| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Division of Pulmonary and Critical Care Medicine (Drs. Mansharamani, Balachandran, Koziel, and Mr. Garland), Department of Medicine, Beth Israel Deaconess Medical Center, and Harvard Medical School, Boston; and Lahey Clinic Medical Center (Dr. Vernovsky), Burlington, MA.
Correspondence to: Henry Koziel, MD, Division of Pulmonary and Critical Care, Palmer Building, Room 108, Beth Israel Deaconess Medical Center, One Deaconess Rd, Boston, MA 02215; e-mail: hkoziel{at}caregroup.harvard.edu
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Design: Prospective observational study.
Setting: Three hundred seventy-five-bed tertiary-care urban referral teaching hospital, and 250-bed community-based referral hospital.
Patients: One hundred seventy-one consecutive confirmed HIV-seronegative hospitalized and ambulatory adults, including 22 patients with active PCP, 8 patients with bacterial pneumonia, 24 persons in two groups considered at high clinical risk, 38 persons in two groups considered at low or undefined risk, and 79 persons in four groups considered not at risk for PCP (including healthy individuals).
Measurements and results: Compared to counts in healthy individuals, median CD4 + counts were significantly decreased in patients with active PCP (61 cells/µL vs 832 cells/µL; p = 0.001) where 91% of patients had a CD4 + count < 300 cells/µL at the time of PCP diagnosis. Median CD4 + counts were also reduced in the high clinical risk groups of recent organ transplant recipients (117 cells/µL; p = 0.007), 64% with < 300 cells/µL, and patients receiving chemotherapy (221 cells/µL; p < 0.01), 80% with < 300 cells/µL. For the low or undefined clinical risk groups, the median CD4 + counts were not significantly reduced, although 39 to 46% of individuals receiving long-term corticosteroid therapy (alone or in combination with other agents) had CD4 + counts < 300 cells/µL. Median CD4 + counts in individuals considered not at risk for PCP were similar to those in healthy subjects. Compared to counts in patients with active PCP, median CD4 + counts were significantly higher in bacterial pneumonia patients (486 cells/µL; p < 0.05), but similar to those in healthy subjects.
Conclusions: These data suggest that for immunosuppressed persons without HIV infection (especially in low or undefined PCP risk groups), CD4 + counts may be a useful clinical marker to identify specific individuals at particularly high clinical risk for PCP and may help to guide chemoprophylaxis.
Key Words: corticosteroids • CD4 + T lymphocytes • immunosuppression • Pneumocystis carinii pneumonia
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Other groups of immunocompromised persons are traditionally considered at risk for PCP, although with rare exception7 the incidence of PCP is lower or undefined, compared to the HIV-positive group. In the absence of chemoprophylaxis, 3.4 to 43% of organ transplant recipients8 and 2.6 to 43% of patients receiving chemotherapy for malignancies9 10 11 12 may develop PCP. Persons receiving long-term treatment with corticosteroids represent another important risk group.1 2 3 13 14 However, the incidence of PCP in persons receiving long-term treatment with corticosteroids is undefined, although likely lower than the estimated incidence of 3.8 to 12% of persons who are prescribed corticosteroids in combination with other immunosuppressive agents.13 15 16 In contrast to HIV-positive individuals, clinically useful biological markers that guide the initiation of chemoprophylaxis are not available.
The importance of identifying specific individuals at high clinical risk for non-HIV PCP relates to the severity of this complication, with reported mortalities of 34 to 58%.2 3 14 17 18 Furthermore, the number of recognized cases of non-HIV PCP may be increasing,19 20 21 22 18% of PCP diagnoses may be postmortem,22 and patients who develop non-HIV PCP are often not receiving chemoprophylaxis.2 17 22 The availability of a biological marker would allow appropriate targeting of chemoprophylaxis, avoid unnecessary chemoprophylaxis and associated toxicity23 in persons who may not be at risk, and limit the potential development of P carinii drug resistance.24 These issues are of particular importance for persons receiving long-term treatment with corticosteroids alone, in which the incidence of PCP is likely relatively low.
Although clinically established for HIV-infected individuals,5 25 the role of the CD4 + count in other immunosuppressed patients has not been fully assessed. Several observations in animal models suggest that CD4 + T lymphocytes mediate an effective host response to PCP in the absence of HIV infection.26 27 For persons considered at risk for PCP, most receive immunosuppressive agents (including corticosteroids),1 2 3 13 which have known adverse effects on T-lymphocyte number or function,28 29 30 31 and limited reports suggest that CD4 + counts may be low in cases of non-HIV-related PCP.2 31 32 The purpose of this study was to assess the potential use of CD4 + counts as a biological marker to identify immunocompromised patients (without HIV infection) at risk for PCP, by comparing CD4 + counts in patients with active PCP to persons in various clinical risk groups for this opportunistic infection.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Study Design
A prospective observational study was performed on consenting
individuals following protocols approved by the hospital institutional
review boards. Consecutive adult patients (
18 years of age) were
identified by members of the Division of Pulmonary and Critical Care
Medicine and the Respiratory Therapy Department, medical records
were reviewed for clinical and demographic information, and all
information was recorded on a standardized form. Identified patients
provided a heparinized blood sample for WBC differential determination
and T-lymphocyte subset analysis, and a serum sample for HIV-antibody
testing. T-lymphocyte subset analysis was performed in the clinical
hematology laboratory by trained personnel using a FACScan flow
cytometer (Becton Dickenson; Mountain View, CA) and commercially
available fluorescent antibodies CD45 (KC56), CD3/CD4, and CD3/CD8
(Pharmingen; San Diego, CA).
Cases of PCP
Active cases of non-HIV PCP were identified by the daily review
of clinical pathology reports that demonstrated cytologic or direct
fluorescent antibody staining for P carinii in
specimens of induced sputa or BAL. All cases were identified
consecutively. No cases were excluded. No presumptive cases were
included. All cases were confirmed to be HIV seronegative by
enzyme-linked immunosorbent assay. Blood for T-lymphocyte subset
analysis was obtained within 2 weeks of the diagnosis of PCP.
Study Subjects for Control Groups
Preliminary observations suggest that CD4 + counts may be low
in persons with active non-HIV PCP.2
To determine the
specificity of CD4 + counts in persons with active PCP,
CD4 + counts were performed for nine control groups in the following
four categories:
(1) Persons considered at high clinical risk for PCP, represented by two groups, including patients with recent (< 6 months of surgical transplant) solid organ transplantation, and hematology/oncology patients receiving chemotherapy. For both groups, the incidence for PCP may be up to 43%.8
(2) Persons considered at low or undefined risk for PCP, represented by two groups, including patients receiving long-term (> 1-month duration) administration of systemic corticosteroids,3 and patients requiring corticosteroids in combination with at least one other immunosuppressive agent. The incidence of PCP in patients with Wegener’s granulomatosis receiving corticosteroids in combination with a second agent may be 3.8 to 12%,13 15 16 whereas for persons receiving long-term corticosteroid treatment alone the incidence is unknown.
(3) Persons considered not at risk for PCP, represented by four groups, including patients with underlying medical diseases but not receiving corticosteroids, patients with medical diseases receiving short-term (< 2 weeks) treatment with corticosteroids, patients with remote (> 12 months) organ transplantation, and healthy asymptomatic individuals. At our institutions, renal, hepatic, and pancreatic transplant recipients require reduced immunosuppressive agents by 12 months, and thus are not routinely prescribed anti-P carinii prophylaxis beyond 12 months.
(4) Persons with bacterial pneumonia, represented by a group of patients with clinical and radiographic evidence for active pneumonia with identified organisms other than P carinii.
Statistical Analysis
Data for comparison of groups were analyzed employing the
Kruskal-Wallis nonparametric analysis of variance test using an INSTAT2
statistical package (Graphpad Software; San Diego, CA) on an IBM PS/2
120 MB computer (IBM; Armonk, NY). Statistical significance was
accepted for p values < 0.05. Dunn’s Multiple Comparisons posttest
was performed if the calculated p values were < 0.05.
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
For the category of persons considered not at clinical risk for PCP, for the group of 24 patients with medical diseases but not requiring corticosteroids, underlying conditions included COPD (n = 8), asthma (n = 6), sarcoidosis (n = 2), non-small cell lung cancer (n = 2), chronic renal failure (n = 2), and one each of IPF, lymphoma, pneumonitis, and congestive heart failure. For the group of 16 patients with medical disease and receiving short-term treatment (< 2 weeks) with corticosteroids, underlying conditions included COPD (n = 7), asthma (n = 3), and one each of scleroderma, laryngeal edema, pneumonitis, CLL, diffuse alveolar hemorrhage, and small cell lung cancer. For the group of 11 patients with remote (> 12 months) organ transplantation, transplanted organs included hepatic (n = 4), renal (n = 4), and one each of cardiac, bone marrow, and combination renal and pancreatic. None of these individuals developed PCP at the 12-month follow-up date. For the group of 18 healthy asymptomatic individuals, none had active pulmonary disease.
For the eight individuals with active bacterial pneumonia (not P carinii), underlying conditions included COPD (n = 4), bronchiectasis (n = 2), diabetes mellitus (n = 1), and remote (> 5 years) renal transplant (n = 1). All had radiographic evidence for lobar pneumonia in the setting of purulent sputum, fever, and leukocytosis.
CD4 + T-Lymphocyte Measurements in Patients With Active PCP
CD4 + counts were performed for all 22 patients with PCP (Table 2) . The mean peripheral total WBC count was 22.8 ± 37 cells/µL
(mean x 103 ± SD). The mean value for
fraction of circulating lymphocytes was 31.0 ± 33% of the total WBC
count, and the lymphocyte fraction was < 20% of total WBC count for
12 of these patients. At the time of PCP diagnosis, eight patients had
CD4 + counts < 50 cells/µL, with undetectable CD4 + counts in
four of these patients. For the 22 patients, the mean CD4/CD8 ratio was
2.1 ± 2.7, and the ratio was < 1.0 in 9 of these patients.
CD4 + T-Lymphocyte Measurements in Control Subjects Without
Active PCP
CD4 + counts were available for 149 control subjects (Table 2) .
For the patients with active PCP, the median CD4 + count of 61
cells/µL was significantly lower (p < 0.001) compared to healthy
subjects and compared to other persons not at risk for PCP. There were
no significant differences comparing CD4 + counts in healthy
individuals to counts in persons considered at low or undefined risk
for PCP.
Compared to counts in healthy individuals, the median CD4 + counts were significantly lower for subjects in the high-risk groups, but not different compared to counts in the group with active PCP (p > 0.05). For the transplant recipients, the median CD4 + count was significantly lower for individuals with recent transplants compared to individuals > 12 months following surgical transplantation (117 cells/µL vs 453 cells/µL, respectively; p < 0.01).
To estimate the specificity of the CD4 + count in pneumonia, peripheral blood CD4 + counts for eight persons with active pneumonia other than P carinii were significantly higher, compared to counts in persons with active PCP (p < 0.05), but not significantly different from counts in healthy individuals or other medical patients without pneumonia.
Longitudinal CD4 + T-Lymphocyte Measurements in Subjects Without
Active PCP
To examine for variability in CD4 + counts over time, serial
CD4 + measurements were available for four healthy individuals and
four organ transplant recipients (Fig 1
). For the healthy individuals, baseline biological variability was
observed, although CD4 + counts remained relatively stable for each
subject over a period of 6 to 8 months. For solid organ transplant
recipients, CD4 + counts were low immediately after organ
transplantation, and increased over the course of 12 months as the
level of immunosuppression was gradually reduced.
|
|
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
For persons in the low or undefined clinical risk group for PCP, the median CD4 + counts were not significantly reduced, compared to healthy persons or persons considered not at risk for PCP, as these individuals demonstrated significant biological variability. However, a subset of this risk group had very low CD4 + counts, where 39 to 47% of persons had CD4 + counts < 300 cells/µL (representing the level encompassing 91% of active PCP cases in this study). This subset of persons receiving long-term treatment with corticosteroids (alone or in combination) with CD4 + counts < 300 cells/µL may thus represent specific individuals at particularly high clinical risk for PCP and who may most benefit from chemoprophylaxis, although additional studies to establish this relationship would be required.
The appropriate prescription of PCP chemoprophylaxis for immunocompromised patients without HIV infection has not been established, and whether all patients in a recognized risk category for PCP should receive chemoprophylaxis remains controversial.3 Recognizing the overall lower PCP incidence in persons without HIV infection, the identification of an individual in a recognized PCP risk category may not be sufficient basis to initiate PCP chemoprophylaxis. By analogy, although HIV-infected persons are considered a very high-risk group, PCP chemoprophylaxis is generally not initiated at the time of HIV-seropositive determination, but rather only when CD4 + counts measure < 200 cells/µL. The current study suggests that CD4 + counts < 300 cells/µL may potentially serve to guide initiation of chemoprophylaxis in immunocompromised persons without HIV infection, although the specific level of CD4 T-lymphocytes that confers protection would require further study.
Establishing a threshold level of CD4 + T lymphocytes may be of particular importance for persons receiving long-term treatment with corticosteroids. Antecedent corticosteroid use is cited in 80 to 94% of cases of non-HIV PCP,1 2 3 13 and corticosteroids for chronic inflammatory medical diseases account for 30 to 31.8% of reported cases of non-HIV-related PCP.2 3 However, as the incidence of PCP for these individuals is unknown, prescription of chemoprophylaxis for all persons receiving long-term treatment with corticosteroids3 18 20 may unnecessarily expose the majority of these patients to the side effects of these agents without clear benefit and potentially engender P carinii drug resistance.33 For these individuals, the potential use of CD4 + counts can be illustrated by the following: for persons with underlying immunologic disorders receiving > 16 to 20 mg/d of prednisone (alone or in combination) for > 1 month,3 8 determination of a CD4 + count < 300 cells/µL may identify persons who may benefit most from chemoprophylaxis, whereas persons with sustained CD4 + counts > 300 cells/µL may not require chemoprophylaxis. This hypothesis would require prospective confirmation.
These data support an important role for CD4 + T lymphocytes in host susceptibility to this opportunistic pathogen in the absence of HIV infection. This finding is consistent with data from animal models of PCP26 30 34 and recent preliminary reports of subjects with non-HIV PCP.2 31 32 Interestingly, the observed threshold level in the current study is similar to the 200 to 300 cells/µL threshold level observed for HIV-positive patients,5 suggesting that this threshold level may not be specific to HIV infection and that a critical level of CD4 + T-lymphocytes is required to mediate a protective host response to P carinii. A recently reported association of P carinii colonization in the airways of non-HIV immunocompromised patients with CD4 + counts < 400 cells/µL35 supports the concept of a critical threshold level, although defining the levels which confer protection or are permissive for disease require further prospective validation.
The finding that CD4 + counts were not universally below a threshold
level in cases of active PCP is similar to that observed for
HIV-positive patients, in which 6.1% had CD4 + counts above the
threshold of 200 cell/µL.5
These observations suggest
that other factors are important in host susceptibility, which may
include adverse effects of immunosuppressive agents on CD4 +
T-lymphocyte function, or adverse effects on other important immune
mediators such as interferon-
, CD8 + T lymphocytes,36
or alveolar macrophages.37
Host susceptibility may also
reflect contributions of P carinii environmental exposures
and virulence factors, although these are poorly understood.
These data also suggest that CD4 + counts may guide the duration of PCP chemoprophylaxis in organ transplant recipients, as the appropriate duration has not yet been established.38 The finding that median CD4 + counts were significantly low in recent organ transplant recipients (considered at high clinical risk), but not in remote recipients (considered not at clinical risk), together with the observed gradual increase in CD4 + counts as the level of immunosuppression is reduced may provide a rational basis for the discontinuation of chemoprophylaxis as the levels rise above a threshold value, analogous to HIV-infected patients receiving highly active antiretroviral therapy.39 Studies would be required to determine the necessary sustained level of CD4 + counts prior to discontinuation of PCP chemoprophylaxis.
Additional limitations of the current study include the observational nature and the relative small size of each study group. The level of CD4 + counts, which may confer protection, although suggested, was not definitively established, and the duration of low CD4 + count prior to the development of PCP was not determined. Although longitudinal data were available for some individuals, data were not available for other groups, including those in the low- or undefined-risk group. For persons with active PCP, CD4 + counts prior to the episode of PCP were not available, as CD4 + counts are not routinely measured in these individuals. Although a comparison to persons with bacterial pneumonia was presented, CD4 + counts for immunocompromised non-PCP patients were not available. Finally, the experience in our two centers may not reflect the patients at centers in other regions or countries, although a broad spectrum of medical diseases was represented.
In conclusion, this study demonstrates that active PCP (in the absence of HIV infection) is associated with markedly reduced CD4 + counts. Importantly, the observation that CD4 + counts are also significantly reduced in persons in high clinical risk categories suggests that low CD4 + counts may predispose to PCP in patients without HIV infection. Furthermore, the finding of very low CD4 + counts in a subset of individuals in the low or undefined risk category suggests CD4 + counts may potentially serve to identify specific patients who may benefit from PCP chemoprophylaxis. These preliminary findings provide the rationale for larger multicenter prospective cohort or case-control studies to examine the role of CD4 + counts as a predictor of risk for PCP in immunocompromised patients without HIV infection. If verified, low CD4 + counts may identify specific individuals who are at particularly high clinical risk (whether in the high-, low-, or undefined-risk categories) and may allow targeting of anti-PCP chemoprophylaxis. Establishing this biological marker may be especially useful for patients requiring long-term corticosteroids.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Presented in part at the 1999 American Thoracic Society International Meeting, San Diego, CA.
Received for publication January 5, 2000. Accepted for publication April 20, 2000.
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and CD8+ T-cells restore host defense against Pneumocystis carinii in mice depleted of CD4+ T-cells. J Immunol 162,2890-2894This article has been cited by other articles:
|
|
 |
|
  S. M. Tolaney, A. H. Partridge, R. G. Sheib, H. J. Burstein, and E. P. Winer Pneumocystis Carinii Pneumonia During Dose-Dense Chemotherapy for Breast Cancer J. Clin. Oncol., November 20, 2006; 24(33): 5330 - 5331. [Full Text] [PDF]
|
|
|
|
 |
|
 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]
|
|
|
|
 |
|
 M. Rodriguez and J. A. Fishman Prevention of Infection Due to Pneumocystis spp. in Human Immunodeficiency Virus-Negative Immunocompromised Patients Clin. Microbiol. Rev., October 1, 2004; 17(4): 770 - 782. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. A. Norris, H. Wildschutte, J. Franko, and K. F. Board Genetic Variation at the Mitochondrial Large-Subunit rRNA Locus of Pneumocystis Isolates from Simian Immunodeficiency Virus-Infected Rhesus Macaques Clin. Vaccine Immunol., November 1, 2003; 10(6): 1037 - 1042. [Abstract] [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]
|
|
|
|
 |
|
 M. R. Nicolls, M. Coulombe, J. Beilke, H. C. Gelhaus, and R. G. Gill CD4-Dependent Generation of Dominant Transplantation Tolerance Induced by Simultaneous Perturbation of CD154 and LFA-1 Pathways J. Immunol., November 1, 2002; 169(9): 4831 - 4839. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. R. Heaton, D. G. Blount, S. J. Mann, P. Devlin, S. Koelsch, B. H. E. Smith, J. Stevenson, E. J. Harper, and J. M. Rawlings Assessing Age-Related Changes in Peripheral Blood Leukocyte Phenotypes in Domestic Shorthaired Cats Using Flow Cytometry J. Nutr., June 1, 2002; 132(6): 1607S - 1609. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. R. Heaton, D. G. Blount, P. Devlin, S. Koelsch, S. J. Mann, B. H. E. Smith, J. Stevenson, and E. J. Harper Assessing Age-Related Changes in Peripheral Blood Leukocyte Phenotypes in Labrador Retriever Dogs Using Flow Cytometry J. Nutr., June 1, 2002; 132(6): 1655S - 1657. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Kovacs, V. J. Gill, S. Meshnick, and H. Masur New Insights Into Transmission, Diagnosis, and Drug Treatment of Pneumocystis carinii Pneumonia JAMA, November 21, 2001; 286(19): 2450 - 2460. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Cardenal, F. J. Medrano, J. M. Varela, A. Ordonez, C. Regordan, M. Rincon, A. Martinez, and E. J. Calderon Pneumocystis carinii pneumonia in heart transplant recipients Eur. J. Cardiothorac. Surg., October 1, 2001; 20(4): 799 - 802. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 CD4 Counts: A Useful Marker for PCP in HIV-Negative Patients Journal Watch Infectious Diseases, November 16, 2000; 2000(1116): 7 - 7. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
300 cells/µL may
represent a level that identifies specific individuals at particularly
high clinical risk for PCP (*p < 0.05 compared to healthy
group).