Affiliations: Bethesda, MD
Dr. Masur is Chief, Critical Care Medicine, and Dr. Shelhamer is Deputy Chief, Critical Care Medicine, National Institutes of Health, Clinical Center.
Correspondence to: Henry Masur, MD, National Institutes of Health, Clinical Center, Critical Care Medicine Department, 10 Center Dr, Room 7D43, Bethesda, MD 20892-1662; e-mail: firstname.lastname@example.org
When Pneumocystis carinii pneumonia (PCP)
appeared in previously healthy patients in the late 1970s, clinicians
quickly recognized that a new form of immunosuppression had occurred,
and, thus, PCP was the most prominent feature in the initial reports of
the syndrome now designated as AIDS. Clinicians promptly recognized
that the PCP they were seeing in patients with AIDS differed in
important ways from the PCP that was seen in other patient populations.
In fact, most opportunistic pathogens such as Candida, cytomegalovirus,
and herpes simplex caused somewhat different manifestations in AIDS
patients compared to patients who had cancer or had undergone organ
In the early 1980s, Kovacs et al1 compared the clinical
features of 49 cases of AIDS-associated PCP with 39 episodes of PCP
occurring in patients with other immunosuppressive disorders. They
described AIDS-related PCP as a more indolent disease process. The
median time to diagnosis was longer for AIDS-related PCP (28 days) than
for PCP related to other disorders (5 days), yet despite this longer
symptomatic period, AIDS-related PCP patients presented with a higher
median room air Pao2 (69 mm Hg)
than the non-AIDS patients (52 mm Hg). Despite these differences, the
survival rate of patients with PCP was similarly dismal in both groups
In this issue of CHEST (see page 704), Mansharamani and
colleagues assess the management and outcome patterns for adults with
AIDS-related and non-AIDS-related PCP during the period 1985 to 1995,
the decade following the report by Kovacs et al.1 They
reviewed 605 confirmed cases of HIV-associated PCP and 33 cases of
non-HIV-associated PCP. They emphasize two interesting observations
about management. First, over the period of their study, a declining
fraction of patients with HIV-related PCP required hospitalization,
while, in contrast, non-HIV-infected patients with PCP almost always
required hospitalization. Second, the mortality rate for patients with
HIV-related PCP fell during the study period from 11.7 to 6.6%, while
the mortality rate for non-HIV-infected PCP patients was much higher
(39%; there were too few patients in this group to assess a
chronologic trend). Mansharamani et al conclude that, despite major
advances in the tools for managing PCP, not nearly as much progress has
been made in improving the prognosis of PCP in non-HIV-infected
patients compared to those with HIV infection.
Is HIV-related PCP truly a less severe disease than PCP that is
associated with cancer, organ transplantation, or high-dose
corticosteroid use? Data from many sources support the concept that,
while HIV-related PCP can progress rapidly and can be severe and
lethal, it is, in fact, much more often a slowly progressive disease
than the PCP that occurs in other immunosuppressed patients. A recent
review of the experience at a Swiss hospital from 1983 to 1998 supports
this concept.3 Thus, the biology of the host-pathogen
interaction seems to differ between HIV-infected patients and other
immunosuppressed individuals, and there is more opportunity to
intervene before hypoxemia becomes severe.
Do patients with HIV-related PCP truly fare better than patients with
PCP and other immunosuppressive illnesses, as Mansharamani et al
report? The number of non-HIV-infected patients in the study by
Mansharamani et al is small (n = 33 over 10 years), so it is possible
that real improvements were not recognized. The recent study from
Switzerland2 found that the PCP-related mortality
rate in the two groups was actually similar (HIV-infected group, 11%;
non-HIV-infected group, 19%). In the Swiss study, however, the authors
document that, while the number of non-HIV-infected patients was also
small (n = 32), during the period of their study the non-HIV-infected
patients with PCP had, in fact, benefited from advances since their
mean length of hospital stay declined during the study period along
with their mortality rate.
There are several additional reasons, as Mansharamani et al comment,
why HIV-infected patients might have better outcomes when they develop
PCP: (1) their period of susceptibility can be predicted by an
objective laboratory test, the CD4 T-lymphocyte count; (2) PCP is by
far the most likely cause of diffuse pulmonary disease in those with
CD4 T-lymphocyte counts < 200/μL, while in patients with other
immunosuppressive disorders the differential diagnosis is much broader,
making a focused diagnostic approach and an appropriate empiric
therapeutic regimen much more difficult to construct; (3) many
HIV-infected patients are well-educated about the initial
manifestations of PCP and the need for prompt medical interventions;
and (4) many HIV-infected patients are free of other confounding
systemic insults from underlying disease, cytotoxic drug treatment, and
concurrent infections. Thus, recognizing the onset of PCP, establishing
the specific diagnosis, and completing a course of therapy are often
much more straightforward in patients with HIV than in patients who
have cancer or have undergone organ transplantation.
Over the past 20 years, much has been learned about the diagnosis and
treatment of PCP. It is possible to identify patients at high risk for
HIV-related PCP based on peripheral blood CD4 T-lymphocyte counts, and
the period of risk for other patients is recognizable in relation to
the immunosuppressive therapies. Effective regimens for prophylaxis
have been developed. Sensitive and specific methods for diagnosis have
allowed for earlier diagnosis in many cases. Adjuvant corticosteroid
therapy has decreased the number of patients with HIV-related PCP who
develop respiratory failure. Finally, the armamentarium of therapeutic
agents has increased to include atovaquone, clindamycin-primaquine, and
There is, however, much work yet to be done, especially in
patients who have cancer and have undergone organ transplantation.
First, cancer and organ transplant patients would benefit from the
development of an objective test of their susceptibility to PCP;
unfortunately, in these populations the circulating CD4 T-lymphocyte
count is not nearly as sensitive a marker as it is for patients with
HIV infection. Second, less invasive diagnostic tests would facilitate
early diagnosis: the use of oral washes tested with sensitive and
specific nucleic acid amplification techniques hold considerable
promise.3Third, as sulfonamide-resistant PCP appears to
be a clinically important occurrence,4tests for clinical
laboratories to identify resistant organisms and new drugs with novel
mechanisms of action need to be developed. There is woefully little
activity in the area of drug development, although
echinocandins5and primaquine analogs hold some
The dual message of Mansharamani et al is both that progress has been
made and that there is much yet to be done. Clearly, with this problem,
as with so many others, there are “miles to go” before the
morbidity and mortality produced by PCP are effectively eliminated.
Become a CHEST member and receive a FREE subscription as a benefit of membership.
Individuals can purchase this article on ScienceDirect.
Individuals can purchase a subscription to the journal.
Individuals can purchase a subscription to the journal or buy individual articles.
Learn more about membership or Purchase a Full Subscription.
Institutional access is now available through ScienceDirect and can be purchased at myelsevier.com.
Some tools below are only available to our subscribers or users with an online account.
Download citation file:
Web of Science® Times Cited:
Customize your page view by dragging & repositioning the boxes below.
Enter your username and email address. We'll send you a reminder to the email address on record.
Athens and Shibboleth are access management services that provide single sign-on to protected resources. They replace the multiple user names and passwords necessary to access subscription-based content with a single user name and password that can be entered once per session. It operates independently of a user's location or IP address. If your institution uses Athens or Shibboleth authentication, please contact your site administrator to receive your user name and password.