Toxoplasmic encephalitis has an incidence of 5% to 10% in patients with AIDS in the United States [5] and a higher incidence—15% to 50%—in areas (such as central and southern European countries, including Spain) that have a high seroprevalence in the general population [5, 6]. Patients with an acute episode of toxoplasmic encephalitis are usually given sulfadiazine (4 to 6 g/d) plus pyrimethamine (50 to 75 mg/d) plus folinic acid (15 mg/d) for 6 to 8 weeks. A daily maintenance regimen of sulfadiazine (2 g/d) plus pyrimethamine (25 mg/d) appears to be useful in preventing relapse [7]. A preliminary, open, nonrandomized study [8] has suggested that a twice-weekly regimen of sulfadiazine-pyrimethamine may be useful as maintenance therapy for toxoplasmic encephalitis, although some noncompliant patients in this study relapsed soon after the discontinuation of allocated treatment.

Our objective was to evaluate, in a randomized trial, the efficacy of the intermittent administration of sulfadiazine-pyrimethamine to prevent relapse of toxoplasmic encephalitis in patients with AIDS.

Methods

Top Methods Results Discussion Author & Article Info References

This multicenter, open, randomized study was done between February 1990 and June 1993 at eight university teaching hospitals in Barcelona, Madrid, and Bilbao, Spain.

Patients and Follow-up

We enrolled patients with human immunodeficiency virus (HIV) infection after each had had resolution of an acute episode of toxoplasmic encephalitis treated with sulfadiazine (1 g four times per day) plus pyrimethamine (50 mg/d) plus folinic acid (15 mg/d) for 4 to 8 weeks. Resolution was defined as a greater than 50% reduction of or the disappearance of brain lesions on computed tomography or magnetic nuclear resonance imaging and the amelioration or disappearance of clinical signs (fever, neurologic findings, or both).

Patients treated with clindamycin-pyrimethamine for an initial episode of toxoplasmic encephalitis who did not meet inclusion criteria or whose follow-up was less than 1 month were not considered eligible for the analysis. In addition, drugs with antitoxoplasmic activity, such as cotrimoxazole, clarithromycin, azithromycin, and atovaquone, were not permitted during the study period.

Patients were randomly assigned to receive sulfadiazine (500 mg four times per day) plus pyrimethamine (25 mg/d) plus folinic acid (15 mg/d) either daily or twice weekly. Randomization was done in a central location using a random number table.

Clinical and biological evaluations were done every 30 to 60 days. Biological values evaluated at baseline and during follow-up were blood cell counts, CD4 counts, alanine aminotransferase levels, alkaline phosphatase levels, and creatinine levels. At every examination, compliance and symptoms of toxicity were evaluated, a physical examination was done, and blood samples were obtained for blood cell counts and biochemical parameters. The frequency with which CD4 counts were done during follow-up was decided by the treating physician.

End Points

Primary end points were toxoplasmic encephalitis relapse, death, and interruption of therapy due to adverse effects. Pneumocystis carinii pneumonia was a secondary end point. Toxoplasmic encephalitis was diagnosed by clinical findings (fever, neurologic symptoms, or both); by the appearance of one or more contrast-enhanced focal brain images on computed tomography or magnetic nuclear resonance imaging, or both; and by response to sulfadiazine-pyrimethamine or clindamycin-pyrimethamine therapy. The same criteria were used to evaluate toxoplasmosis relapse. Clinical and radiographic findings suggestive of P. carinii pneumonia (preferably with microbiological confirmation) were required for the diagnosis of P. carinii pneumonia. Inhalation of pentamidine for prophylaxis of P. carinii pneumonia was not a criterion for exclusion, but avoidance of it during the study period was recommended. Only three patients (two receiving the twice-weekly regimen) received inhaled pentamidine during the follow-up period. No other drugs with activity against P. carinii were given during the study period.

Only adverse reactions requiring the discontinuation of therapy (at the discretion of the physician) were recorded for toxicity analysis.

Two independent, blinded investigators evaluated the clinical records, the computed tomographic brain scans or magnetic nuclear resonance images, and the therapeutic responses of persons initially diagnosed as having toxoplasmic encephalitis.

Statistical Analysis

Baseline characteristics were compared using the Student t-test (quantitative variables) or the chi-square or Fisher exact test when necessary (qualitative variables).

To evaluate the efficacy of both regimens, we analyzed all patients who fulfilled the inclusion criteria, were randomly assigned to treatment, and were seen at least once after treatment assignment. Thirteen patients who never received the study medication or who were erroneously assigned to treatment (they did not meet inclusion criteria) were not analyzed. All remaining patients (n = 105) were included in the analysis according to the regimen to which they had been initially assigned, and they were followed until the end of the study even if they switched from their initially allocated treatment or stopped receiving the study medication. We assumed that 95% of patients receiving the daily regimen would remain free of relapse after 1 year of follow-up, and we estimated that almost 75 evaluable patients had to receive each regimen so that we could detect a difference of 15% or more with 80% certainty and a 5% significance level.

We had planned to do only one interim analysis after the first 100 patients were randomly assigned to treatment and to terminate the study if relapse ratio or mortality rate differed significantly (P 0.017; one-tailed test), because before starting the trial we assumed that the daily regimen would be equal to or superior to the twice-weekly regimen. However, on the advice of a blinded, independent safety-monitoring board, we terminated the trial when the P value for the difference was only 0.027.

Relapses and survival were compared using the Mantel-Cox log-rank test of the Kaplan-Meier product-limit estimates. Adjusted risk ratio and its 95% CIs were calculated using the multivariate Cox proportional-hazards method. Covariates used in both analyses were age ( 30 years vs. >30 years), sex, risk groups for HIV infection (drug abusers vs. other risk groups), previous diagnosis of AIDS, P. carinii pneumonia prophylaxis administered before the initial episode of toxoplasmic encephalitis, CD4 counts ( 100 x 10⁶/L vs. >100 x 10⁶/L), baseline number of lesions on computed tomography ( 2 lesions vs. >2 lesions), radiologic sequelae after initial toxoplasmosis episode, antiretroviral therapy during follow-up, and maintenance regimen. Calculations were done using the 1L and 2L routines of the BMDP statistical package (BMDP Statistical Software, Los Angeles, California). All P values were two-sided.

The relapse risk per 100 patient-years and the 95% CIs (Poisson distribution) were calculated using the STATA program (STATA, Santa Monica, California).

Results

Top Methods Results Discussion Author & Article Info References

Patients

Initially, 118 patients were randomly assigned to treatment. Of these, 13 were excluded from the analysis: 9 who were receiving the twice-weekly regimen (5 did not fulfill the inclusion criteria and 4 were not seen after treatment assignment) and 4 who were receiving the daily regimen (2 did not fulfill the inclusion criteria and 2 were not seen after treatment assignment). Sixty patients receiving the daily regimen and 45 patients receiving the twice-weekly regimen were considered eligible for the study. The baseline characteristics of the patients are described in Table 1.

Efficacy

Overall, 11 patients (10.4%) had a second episode of toxoplasmic encephalitis after a median follow-up period of 11 months (range, 1 to 39 months). Ten patients improved with sulfadiazine-pyrimethamine therapy. Only one patient, who presented with neurologic signs and several contrast-enhanced lesions on computed tomography, died a few days after therapy was initiated. An intention-to-treat analysis showed that patients receiving the intermittent regimen had a higher rate of relapse than patients receiving the daily regimen (19.5 per 100 patient-years [CI, 8.4 to 38.4 per 100 patient-years]) compared with 4.4 per 100 patient-years [CI, 0.9 to 13.1 per 100 patient-years]), representing an incidence rate ratio of 4.36 (CI, 1.05 to 25.5; P = 0.024). The estimated cumulative percentages of relapse at 12 months were 30% for patients receiving the twice-weekly regimen and 6% for patients receiving the daily regimen (Mantel-Cox, P = 0.029; Figure 1). Risk ratio, calculated using the multivariate Cox proportional-hazards method after adjusting for age, sex, risk group, previous diagnosis of AIDS, P. carinii pneumonia prophylaxis administered before the initial toxoplasmosis episode, number of lesions on computed tomography at baseline, radiologic sequelae, CD4 counts, and antiretroviral therapy during follow-up, was 5.6 (CI, 1.2 to 25.6; P = 0.028).

View larger version (9K): [in this window] [in a new window]   Figure 1. Kaplan-Meier plots of the time to relapse of toxoplasmic encephalitis. The line with triangles denotes the twice-weekly regimen; the line with circles denotes the daily regimen. The estimated percentages of relapse after 12 months were 30% in the group receiving the twice-weekly regimen and 6% in the group receiving the daily regimen.

If we consider patients only while they were receiving study medication—that is, excluding the five patients with poor compliance and considering the period of follow-up before switching to another regimen in the seven patients who changed therapies—the estimated cumulative percentages of relapse at 12 months were 3% for patients receiving the daily regimen and 17% for patients receiving the intermittent regimen.

Overall, 55 patients died during the follow-up period. No statistical differences in survival rates or in causes of death (data not shown) were found between the two groups. Only 1 patient died of toxoplasmic encephalitis during the study period. The median survival time was 11.9 months in the group receiving the twice-weekly regimen and 15.2 months in the group receiving the daily regimen (Mantel-Cox, P = 0.1022; Figure 2, Table 2). The adjusted risk ratio was 1.6 (CI, 0.9 to 2.9; P = 0.11). Ten patients receiving the daily regimen and 7 patients receiving the twice-weekly regimen had developed P. carinii pneumonia before study entry. No episodes of P. carinii pneumonia occurred during follow-up. Thus, excluding 2 patients who concomitantly received inhaled pentamidine to prevent P. carinii pneumonia, 43 patients who only received twice-weekly therapy with sulfadiazine-pyrimethamine for primary or secondary P. carinii prophylaxis were free of P. carinii after 368 patient-months of follow-up (the upper limit for 97.5% CI was 9.0; see Table 3).

View larger version (10K): [in this window] [in a new window]   Figure 2. Kaplan-Meier plots of the time to death. The line with triangles denotes the twice-weekly regimen; the line with circles denotes the daily regimen. Median survival was 11.9 months in the group receiving the twice-weekly regimen and 15.2 months in the group receiving the daily regimen.

No statistical differences in severe toxicity were seen between groups. Six patients (5.7%) had to discontinue therapy because of adverse effects: Four (6.6%) were receiving the daily regimen (three had skin rash and one had gastric intolerance), and two (4.4%) were receiving the intermittent regimen (both had skin rash) (P = 0.62; Table 2).

Discussion

Top Methods Results Discussion Author & Article Info References

Previous studies [6-11] have shown that as many as 80% of patients who have had an episode of toxoplasmic encephalitis will have relapse without maintenance therapy, whereas daily administration of sulfadiazine-pyrimethamine may decrease the relapse rate to less than 10%.

In our study, only 6% of patients who received the daily regimen of sulfadiazine-pyrimethamine had a relapse of toxoplasmic encephalitis at 12 months; this regimen was statistically more effective than the twice-weekly regimen at the given doses. However, the twice-weekly regimen had some efficacy in preventing relapse. Patients receiving this regimen had a 30% relapse rate at 12 months; the relapse rate reported in patients who do not receive secondary prophylaxis is 50% to 80% [8, 10, 11].

A twice-weekly regimen of sulfadiazine-pyrimethamine may be useful because sometimes (as with cotrimoxazole) patients who cannot tolerate high doses of sulfadiazine because of allergy or gastric disturbances may tolerate lower doses [12]. In addition, patients who do not comply with daily regimens may prefer a twice-weekly regimen, even if it has lower efficacy. Moreover, clindamycin-pyrimethamine, which is the usual alternative to sulfadiazine-pyrimethamine, is associated with a relapse rate of 28%, presents considerable toxicity [13], and does not prevent P. carinii pneumonia [14]. The role of other drugs, including atovaquone and new macrolides, such as azithromycin or clarithromycin, alone or combined with pyrimethamine, remains to be defined [15-18].

Ours is the first randomized study to evaluate the efficacy of an intermittent regimen of sulfadiazine-pyrimethamine to prevent relapse of central nervous system toxoplasmosis in patients with AIDS. In the only article published to date about an intermittent regimen of sulfadiazine-pyrimethamine in secondary prophylaxis of toxoplasmic encephalitis [8], a twice-weekly regimen of sulfadiazine (4 g/d) and pyrimethamine (25 mg/d) seemed to be as effective as a daily regimen of this combination. However, this article reported a nonrandomized, retrospective study of a small number of patients. The investigators used a higher dose of sulfadiazine, but we do not know if this was why their results were apparently better than those obtained in our study, as pyrimethamine is considered the main drug in both the treatment and secondary prophylaxis of toxoplasmosis in patients with HIV infection. On the other hand, a rapid relapse was also seen in 3 (20%) noncompliant patients [8].

As shown in previous studies [19, 20], pyrimethamine levels—not measured in our study—may vary among patients. Perhaps levels were too low in some patients, principally those with poor compliance. Irregular compliance with intermittent prophylactic regimens may more easily lead to suboptimal therapeutic doses than irregular compliance with daily regimens. Periodic questioning of patients as a method for evaluating compliance has important limitations. Thus, caution is necessary when evaluating data about on-treatment analysis. Although five of eight patients who had relapse during intermittent therapy denied having poor compliance, it is difficult to be sure whether they took all or most of their pills; the same is true for patients who did not have relapse. Consequently, we decided to report only on-treatment results in a purely descriptive manner.

Sulfadiazine-pyrimethamine therapy has previously been found to be effective in preventing P. carinii pneumonia [21]. One interesting finding of our study is that even an intermittent regimen of sulfadiazine-pyrimethamine seems to be sufficient to prevent P. carinii pneumonia: No patient developed this opportunistic infection during the study period. The estimated incidence of a first episode of P. carinii pneumonia in patients with a CD4 cell count of less than 200 x 10⁶/L was about 15% to 47% at 12 months [22]. Moreover, at least 60% of patients with an episode of P. carinii pneumonia will have relapse in the following year without maintenance therapy [12]. Although caution must be used when comparing present data with historic controls, we believe that our results suggest the possible efficacy of the twice-weekly regimen in the prevention of this infection. Larger studies are needed to confirm these results.

Our study had some limitations, such as the relatively small number of patients included and the small number of relapses that occurred. However, our sample was large enough to show a statistical difference in favor of the daily regimen. Biases may occur in a nonblinded study; to limit these, we used strict and objective criteria for end point diagnosis, which, in addition, was reviewed by blinded, independent investigators. In summary, the twice-weekly regimen of sulfadiazine-pyrimethamine used in this trial as maintenance therapy for toxoplasmic encephalitis was less effective than the standard daily regimen. However, it could be useful for selected patients, such as those who do not tolerate or are not compliant with daily regimens of pyrimethamine plus sulfadiazine or clindamycin. Moreover, a twice-weekly regimen of sulfadiazine-pyrimethamine might also be useful for P. carinii pneumonia prophylaxis. It should be noted that compliance must be closely monitored when such a regimen is recommended.

An intermittent regimen of sulfadiazine-pyrimethamine using higher doses or more frequent administration (thrice-weekly instead of twice-weekly) may be more effective than that used in our study. A trial comparing thrice-therapy with the standard daily regimen is ongoing.

Appendix

Members of the Spanish Toxoplasmosis Study Group are D. Podzamczer (coordinator), F. Bolao, and F. Gudiol, Ciutat Sanitaria de Bellvitge, Barcelona, Spain; J.M. Miro (coordinator), A. Guelar, and J.M. Gatell, Hospital Clnic, Barcelona, Spain; P. Domingo and J.L. Barrio, Hospital Sant Pau, Barcelona, Spain; G. Sirera and B. Clotet, Hospital Germans Tras, Badalona, Spain; J. Santamara, Hospital de Cruces, Bilbao, Spain; J. Verdejo, Hospital Princesa, Madrid, Spain; F. Laguna and J. Gonzalez-Lahoz, Instituto Carlos III, Madrid, Spain; and J. Cosn, Hospital Gregorio Maranon, Madrid, Spain.

Presented in part at the Tenth International Conference on AIDS, Yokohama, Japan, 7-12 August, 1994 (Abstract PB0630).