|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Division of Pulmonary and Critical Care Medicine, Department of Medicine, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, MA.
Correspondence to: Henry Koziel, MD, Division of Pulmonary and Critical Care Medicine, Palmer Building, Room 108, Beth Israel Deaconess Medical Center, One Deaconess Rd, Boston, MA 02215; e-mail: hkoziel{at}caregroup.harvard.edu
| Abstract |
|---|
|
|
|---|
Design: Retrospective review of medical records.
Setting: A 375-bed tertiary-care urban teaching hospital and referral center.
Patients: All adult patients with morphologically confirmed PCP from 1985 to 1995.
Measurements and results: From 1985 to 1995, 638 confirmed cases of PCP were identified, including 605 cases in 442 HIV-positive persons (HIV + PCP), and 33 cases in 33 non-HIV patients (non-HIV PCP). For HIV + PCP cases, a peak of 104 cases occurred in 1987, with a gradual decline to 23 in 1995. The proportion of cases requiring hospitalization declined from a peak of 91.6% in 1987 to a low of 51.6% in 1992. ICU admission was required for 6.3 to 8.2%, and mechanical ventilation for 4.7 to 5.7%. Overall mortality improved from 11.7 to 6.6%, although mortality for intubated patients remained at 50 to 60%. For the non-HIV PCP cases, 97% occurred from 1989 to 1995 with similar annual frequency, 97% required hospitalization, 69% required ICU admission, and 66% required intubation. Overall mortality was 39%, and mortality for intubated patients was 59%.
Conclusions: Despite major advances in diagnosis and management, PCP remains a significant problem in non-HIV-infected patients, and respiratory failure remains associated with a high mortality rate for patients with both HIV + PCP and non-HIV PCP.
Key Words: AIDS HIV immunosuppression mechanical ventilation mortality Pneumocystis carinii pneumonia
| Introduction |
|---|
|
|
|---|
In contrast to AIDS-related cases, cases of PCP in patients with other predisposing immunodeficiency states17 18 19 20 21 22 23 (uncommon prior to the AIDS epidemic) may be increasing,24 25 and the associated mortality may be > 50%.19 20 21 25 Furthermore, management of PCP associated with other immunodeficiency states (non-HIV PCP) is less standardized. For example, PCP chemoprophylaxis is routinely prescribed for recent organ transplant recipients (traditionally recognized at high clinical risk for PCP), while the role for chemoprophylaxis for individuals receiving long-term corticosteroid therapy (traditionally considered at some, although undefined, risk) has not been established.23 Furthermore, the duration of prophylaxis required for solid organ transplant recipients26 and the role for acute heightened-dose adjunctive corticosteroids for the treatment of non-HIV PCP remain undefined.22
Recognizing differences in clinical and laboratory characteristics comparing HIV + PCP to non-HIV PCP,19 27 a prior report suggested similar high mortalities of 57% and 50%, respectively.19 However, as with most recent large series,19 21 24 28 these reports included data prior to the major advances in the management of AIDS-related PCP. Of the reports that include cases since 1990,23 25 no comparison to HIV + PCP is included. Although several investigators have suggested that advances in the management of HIV + PCP may be applied to non-HIV PCP,20 29 30 specific studies related to non-HIV PCP are limited; whether the significant changes in management and outcomes of non-HIV PCP paralleled the changes associated with advances in HIV + PCP management remains to be determined. Reviewing an 11-year period during the AIDS epidemic, and including periods before and after the introduction of major advances in the diagnosis and treatment of AIDS-related PCP, the purpose of this study was to examine the management patterns and clinical outcomes of PCP at a tertiary-care urban hospital comparing cases in HIV-infected persons to cases in other immunocompromised persons without HIV infection.
| Materials and Methods |
|---|
|
|
|---|
Identification of PCP Cases
Confirmed cases of PCP were identified by a retrospective review
of detailed microbiological and pathologic records that identified PCP
in biological specimens for the period from 1985 to 1995. These years
represented a period for which complete medical information was
available, and also included the period preceding and following major
advances in the care of AIDS-related PCP. Biological specimens from
inpatient and outpatient services included respiratory tract specimens
(induced sputa, BAL, transbronchial biopsies), pleural fluid analysis,
and surgical tissue biopsy. All requests for induced sputum and BAL
specimens were administered through the Division of Pulmonary Medicine
and were detailed in comprehensive log books.
Only cases demonstrating P carinii organisms by immunofluorescence, modified Giemsa, or methenamine silver staining in specimens of induced sputum, BAL, transbronchial or open lung biopsy, or other biological specimens were included in the study. Presumptive cases of PCP were excluded. The demonstration of typical P carinii organisms in respiratory tract specimens within 30 days of a documented episode of PCP was considered as a single event.
Patient Demographics and Clinical Outcomes
Outpatient and inpatient medical records of all identified
confirmed cases in patients > 18 years of age were reviewed, and
information was recorded on standardized forms, including age, gender,
underlying medical conditions, HIV status and risk factors, peripheral
CD4 T-lymphocyte count, use of immunosuppressive agents prior to PCP
diagnosis, and use of chemoprophylaxis against P carinii.
Ethnicity data were not available.
The measured outcomes included requirement for hospitalization, ICU admission, intubation, and associated mortalities. To accurately identify all patients requiring intubation and mechanical ventilation, medical records data were cross-referenced with the log books of the ICU and respiratory therapists. Respiratory failure was defined as requiring endotracheal intubation and mechanical ventilation support. All persons with HIV + PCP were documented HIV seropositive. All non-HIV cases were documented HIV seronegative.
Statistical Analysis
For the statistical analysis, continuous data were compared with
use of the Students t test31
and confirmed by
the Mann-Whitney test. Statistical analysis of noncontinuous
dichotomous data were compared by the
2 test
(with Yates correction) or the Fishers Exact Test as
appropriate.32
All statistical analysis were performed on
an IBM PS/2 computer (IBM; Armonk, NY) utilizing INSTAT2 statistical
program (Graphpad Software; San Diego, CA). All p values were two
sided, and statistical significance was accepted for
p < 0.05.
| Results |
|---|
|
|
|---|
|
|
|
Mortality for PCP
The overall mortality for the HIV + PCP group was 9.6%, and
hospital mortality was 13.2%. For patients with acute respiratory
failure requiring intubation, the mortality was 56.2%. In comparison,
the overall mortality for the non-HIV PCP patients was 39.4%
(p < 0.0001, compared to HIV + PCP), in-hospital mortality was
40.6% (p < 0.0001), and the mortality for non-HIV PCP patients who
required intubation and mechanical ventilation was 59.1% (p = 1.0).
Trends in Outcomes in PCP From 1985 to 1995
The trends for ICU admission, intubation, and mortality rates are
presented in Figure 2
. For HIV + PCP cases, the proportion of patients requiring
hospitalization declined from a peak of 91.6% in 1987 to a low of
51.6% in 1992. For the hospitalized patients, the need for ICU
admission and mechanical ventilation peaked in 1991, corresponding to
the introduction of adjunctive corticosteroids (Fig 2
, top,
A). Overall mortality ranged from 3.4 to 14.8%, and
mortality for intubated patients was 0 to 80%. In comparison, for
non-HIV PCP, all patients since 1989 required hospitalization, and the
rates of ICU admission, intubation, and mortality remained consistently
high (Fig 2
, bottom, B).
|
|
|
| Discussion |
|---|
|
|
|---|
In contrast, the number of confirmed cases of PCP identified in patients with other immunocompromised states increased over the period from 1985 to 1995, particularly during the latter half of this period. Only one case was identified prior to 1989, whereas 97% of confirmed cases were identified since 1989 at this institution. Importantly, the rates of hospitalization, ICU admission, intubation with mechanical ventilation, and mortality for the non-HIV PCP group remained high, where 97% of the confirmed cases required hospitalization, 68.8% required ICU admission, and 65.6% required intubation and mechanical ventilation. The overall mortality for these patients was 39.4%, and the mortality for those experiencing acute respiratory failure and requiring intubation was 59.1%. For the period from 1990 to 1995, these outcomes were worse than for HIV + PCP in this study, and these outcomes were similar to published reports from other institutions made prior to the advances for HIV + PCP.21 25 28
The explanation for the progressive decline in the number of cases of HIV + PCP was not specifically addressed in this retrospective review. The decline was not associated with a decline in the population of HIV-positive patients followed at our institution. Possible explanations may include the appropriate identification of persons at high clinical risk, the targeted prescription of effective chemoprophylaxis, and more frequent use of empirical treatment for suspected PCP. At this institution, for example, over a period from 1989 to 1993, empirical treatment for PCP increased from 10 to 20% (H. Koziel, MD; unpublished data; October 1996). Although the decline in the incidence of HIV + PCP may reflect the efficacy of HAART controlling viral replication, the observed decline in PCP cases occurred prior to the introduction of potent protease inhibitors,15 suggesting that other factors contributed to the overall decline in HIV + PCP cases.
The explanation for the increase in non-HIV PCP cases was not specifically addressed in this study. Similar increased incidences have been reported by others.24 25 There were no changes in the transplantation programs, changes in the composition or number of non-HIV-infected patients, or changes in the diagnostic approach or management style toward immunosuppressed patients at our institution to account for the observed increased incidence of non-HIV PCP. Possible explanations may include heightened clinical suspicion in the advent of the AIDS epidemic; improved diagnostic techniques, together with increased use of immunosuppressive agents; and expansion of an aging population.34 Although our understanding of P carinii epidemiology is limited, increased cases may reflect in part environmental exposures to P carinii and P carinii virulence factors (although these are poorly understood).
The decline in hospitalization rates for HIV + PCP may reflect more widespread use of chemoprophylaxis, and improved practice standards as physicians developed experience managing PCP over time, coupled with earlier identification of cases and earlier institution of specific therapy based on a higher clinical index of suspicion. A similar decline in the hospital management of HIV + PCP has been reported by other investigators.14 35 The high hospitalization rate for non-HIV PCP may reflect the severity of PCP in this patient population, host response to the organism,27 underlying immunodeficiency state, advanced age,34 underlying medical comorbidities, and the more acute presentation for non-HIV PCP.19
Respiratory failure was the most frequent indication for ICU admission for both groups, representing 74.4% of HIV + PCP and 95.5% of non-HIV PCP ICU admissions. The overall incidence of acute respiratory failure requiring intubation was much higher in non-HIV PCP cases (63.6%), compared to HIV + PCP cases (5.2%). The higher respiratory failure rate is consistent with published reports,21 25 and may represent differences in the host-parasite interaction comparing HIV-positive to non-HIV-infected patients,19 27 may be related to an influx of neutrophils in the alveolar airspace,27 or may reflect the underlying medical conditions associated with non-HIV PCP.21 The high mortality associated with respiratory failure complicating HIV + PCP and non-HIV PCP in the current and other studies21 33 35 36 clearly highlights the severity of this complication and underscores the need to better understand the pathophysiology to develop new or improved therapeutic interventions.
In the current study, the observed overall mortality for HIV + PCP was 9.6%, similar to that reported by other investigators.8 37 In contrast to HIV-infected persons, the overall mortality rate for non-HIV PCP in our study was 39.4%. The poorer outcomes for non-HIV PCP in this study are unlikely due to inexperience in the diagnosis and management of PCP at our institution. Similar high mortality rates reported in other institutions19 20 21 38 39 likely in part represent the lack of a standardized approach to the management of non-HIV PCP over the period of this study. The persistently high morbidity and mortality associated with non-HIV PCP observed in the current study suggest that patients with non-HIV PCP have not benefited from the advances in the management of HIV + PCP, although other factors may be important. A recent report comparing HIV + PCP to non-HIV PCP suggests that non-HIV PCP mortality may have decreased since 1995.39
For recurrent episodes of HIV + PCP, the mortality rates remain low and relatively stable for up to four episodes, with 79 to 88% survival, similar to report from other centers.33 40 Interestingly, no recurrent episodes of non-HIV PCP were identified, which may in part be attributed to the routine administration of specific prophylaxis on recovery from PCP (none of the non-HIV-infected patients were taking prophylaxis at the time of PCP diagnosis), or perhaps to the transient and reversible nature of the underlying immunodeficiency. Recognizing the potential importance of organism burden in AIDS-related PCP,41 other factors that may influence recurrence rates in the non-HIV-infected population may relate to the relatively low "load" of P carinii organisms (compared to cases of HIV + PCP).27
Other limitations of the current study include the retrospective nature of the investigation, the relative small number of non-HIV PCP cases compared to HIV + PCP cases, and the exclusion of presumptive PCP cases. Although a comparison of experiences of HIV + PCP to non-HIV PCP at a single institution is valuable, the findings at a single institution may not reflect the experience at other institutions with more diverse patient populations. The findings for HIV + PCP may in part reflect the predominantly homosexual risk group and may not be representative of other HIV risk groups. For the non-HIV PCP cases, the advanced age and the greater proportion of women likely reflect the demographics of persons at risk for acquired immunodeficiency in the absence of HIV infection and are consistent with other reports.24 25 38 42 43 Although the study included the period encompassing effective combination antiretroviral therapy,13 this study did not estimate the impact of newer antiretroviral agents, as the routine prescription of protease inhibitors occurred after 1995. Although the improved HIV + PCP outcomes in the current study were observed prior to the routine use of protease inhibitors, the introduction of HAART may have additional favorable impact on the incidence and outcomes of HIV + PCP.15 16
In summary, this study demonstrates that PCP remains a significant problem in the immunocompromised host and that significant differences remain in PCP outcomes comparing HIV + PCP to non-HIV PCP at our institution. Whereas for HIV + PCP, the annual number of cases, hospitalization, and overall mortality have declined over the period from 1985 to 1995, the number of non-HIV PCP cases has increased, and the need for hospitalization, ICU admission, intubation, and the overall mortality rates remain high. Furthermore, respiratory failure remains a formidable complication associated with high mortality rates for both HIV-infected patients and persons with other immunodeficiency states. This study suggests that whereas improved management and clinical outcomes for HIV + PCP may in part reflect advances in the diagnosis and treatment of this important opportunistic infection, similar benefits have not been apparent for non-HIV PCP. As extrapolation of HIV-related data to cases of non-HIV PCP may not be appropriate, further studies are needed to better identify non-HIV-infected persons at risk, and further define the use of adjunctive agents such as corticosteroids,22 and to identify and treat patients at risk for respiratory failure complicating PCP.
| Footnotes |
|---|
These data were presented in part at the 1997 ATS/ALA International Conference in San Francisco, CA.
Received for publication October 11, 1999. Accepted for publication March 20, 2000.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. Zhang, S. D. Tachado, N. Patel, J. Zhu, A. Imrich, P. Manfruelli, M. Cushion, T. B. Kinane, and H. Koziel Negative regulatory role of mannose receptors on human alveolar macrophage proinflammatory cytokine release in vitro J. Leukoc. Biol., September 1, 2005; 78(3): 665 - 674. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Festic, O. Gajic, A. H. Limper, and T. R. Aksamit Acute Respiratory Failure Due to Pneumocystis Pneumonia in Patients Without Human Immunodeficiency Virus Infection: Outcome and Associated Features Chest, August 1, 2005; 128(2): 573 - 579. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. E. Evans, P. Y. Hahn, F. McCann, T. J. Kottom, Z. V. Pavlovic', and A. H. Limper Pneumocystis Cell Wall {beta}-Glucans Stimulate Alveolar Epithelial Cell Chemokine Generation through Nuclear Factor-{kappa}B-Dependent Mechanisms Am. J. Respir. Cell Mol. Biol., June 1, 2005; 32(6): 490 - 497. [Abstract] [Full Text] [PDF] |
||||
![]() |
Z. Wang, A. Foye, Y. Chang, P. R. Chess, T. W. Wright, S. Bhagwat, F. Gigliotti, and R. H. Notter Inhibition of surfactant activity by Pneumocystis carinii organisms and components in vitro Am J Physiol Lung Cell Mol Physiol, June 1, 2005; 288(6): L1124 - L1131. [Abstract] [Full Text] [PDF] |
||||
![]() |
R J Boyton, D M Mitchell, and O M Kon The pulmonary physician in critical care * Illustrative case 5: HIV associated pneumonia Thorax, August 1, 2003; 58(8): 721 - 725. [Full Text] [PDF] |
||||
![]() |
G. S. Gilmartin and H. Koziel Pneumocystis Carinii Pneumonia in Adult Non-HIV Disorders J Intensive Care Med, November 1, 2002; 17(6): 283 - 301. [Abstract] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |