OI Update: Pneumocystis carinii Pneumonia

The Incidence Is Down, the Treatments Are Better, But the Threat Remains

Maximally suppressive antiretroviral therapy has revolutionized the treatment of HIV disease. Patients who respond well to the potent multidrug regimens that are now standard therapy for individuals with advanced HIV infection not only achieve clinical stability, they seem to experience some degree of immune reconstitution. We now know that the initiation of highly suppressive therapy leads to a dramatic rise in CD4 count -- an increase that results from peripheral expansion of pre-existing lymphocytes. This rapid expansion is followed, after a year or more of therapy, by a gradual, selective rise in naïve T-cells.

What we do not yet know is how much of the immune system's old memory repertoire is regained in this process, but the evidence at hand suggests that some immune function is regained. This would explain the drastic reduction we have seen in the incidence of AIDS-related opportunistic infections in patients with late-stage HIV disease. Although OI rates began to drop before the advent of the protease inhibitors -- thanks to a combination of earlier intervention, more effective prophylaxis, combination nucleoside therapy, and improved clinical care -- truly dramatic reductions occurred only after the widespread adoption of the protease-inhibitor-containing therapies.

Overall, the incidence of OIs has dropped a stunning 70% since the protease inhibitors became widely available two and a half years ago. Evidence continues to accumulate that maximally suppressive antiretroviral therapy has a significant impact on AIDS-related morbidity. There have been multiple anecdotal reports that such therapy has a positive therapeutic impact on previously untreatable OIs like progressive multifocal leukoencephalopathy, cryptosporidiosis, and microsporidiosis (see "Regression of PML in an aggressively treated patient," Vol. 3, No. 2). In the February 1998 issue of HIV Newsline we reported that highly active antiretroviral therapy -- combined with a year-long course of antimycobacterial agents -- can lead to the apparent eradication of disseminated MAC (see "HAART therapy leads to resolution of MAC infection," Vol. 4, No. 1).

That's the good news. The bad news is that OIs continue to pose a threat to individuals with advanced HIV disease. There have, for example, been reports of fulminant CMV retinitis in patients whose CD4 counts had risen, as a result of highly active antiretroviral therapy, well above the supposed threshold level for breakthrough infections. To keep clinicians abreast of new developments in the treatment of these old threats to their patients, the editors of HIV Newsline are inaugurating "OI Update," a series of succinct and timely reports on the current standard of care for a wide range of opportunistic infections, beginning with the best-known OI of all -- PCP.

-- The Editors

Pneumocystis carinii pneumonia -- once the most common presenting symptom in HIV-infected patients and the leading cause of AIDS-related deaths -- continues to be a significant cause of morbidity and mortality in patients with AIDS despite the availability of prophylaxis and highly active antiretroviral therapy.

Previously categorized as a protozoan and now classified as a fungus, P. carinii can be isolated from the environment by means of a spore trap and identified by polymerase chain-reaction assay. Since pneumocystis cannot be cultivated, it is not known if the forms identified by PCR are infectious. Molecular typing suggests that repeated episodes of PCP are probably related to acquisition of different strains rather than to reactivation of an existing infection.

Pneumocystis is known to be host-species-specific, which suggests that human infection is unlikely to be a zoonosis. It is not known if PCP can be transmitted from person to person, but because the likelihood of nosocomial infection is low, respiratory isolation is not mandated for patients with PCP when they are hospitalized.

PCP is typically diagnosed by visualization of the organisms in clinical samples of induced sputum and/or bronchoalveolar lavage using immunofluorescence or Giemsa, toluidine blue, or Gomori methenamine silver stains (Figure). PCR assays to detect both single-copy genes and multiple-copy genes are currently under development. To date, the results of these assays have been disappointing, with sensitivities varying from zero to 100% in different studies of the same assay.

The fungus that causes P. carinii pneumonia in patients with advanced HIV disease can be diagnosed by visualization of the organisms in clinical samples of induced sputum and/or bronchoalveolar lavage using immunofluorescence or one of these stains: Giemsa, toluidine blue, or Gomori methenamine silver. When the latter method is used, the P. carinii particles show up as dark, ovoid shapes against a blue-green background.

Current treatment recommendations

Trimethoprim-sulfamethoxazole (Septra®) -- dosed as TMP 5 mg/kg IV q 8 hours for 21 days -- remains the preferred treatment for PCP. Mild cases of pneumocystis may be treated on an outpatient basis with two double-strength tablets of TMP-SMX three times a day. Alternative therapies are listed in Table 1. There are no studies evaluating the use of combination therapy in patients failing to respond to TMP/SMX. When patients develop progressive disease while on therapy, the explanation is likely to be some combination of host immunity, underlying pulmonary dysfunction, and severity of disease, not drug resistance.

Table 1. Treatment for Acute PCP
First-Line Therapy:
     (dosed as TMP 5 mg/kg IV q 8 hours)
     for 21 days
Alternative Therapies:
     Pentamidine (3-4 mg/kg/day IV)
     TMP (5 mg/kg q 8 hours)
     plus dapsone (100 mg daily)
     Clindamycin (600 mg IV/po q 8 hours)
     plus primaquine (30 mg daily)
     Atovaquone suspension
     (750 mg po b.i.d. for mild-to-moderate PCP)
     Trimetrexate (45 mg/m2 IV daily)
     plus leucovorin (20 mg/m2 po q 6 hours)

Corticosteroids should be given to patients with PCP who are hypoxic (i.e. have a pO2 less than 70 mm Hg or an alveolar-arterial pO2 difference exceeding 35 mm Hg). The recommended treatment is 40 mg of prednisone b.i.d. for five days, then 40 mg daily for five days, then 20 mg daily for 11 days. The use of adjunctive corticosteroids for the treatment of PCP does not increase mortality or the risk of common HIV-associated complications.

Prophylaxis for PCP should be initiated for all HIV-infected individuals who have CD4 counts below 200 cells/mm3 and for all patients with unexplained fever (i.e. temperature exceeding 100.0 F or 37.70 C) for longer than two weeks, a history of oral candidiasis, or a previous episode of PCP. Although the majority of individuals who develop PCP are unaware of their HIV status at the time of diagnosis, approximately 15% of PCP cases occur in patients whose serostatus is known but who do not meet the standard criteria for prophylaxis. This suggests that clinicians should interpret the current prophylaxis guidelines conservatively, and err on the side of caution when prescribing preventive therapy -- even in patients who respond well to combination antiretroviral therapy.

TMP/SMX is the preferred prophylactic agent, at a dose of one double-strength tablet daily. In addition to its proven efficacy and low cost, TMP/SMX reduces the incidence of toxoplasmosis and bacterial infections. Single-strength tablets may also be effective, and they may be better tolerated.

As all clinicians who treat HIV-infected patients know only too well, adverse reactions to TMP-SMX -- principally severe rash -- are seen in a significant percentage of the individuals treated with this sulfa-based drug. In some studies of PCP prophylaxis, up to 35% of the participants found that they were unable to tolerate TMP-SMX (see "Update: Pneumocystis carinii Pneumonia," Vol. 2, No. 5).

As ACTG 268 clearly demonstrated, the gradual initiation of TMP/SMX prophylaxis improves its tolerability. In that trial, patients started on the full dose of TMP-SMX were 2.3 times more likely to discontinue therapy than those who received the ramped dose outlined in Table 2. (For further information on this dosing schedule, see "ACTG 268: Gradual vs. Routine Initiation of PCP Prophylaxis," Vol. 3, No. 2) Clinicians may want to consider this therapeutic option when initiating TMP-SMX therapy, especially in patients with CD4 counts below 50 cells/mm3, since low CD4 counts are predictive of therapeutic failure.

Table 2. Gradual Initiation
of TMP/SMX Therapy
   Dosing Schedule:
   1 cc po for 3 days, then
   2 cc po for 3 days, then
   5 cc po for 3 days, then
   10 cc po for 3 days, then
   20 cc (or 1 double-strength tablet)
   po daily or three times a week

More recently, Community Programs for Clinical Research on AIDS Trial 006 found that initiating PCP prophylaxis with thrice-weekly doses -- and then escalating to daily TMP/SMX after the first month of treatment -- reduces the hypersensitivity reactions that commonly occur during the first month of prophylaxis.

Individuals who are initially intolerant of TMP/SMX should be rechallenged according to the dosing schedule in Table 2. Patients who are at risk for toxoplasmosis as well as PCP and who are persistently intolerant of TMP/SMX should be offered one of two alternative prophylaxis regimens: dapsone (50-100 mg qd) plus pyrimethamine (50 mg weekly) and leucovorin (25 mg weekly); or dapsone (200 mg/day) plus pyrimethamine (75 mg weekly) and leucovorin (25 mg weekly) (Table 3). The relatively small number of patients who are intolerant of both TMP/SMX and dapsone should be prophylaxed with either aerosolized pentamidine or atovaquone suspension.

Table 3. Prophylaxis for PCP
First-Line Therapy:
     (1 double-strength tablet po daily)
Alternative Therapies:
     (1 double-strength tablet po
     three times a week)
     (50 mg po b.i.d. or 100 mg po daily)
     Aerosolized pentamidine
     (300 mg monthly)
     Atovaquone suspension
     (1500 mg po daily)

The patients most likely to experience a breakthrough infection while on PCP prophylaxis are those with CD4 counts below 50 cells/mm3, but nonadherence and previous treatment with regimens other than TMP/SMX are also predictive of prophylaxis failure. Locke and colleagues have reported that patients receiving TMP/SMX or dapsone are more likely to develop dihydropteroate synthase gene mutations than those not receiving prophylaxis. Despite the presence of this mutation, these patients respond to therapy with TMP/SMX, and the role of resistance in breakthrough PCP is unknown.

PCP prophylaxis should be offered to all pregnant HIV-positive women. Once again, TMP/SMX is the agent of choice. Although concerns about the teratogenicity of this agent are purely theoretical, clinicians may wish to offer pregnant patients aerosolized pentamidine, rather than TMP-SMX, during the first trimester.


Although pneumocystis is less of a threat to people with HIV than it once was, PCP still causes significant morbidity and mortality in this population. Clinicians should therefore prescribe prophylaxis for all at-risk patients, and they should encourage consistent compliance with whatever regimen patients are assigned. This advice applies to all patients, even those who have experienced a significant degree of immune reconstitution as a result of highly active antiretroviral therapy.

Judith A. Aberg, M.D., is with the UCSF AIDS Program, San Francisco General Hospital.