PCP Prohylaxis Revised with Update From IDSA Nov12-15

Two studies presented at ICAAC on this subject are of interest. The data of both studies suggest that as a result HAART induced and sustained CD4 increases from below 200 to well above 200 PCP is likely not to occur. The studies have follow-up of only 1 year for those individuals who had never before experienced PCP. In one of the two studies the followup was only 6.6 months for those who had a prior episode of PCP. However, risk is not eliminated as suggested by an experience related by Dr Fred Valentine at the post Geneva Forum conducted by NATAP in July at NYU Med Ctr.

In a retrospective study of 197 patients where 21 stopped prophylaxis and 176 continued prophylaxis, Yangco and colleagues compared the rates of PCP between patients who stopped PCP prophylaxis with those who continued prophylaxis following clinical improvement with HAART. All patients who stopped prophylaxis took the regimen for primary prophylaxis while patients who had prior PCP (secondary) continued PCP prophylaxis even with clinical improvement on HAART. The CD4 nadir (lowest CD4 count reached) of those who discontinued prophylaxis was 105 cells. The median follow-up was 13 months. No episode of PCP occurred for individuals who discontinued PCP prophylaxis or for those who did not discontinue prophylaxis. Those who stopped prophylaxis had a sustained CD4 count of 431 for 11 months, on average.

Schneider and colleagues reported similar results. In contrast to the Yangco study, in this prospective observational study of 73 individuals, primary and secondary prophylaxis was discontinued in patients whose CD4s increased to above 200 cells (documented twice with an interval of at least 1 month) as a result of HAART. So far, no episodes of PCP have occurred after a follow-up of 1 year in those who received primary prophylaxis and 6.6 months in those who received secondary prophylaxis.

 Schneider study


 Primary Proph

 Secondary Proph





 CD4 count at discontinuation




 Time since CD4 count >200 (mos)




 Lowest CD4 count (nadir)




 Patients with undetectable HIV RNA




 Time since HAART (mos)




 Follow-up time (mos)








At the NATAP post Geneva forum, Dr Fred Valentine presented data on one person whose CD4 count had increased from <100 cells to 440 due to HAART when he experienced a mild case of PCP. The person stopped taking prophylaxis. A table of data and further discussion is in article Restoring the Immune System (click here to read it).

I think the data from the two studies reported by Yangco and Schneider are preliminary. Valentine's experience shows there is a risk to discontuing prophylaxis. One year follow-up may be too little upon which to base a decision to discontinue PCP prophylaxis. But a number of doctors in the community are discontinuing prophylaxis for PCP when CD4 increases are adequate. The ACTG is also conducting studies to address this question. As you may know, Bactrim also has additional anti-bacterial effects besides on PCP.

At the recent IDSA meeting in Denver on Nov 12-15, 1998 Mark Dworkin and colleagues from the CDC reported results from their data analysis suggesting that individuals with low CD4 nadirs (0-24) prior to antiretroviral therapy (ART) have a higher risk of developing PCP despite increases of 100 CD4s to above 200 due to ART. All patients in the Dworkin study received antiretroviral therapy.

Dworkin analyzed data from ASD, a national longitudinal medical record surveillance study, and compared two groups-group 1 (n=626) had a CD4 count <200 followed by an increase of 100 CD4 cells or more to above 200 CD4s associated with antiretroviral therapy; group 2 (n=3,473) never had a recorded CD4 count <200 but had been prescribed ART. The analysis of group 1 started only after patients were told they could stop PCP prohylaxis. They controlled for HIV transmission mode, age, sex, race/ethnicity, city, CD4 count, history of AIDS, OI, ART regimen (mono, dual, or triple therapy. Patients with a history of PCP were excluded.

Group 1 (3 PCP cases, 471 person-years) did not have a statisically significant increased risk for PCP compared to group 2 (32 PCP cases, 4,234 person-years).

However, Dworkin reports there may be an increased risk for developing PCP if a person's lowest CD4 (nadir) was very low, although his study results are not statistically significant. You will note in the table below that Dworkin's data suggests that individuals with a CD4 nadir of 25-199 had less of a chance of developing PCP (hazard ratio 0.5) than individuals in group 2. This may be because the individuals with CD4 nadir 25-199 were obviously a group who their doctor felt had a good response to ART including presumably a good viral load response. They may have had a CD4 increase to 300 or more CD4s.

Influence of CD4 Nadir on Risk for PCP above 200 CD4 Cells Following an increase of at least 100 CD4 to above 200 CD4 due to ART

 CD4 Nadir

 PCP cases/person yrs

 Incidence, 95% CI*

 Hazard Ratio (increased risk) 95% CI*



 4.4 (0-12.9)

 3.7 (0.5-30.0)



 0.5 (0-1.1)

 0.5 (0.1-2.0)

* Group 2 is the comparison (32 PCP/4,234 person years); in other words, a person with a CD4 nadir of 0-24 cells in group 1 has a 3.7 increased risk of developing PCP than individuals in group 2. CI= confidence intervals.