are already receiving it. However, among those not receiving dapsone, atovaquone is better tolerated and may be the preferred choice for prophylaxis against P. carinii pneumonia.
Atovaquone, an agent with antipneumocystis activity in studies in animals,8 has been shown to be effective and reasonably well tolerated for the treatment of mild-to-moderate P. carinii pneumonia in HIV-infected patients.9 However, there have been no large studies of the efficacy and tolerability of atovaquone for the prevention of P. carinii pneumonia.
The Community Program for Clinical Research on AIDS (CPCRA) and the AIDS Clinical Trials Group (ACTG) initiated a randomized study comparing atovaquone suspension with dapsone for the prevention of P. carinii pneumonia among HIV-infected patients who could not tolerate trimethoprim, sulfonamides, or both.
Methods
Study Population
HIV-infected patients 13 years of age or older were eligible for the study if they had a history of P. carinii pneumonia (this group received secondary prophylaxis), or if they had no history of P. carinii pneumonia and had a CD4+ lymphocyte count no higher than 200 per cubic millimeter or no more than 15 percent of the total lymphocyte count (this group received primary prophylaxis). Patients were also required to have a history of a treatment-limiting reaction to sulfonamides or trimethoprim, no history of intolerance of dapsone or atovaquone, and adequate glucose-6-phosphate dehydrogenase levels.
Study Design
The study was a randomized, open-label clinical trial comparing the efficacy of atovaquone (Mepron, Glaxo Wellcome, Research Triangle Park, N.C.; given as a 1500-mg suspension once daily) and dapsone (Jacobus Pharmaceuticals, Princeton, N.J.; 100-mg tablets, given once daily) in delaying or preventing the onset of confirmed or probable P. carinii pneumonia. Patients randomly assigned to receive dapsone who had CD4+ cell counts under 100 per cubic millimeter and positive results on serologic tests for toxoplasma were encouraged to take pyrimethamine (50 mg) and leucovorin (15 mg) each week. Patients in whom P. carinii pneumonia developed or who had an adverse event necessitating the discontinuation of the assigned treatment could be switched to the other study treatment. Patients were randomly assigned to treatment in a 1:1 ratio according to a permuted-block design, with stratification according to the clinical center, history with respect to P. carinii pneumonia, and serologic status with respect to toxoplasma.
The sample size and duration of follow-up (18 months from the time the last patient was enrolled) were calculated to provide the study with 80 percent power to detect a 50 percent decrease in the risk of P. carinii pneumonia among the patients assigned to atovaquone, as compared with those assigned to dapsone, at a 0.05 level of significance (in a two-sided test). It was assumed that P. carinii pneumonia would develop in 20 percent of the patients receiving dapsone for primary prophylaxis and in 40 percent of those receiving dapsone for secondary prophylaxis. It was also assumed that 80 percent of the participants would receive primary prophylaxis, and 20 percent secondary prophylaxis. Overall mortality at two years in the absence of P. carinii pneumonia was assumed to be 20 percent and 40 percent for those receiving primary and secondary prophylaxis, respectively. The study was approved by each medical center's institutional review board, and the participants were required to provide written informed consent in order to participate.
End Points and Follow-Up
The primary end point was the development of confirmed or probable P. carinii pneumonia. Confirmation required histologic or cytologic identification of P. carinii in bronchoalveolar-lavage, lung-biopsy, or sputum specimens. A probable diagnosis was based on a history of either exertional dyspnea or nonproductive cough; evidence of diffuse bilateral pulmonary disease on either a chest roentgenogram or a gallium scan, or abnormal arterial-blood gases; and the absence of another condition that could account for the findings. The end points were reviewed by a clinical-events committee whose members were unaware of the patients' treatment assignments. The secondary end points were intolerance of the study drug necessitating the permanent discontinuation of the drug, death, the combined end point of P. carinii pneumonia or death, and confirmed or probable toxoplasmosis. A four-point scale was used to grade the severity of adverse events (grade 4 events were severe and potentially life-threatening).
Statistical Analysis
Investigators were unaware of all interim results, which were reviewed by an independent data and safety monitoring board. The treatment groups were compared by means of Kaplan–Meier event–time curves, log-rank tests, and proportional-hazards regression models.10 Primary analyses were performed according to the intention-to-treat principle and were stratified according to the trial group (CPCRA or ACTG), history with respect to P. carinii pneumonia, and results of serologic tests for toxoplasma. Relative risks (atovaquone vs. dapsone) are given with 95 percent confidence intervals. Secondary analyses were also performed. In an "on-treatment" analysis, follow-up data were censored 30 days after the assigned study treatment was discontinued. In a second analysis, which was conducted because of the low rate of P. carinii pneumonia during the later follow-up period, follow-up data were censored when therapy was initiated with protease inhibitors. The proportional-hazards model was used to estimate the risk of P. carinii pneumonia for patients who discontinued the study treatment as compared with those who did not. In this analysis, a single time-varying indicator of discontinuation of all study treatment was included in the regression model. All P values are two-tailed; all median times reported account for censoring.
Results
Characteristics of the Patients
Between October 4, 1994, and May 5, 1995, 1057 patients were enrolled in the study, 194 from 11 CPCRA units and 863 from 37 ACTG units. The two treatment groups were well balanced with regard to base-line characteristics (Table 1). Overall, 21.9 percent were black, 12.1 percent were Hispanic, 17.7 percent were injection-drug users, and 12.4 percent were female. A history of P. carinii pneumonia was reported by 298 patients (28.2 percent). Overall, the median base-line CD4+ lymphocyte count was 60 per cubic millimeter, and 16.1 percent of the patients had positive serologic tests for toxoplasma. At base line, 73.4 percent of the patients were receiving prophylaxis against P. carinii pneumonia (51.7 percent were receiving dapsone, 18.5 percent pentamidine, and 3.2 percent other treatments or a combination of treatments), and 58.6 percent of the patients were receiving antiretroviral treatment (46.9 percent were receiving nucleoside monotherapy, and 11.7 percent nucleoside combination therapy). At the 12-month follow-up visit, 70.5 percent of the patients were receiving antiretroviral therapy, with 31.3 percent receiving combination nucleoside therapy. At the 24-month follow-up visit, 93.0 percent of the patients were receiving antiretroviral therapy, with 17.2 percent taking combination nucleoside regimens and 71.9 percent (71.7 percent of the atovaquone group and 72.0 percent of the dapsone group) using regimens containing protease inhibitors.
|
At the end of the study, on April 11, 1997, the median follow-up was 27 months for each treatment group. At this time, the status with respect to P. carinii pneumonia was unknown for 5.0 percent of patients assigned to atovaquone and 7.1 percent of those assigned to dapsone. The vital status of 5.0 percent of the patients in the atovaquone group and 6.7 percent of those in the dapsone group was unknown.
Of the 617 patients who were alive at the close of the study, 289 (47 percent) were receiving study-supplied dapsone or atovaquone: 214 (35 percent) were still taking the originally assigned drug, and 75 (12 percent) had been switched to the alternative treatment. The median time during which patients took the assigned drug was 7.2 months for atovaquone and 7.4 months for dapsone. The median time during which they took either the assigned drug or the alternative was 10.8 months for those assigned to atovaquone and 11.9 months for those assigned to dapsone (P=0.22).
P. carinii Pneumonia
A total of 257 patients had at least one episode of confirmed or probable P. carinii pneumonia: 122 in the atovaquone group (15.7 cases per 100 person-years) and 135 in the dapsone group (18.4 per 100 person-years; relative risk for atovaquone vs. dapsone, 0.85; 95 percent confidence interval, 0.67 to 1.09; P=0.20) (Table 2). The cumulative percentages of patients who had had at least one episode of confirmed or probable P. carinii pneumonia after 6, 12, and 24 months were 8.9 percent, 19.5 percent, and 26.5 percent for patients assigned to atovaquone and 13.5 percent, 21.3 percent, and 29.5 percent for patients assigned to dapsone (Figure 1). In both treatment groups, the risk of P. carinii pneumonia declined after the first year of follow-up. The relative risk of P. carinii pneumonia in the two groups did not differ significantly when data were censored at the time of initiation of therapy with protease inhibitors (relative risk, 0.84; P=0.19).
|
|
Among patients who were receiving their originally assigned treatment or who had discontinued or switched treatment not more than 30 days previously (on-treatment analysis), P. carinii pneumonia developed in 83 patients assigned to atovaquone (17.0 cases per 100 person-years) and in 101 patients assigned to dapsone (20.8 per 100 person-years; relative risk, 0.80; 95 percent confidence interval, 0.59 to 1.07; P=0.13). Patients who discontinued all study treatment had a 37 percent lower rate of P. carinii pneumonia than those who continued to receive the study-supplied medication (P=0.02).
Toxoplasmosis
Confirmed or probable toxoplasmosis developed in 7 of the 170 patients who were seropositive for toxoplasma at base line (4 assigned to atovaquone and 3 assigned to dapsone) (relative risk, 1.18; 95 percent confidence interval, 0.26 to 5.30; P=0.83).
Mortality
A total of 440 patients died during the study. Death rates were similar for the two treatment groups; there were 232 deaths (26.2 per 100 person-years) among patients assigned to atovaquone and 208 (24.1 per 100 person-years) among those assigned to dapsone (relative risk, 1.07; 95 percent confidence interval, 0.89 to 1.30; P=0.45) (Table 2).
P. carinii Pneumonia or Death
A total of 555 patients had confirmed or probable P. carinii pneumonia or died: 281 assigned to atovaquone (36.1 events per 100 person-years) and 274 assigned to dapsone (37.0 per 100 person-years; relative risk, 0.98; 95 percent confidence interval, 0.83 to 1.16; P=0.80).
Subgroup Analyses
Among patients receiving primary prophylaxis, the relative risk of P. carinii pneumonia for atovaquone versus dapsone was 0.81 (P=0.20), and the relative risk of death was 1.25 (P=0.07). Among patients receiving secondary prophylaxis, the relative risk was 0.92 (P=0.65) for P. carinii pneumonia and 0.85 (P=0.28) for death (Table 2). The relative risk of death was significantly different in the subgroup of patients receiving primary prophylaxis and the subgroup receiving secondary prophylaxis (P=0.03). The relative risk of P. carinii pneumonia in the atovaquone group versus the dapsone group was 0.99 among patients already receiving dapsone at base line (P=0.95) and 0.68 among those not receiving dapsone at base line (P=0.05).
Adverse Events and Discontinuation of Study Drugs
Among the patients assigned to atovaquone, 436 (81 percent) discontinued the study medication, including 146 who were switched to dapsone. The most common reasons for discontinuation were intolerance (25 percent), the patient's request (18 percent), development of P. carinii pneumonia (16 percent), and death (14 percent). Among the patients assigned to dapsone, 407 (78 percent) discontinued the medication, including 149 who were switched to atovaquone. For these patients, the most common reasons for discontinuation were intolerance (26 percent), development of P. carinii pneumonia (17 percent), death (17 percent), and the patient's request (9 percent).
The overall rates of discontinuation because of an inability to tolerate the study drug did not differ significantly between treatment groups (relative risk for atovaquone vs. dapsone, 0.94; P=0.59) (Table 3). Among patients who discontinued atovaquone because of intolerance, 74 percent had grade 1 or 2 adverse events, and 25 percent had grade 3 or 4 adverse events. In the dapsone group, 63 percent of the patients who discontinued the drug because of intolerance had grade 1 or 2 adverse events, and 36 percent had grade 3 or 4 adverse events. (Grades were not available for three patients.) The specific adverse events varied according to the treatment group. As compared with patients assigned to dapsone, more patients assigned to atovaquone discontinued the drug because of upper gastrointestinal symptoms (relative risk, 6.79; P<0.001) or diarrhea (relative risk, 16.6; P=0.01). However, fewer patients assigned to atovaquone discontinued the study drug because of hypersensitivity reactions, such as rash, fever, allergic reaction, pruritus, and dermatitis (relative risk, 0.50; P<0.001), or anemia (0 vs. 2.1 per 100 person-years; P=0.001).
|
A significantly larger number of patients assigned to atovaquone requested that the study medication be discontinued in the absence of an adverse event or P. carinii pneumonia (Table 3). However, the proportion of patients who discontinued the study drug either at their own request or because of intolerance did not differ significantly between the atovaquone and dapsone groups (relative risk, 1.21; P=0.05). The risk of discontinuation for either of these reasons was significantly influenced by base-line use of dapsone (Table 3).
The median time during which patients received the study medication was similar in the two treatment groups but differed significantly according to dapsone use at base line (Figure 2). The median time that patients continued to use assigned dapsone was 10.9 months for those who were receiving dapsone at base line and 3.9 months for those who were not (P<0.001). The median time that patients continued to use atovaquone was similar in these two subgroups: 7.0 months for those who were receiving dapsone at base line and 7.4 months for those who were not.
|
In patients with HIV infection, the use of trimethoprim–sulfamethoxazole as prophylactic treatment has been limited by adverse events.6,7 Various regimens for desensitization to trimethoprim–sulfamethoxazole have been recommended,11 and dose escalation has been attempted.12,13 These strategies may be inadvisable or unsuccessful in some patients, necessitating the use of alternative agents, including dapsone, dapsone plus pyrimethamine, and aerosolized pentamidine.14 In this study, atovaquone suspension and dapsone at a 100-mg daily dose were similarly effective for the prevention of P. carinii pneumonia among patients who could not tolerate trimethoprim–sulfamethoxazole. This finding was consistent for both patients who were receiving primary prophylaxis and those who were receiving secondary prophylaxis and for both patients who were receiving dapsone at base line and those who were not.
The efficacy of dapsone as prophylaxis against P. carinii pneumonia has been evaluated in several studies that used various doses of dapsone, with or without pyrimethamine.6,7 With the exception of one small study, none of these studies were specifically designed to enroll patients who could not tolerate trimethoprim–sulfamethoxazole.15
A potential advantage of atovaquone for the prevention of P. carinii pneumonia is its antitoxoplasma activity.14 Although we found that atovaquone and dapsone had similar efficacy for the prevention of P. carinii pneumonia, we were unable to evaluate the efficacy of atovaquone for the prevention of toxoplasmosis because of the low seroprevalence of toxoplasma among the study participants and the occurrence of very few cases of toxoplasmosis.
The rates of P. carinii pneumonia decreased with increasing duration of follow-up in both treatment groups, perhaps as a result of the introduction of more potent antiretroviral therapies during the course of the study. (Seventy-two percent of the participants at two years of follow-up were receiving regimens containing protease inhibitors.) There was also a substantial decline in the rates of HIV-related opportunistic events during this time.16,17 On the other hand, it is not clear why there was a greater decline in the rate of P. carinii pneumonia among the patients who discontinued the study medications than among those who continued to receive them. The regimens used as prophylaxis against P. carinii pneumonia (largely aerosolized pentamidine and dapsone) by patients who discontinued the assigned treatment are unlikely to explain the lower rate of P. carinii pneumonia. It is also unlikely that patients who discontinued the study medications were those with a more favorable response to antiretroviral therapy and a lower risk of P. carinii pneumonia. It is possible that those who discontinued treatment and became lost to follow-up (about 6 percent of the patients) had a higher rate of P. carinii pneumonia. The similar rate of loss to follow-up in the two treatment groups reduces the likelihood of a bias in the treatment effects.
A substantial proportion of participants had adverse events that necessitated discontinuation of the study drugs. Overall, there was no significant difference between the atovaquone and dapsone groups in the rate of adverse events or in the median time during which patients received the assigned treatment. However, dapsone use before randomization was an important determinant of the relative tolerance of these two regimens. Patients who were receiving dapsone at base line were randomly assigned to either the continued use of dapsone or the initiation of atovaquone. Because few patients were receiving atovaquone before randomization, a comparable group of patients receiving atovaquone at base line could not be studied. Patients who were receiving dapsone at base line were more likely to tolerate the continued use of dapsone than the initiation of atovaquone. The opposite was true for those not receiving dapsone at base line: they tolerated atovaquone better than dapsone. The results for the patients in the latter group, who did not already have a demonstrated tolerance of dapsone, may be more applicable to patients who cannot tolerate trimethoprim–sulfamethoxazole and who are therefore newly eligible for alternative prophylactic regimens.
Each treatment was associated with distinct patterns of adverse events. With atovaquone, there were significantly higher rates of upper gastrointestinal symptoms and diarrhea among the patients who were receiving dapsone at base line. With dapsone, hypersensitivity reactions and anemia predominated.
Survival rates were similar whether patients were assigned to atovaquone or to dapsone. A previous study reported higher mortality among patients receiving dapsone than among those receiving aerosolized pentamidine for secondary prophylaxis against P. carinii pneumonia.18 However, this result was not confirmed in our study or in a recent meta-analysis.19
In conclusion, this large randomized study compared atovaquone with the most widely used alternative, dapsone, as prophylaxis against P. carinii pneumonia in HIV-infected patients with a history of intolerance of trimethoprim–sulfamethoxazole. The rates of P. carinii pneumonia, survival, and tolerance were similar in the atovaquone and dapsone groups. However, among patients who were not receiving dapsone at base line, atovaquone was better tolerated, and it may be the preferred therapy for such patients. Patients who were already receiving dapsone tolerated it better than atovaquone, a result indicating that such patients should continue to take dapsone. The high rates of P. carinii pneumonia and intolerance with both medications suggest that further research is needed to find better options for prophylaxis against P. carinii pneumonia among patients who cannot tolerate trimethoprim–sulfamethoxazole.
Supported in part by the CPCRA (funded by the National Institute of Allergy and Infectious Diseases) and the ACTG (grant no. AI38858) and by the General Clinical Research Center Units funded by the National Center for Research Resources (grant no. MO1RR0048).
We are indebted to the patients for their participation in this study and to the following for their contributions: James Neaton, Ph.D., Kinley Larntz, Ph.D., Janice Walker, R.N., M.N., Ana Martinez, R.PH., Mira Madans, David Jacobus, M.D., Michael Rogers, Ph.D., Jeffrey Chulay, M.D., Marjorie Dehlinger, D.N.Sc., Lawrence Deyton, M.S.P.H., M.D., Mari-Joy Morrison, R.N., Frederick Nunley, Jerome Ernst, M.D., Charles Farthing, M.D., Dianne Finkelstein, Ph.D., and Jill Ramser, R.N., M.Ed. We are also indebted to Glaxo Wellcome for providing the atovaquone suspension and to Jacobus Pharmaceuticals for providing dapsone for this study.
Source Information
From the Harlem Hospital Center and Columbia University College of Physicians and Surgeons, New York (W.M.E.); Northwestern University, Chicago (R.L.M.); the Division of Biostatistics, School of Public Health (T.M.Y.), and the School of Medicine (H.H.B.), University of Minnesota, Minneapolis; Saint Joseph's Medical Center, Chicago (R.L.-H.); Mount Sinai Medical Center, New York (T.W.C.); East Orange Veterans Affairs Hospital, East Orange, N.J. (R.E.); the University of Washington, Seattle (T.M.H.); the Medical College of Virginia, Virginia Commonwealth University, Richmond (T.M.K.); Illinois Masonic Medical Center, Chicago (M.S.); the University of North Carolina, Chapel Hill (C.H.); and the Division of AIDS, National Institutes of Allergy and Infectious Diseases, Bethesda, Md. (R.H.).
Address reprint requests to Dr. El-Sadr at the Division of Infectious Diseases, Rm. 3107, Harlem Hospital Center, 506 Lenox Ave., New York, NY 10037.
References
The collaborating sites and investigators participating in this study were as follows. CPCRA units — AIDS Research Alliance, Chicago: D. Moore, R. Krupa, and R. Verheggen; AIDS Research Consortium of Atlanta: L. Mahon, H. Bohn, and M. Thompson; Community Consortium of San Francisco: J. Pannell, C. Brosgart, and F. Strauss; Harlem AIDS Treatment Group, New York: J. Wennberg, S. Caras, and C. Guity; Henry Ford Hospital, Detroit: L. Faber, L. Makohon, and D. Mastro-Polak; Louisiana Community AIDS Program, New Orleans: C.L. Besch, Sr., S. Pablovich, and S. Leblanc; North Jersey Community Research Initiative, Newark, N.J.: C. Forrester, N. Regevik, and R. Nahass; Richmond AIDS Consortium, Richmond, Va.: T.M. Kerkering, C. Webster, and J. Brooks; Research and Education Group, Portland, Oreg.: R. Lusk, C. Salveson, and J.H. Sampson; Washington Regional AIDS Program, Washington, D.C.: F. Nunley, K. Shade, and B. Staton; Wayne State University, Detroit: L. Crane, R. MacArthur, and G. Deisinger. ACTG units — Case Western Reserve University, Cleveland: R. Kalayjian, M. Chance, and N. Frantz; Georgetown University, Washington, D.C.: P. Pierce, P. Kumar, and J. Timpone; Harvard Medical School and Massachusetts General Hospital, Boston: D. Joyce-Otis; Boston Medical Center, Boston: D. Ives; Harvard Medical School and Beth Israel Deaconess Medical Center, Boston: H. Heller; Howard University, Washington, D.C.: R. Delapenha, J. McNeil, and Y. Butler; Indiana University Hospital, Indianapolis: J. Black, S. Ryan, and H. Nixon; Johns Hopkins University, Baltimore: M. Higgins, P. Brooks, and J. Feinberg; Mount Sinai Medical Center, New York: D. Mildvan, B. Simpson, and V. Koslowski; New York University, New York: V. Rosenwald and R. Gulick; Northwestern University, Evanston, Ill.: P. Donath, N. French, and B. Sha; Ohio State University, Columbus: R.J. Fass, C.O. Mills, and J.L. Neidig; Stanford University, Palo Alto, Calif.: T.C. Merigan, Jr., D. Israelski, and D. Slamowitz; University of Alabama at Birmingham: L. Alldredge, S. Wells, and K.E. Squires; UCLA School of Medicine: G. Mathisen and G. Beall; Harbor–UCLA Medical Center: M. Guerrero; University of California, San Diego: R. Haubrich, R. Snyder, and D. Richman; University of California, San Francisco: E. San Juan, H. Woodring, and D. Gary; University of Cincinnati, Cincinnati: M. Dohn, J. Black, and J. Sanchez; University of Colorado Health Sciences Center, Denver: G. Ray, B. Putnam, and S.B. Canmann; University of Hawaii, Honolulu: C.M. Shikuman, M. Millard, and S. Souza; University of Miami School of Medicine, Miami: D.T. Jayaweera, J. Cole, and L. Thompson; University of Minnesota, Minneapolis: J.T. Stapleton, S. Swindells, and C. Kumekawa; University of North Carolina, Chapel Hill: D. Ragan and B. Longmire; University of Pennsylvania, Philadelphia: S. Hauptman, R.R. MacGregor, and E. Donaghy; University of Rochester Medical Center, Rochester, N.Y.: R. Reichman, R. Hewitt, and D. Blair; University of Southern California, Los Angeles: M.P. Dube, V.T. Clemente, and J.M. Leedom; University of Texas, Galveston: R.B. Pollard, M.J. Borucki, and A. Caballero; University of Washington, Seattle: B.A. Royer, J. Lund, and A.C. Collier; Vanderbilt University, Nashville: M.A. South, S. Raffanti, and D. Haas; Washington University, St. Louis: W.G. Powderly, C.J. Fichtenbaum, and G. Hamilton; Yale University, New Haven, Conn.: G. Friedland, E. Cooney, and M.D. McDonough.
| |||||||||||||||||||||||||||||||||||||||||
Related Letters:
Atovaquone Compared with Dapsone to Prevent Pneumocystis carinii Pneumonia
Horowitz H. W., Wormser G. P., El-Sadr W.
Extract |
Full Text
N Engl J Med 1999;
340:1512-1513, May 13, 1999.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
