|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Division of Pulmonary and Critical Care Medicine, Georgetown University Medical Center, Washington, DC.
Correspondence to: Anne E. ODonnell, MD, FCCP, Division of Pulmonary and Critical Care Medicine, Georgetown University Medical Center, 3800 Reservoir Rd, NW, Washington, DC 20007-2197; e-mail: odonnela{at}georgetown.edu
| Abstract |
|---|
|
|
|---|
Design: Retrospective chart review.
Setting: A tertiary care university hospital.
Patients: All HIV-infected patients referred to the Pulmonary and Critical Care Medicine Service from January 1, 1993, through December 31, 1995 (era 1) and from July 1, 1997, through June 30, 2000 (era 2).
Interventions: Inpatient and outpatient charts were reviewed for data regarding patient demographics, CD4 cell counts, viral load levels, duration of HIV seropositivity, history of opportunistic infections, and final diagnosis.
Results: Pneumocystis carinii pneumonia (PCP) was less common in the HAART era than in the pre-HAART era, whereas bacterial pneumonia and non-Hodgkins lymphoma (NHL) were more common in the HAART era than in the pre-HAART era. HAART was protective against PCP (odds ratio [OR], 0.37; confidence interval [CI], 0.16 to 0.89) in a manner dependent on the CD4 cell count. Patients receiving HAART were at increased risk for the development of bacterial pneumonia (OR, 2.41; CI, 1.12 to 5.17) and NHL (OR, 15.11; CI, 3.14 to 28.32). A history of PCP indicated a risk factor for bacterial pneumonia (OR, 2.14; CI, 1.13 to 4.04). A history of cytomegalovirus infection indicated a risk factor for NHL (OR, 6.0; CI, 1.27 to 28.32).
Conclusions: There have been significant changes in the spectrum of HIV-related pulmonary complications seen by our Pulmonary and Critical Care Medicine Service in the HAART era.
Key Words: bacterial pneumonia highly active antiretroviral therapy HIV non-Hodgkins lymphoma Pneumocystis carinii
| Introduction |
|---|
|
|
|---|
Fewer opportunistic infections occur in patients who respond to HAART.7 8 9 10 11 12 13 14 Cessation of primary and secondary prophylaxis against some of these opportunistic infections may be safe15 16 17 18 19 20 in these patients.
However, what effect HAART has had on noninfectious complications of HIV infection is uncertain. Kaposis sarcoma has become less common.21 22 23 24 25 26 27 There were predictions that successful antiretroviral therapy might prolong survival but might lead to an increased risk of non-Hodgkins lymphoma (NHL) and other HIV-associated malignancies.28 29 30 Subsequent reports have not yielded consistent conclusions. Some studies suggest little change, whereas other studies have shown a decline of lesser magnitude than that seen for opportunistic infections.22 23 24 25 26 27 31 32
To clarify the impact of HAART on our patient population, we determined the relative frequencies of HIV-related complications in the patients referred to our Pulmonary and Critical Care Medicine Service.
| Materials and Methods |
|---|
|
|
|---|
We collected information regarding standard demographic variables, HAART status, CD4 cell count, viral load (available only during era 2), history of previous opportunistic infections, and ultimate diagnosis for each patient. We defined HAART as the combination of any two reverse transcriptase inhibitors with a protease inhibitor. If there was mention of noncompliance with therapy, non-HAART status was assigned. A past or current diagnosis of Pneumocystis carinii pneumonia (PCP) or cytomegalovirus pneumonia (CMVP) required typical pathologic findings on BAL or lung biopsy specimens. Prior cytomegalovirus (CMV) infection was defined as having had documented CMV pneumonitis, retinitis, colitis, or disseminated infection. Bacterial (community or hospital acquired) pneumonia was diagnosed in those patients with typical clinical, laboratory, and radiographic presentations as well as a response to antibiotics not expected to treat PCP. A diagnosis of NHL required tissue biopsy.
Several patients were seen on more than one occasion within a set time period; only data from the first visit were included for analysis. A few patients were seen during both eras. For these patients, data from the first visit within each time period were included.
Data Analysis
Statistical analysis was performed statistical SPSS software
(SPSS Standard version 8.0; SPSS; Chicago, IL). Categorical data were
compared using
2 testing, and numerical data
were compared using independent samples t testing. Logistic
regression was performed to find independent associations and to obtain
odds ratios (ORs) and confidence intervals (CIs). Significance was set
at p < 0.05.
| Results |
|---|
|
|
|---|
|
0.001). No difference
was found between the two groups with regard to having a prior
diagnosis of PCP or CMV.
|
|
|
|
When the analysis was limited to only the 51 patients receiving HAART (Table 5) , there were no significant associations between PCP, bacterial pneumonia, or NHL, and age, race, gender, CD4 cell count, viral load, duration of HIV positivity prior to referral, or history of PCP or CMV. However, three nonsignificant trends for the development of lymphoma emerged. Lymphoma tended to be more common with lower CD4 cell counts (p = 0.068), higher viral loads (p = 0.055), and history of CMV infection (p = 0.071).
Table 6 shows the results of logistic regression analysis. ORs and CIs for the development of PCP, bacterial pneumonia, and NHL are shown for patients receiving HAART compared with patients in the pre-HAART era. Age, race, gender, CD4 cell count, duration of HIV positivity before referral, and history of prior opportunistic infections were included in the regression model. HAART was protective against PCP (OR, 0.37; CI, 0.16 to 0.89) only when CD4 cell count was not included in the regression model (data not shown). HAART was associated with increased risk of developing bacterial pneumonia (OR, 2.41; CI, 1.12 to 5.17) and NHL (OR, 15.11; CI, 3.14 to 28.32).
|
| Discussion |
|---|
|
|
|---|
We found, as have others, that PCP has become a less common diagnosis in the years since HAART has become widely available. In the pre-HAART era, at our institution, 36.3% of the patients referred to us received a diagnosis of PCP. Only 17.6% of the patients who were receiving HAART received a diagnosis of PCP. Initial regression analysis showed that the receipt of HAART served as a protective factor against the development of PCP. This factor lost significance when CD4 cell count was included in the model. This is not surprising because CD4 cell count is a strong predictor of PCP.
In our select patient population, a diagnosis of bacterial pneumonia was more common for patients receiving HAART than it had been in the pre-HAART era. HAART status and a history of PCP were both predictive factors for the development of bacterial pneumonia. Infection with nonopportunistic organisms is less dependent on the immune status of the patient; therefore, HAART-induced reconstitution of the immune system may not be as protective against bacterial pneumonia pathogens as it is against PCP.
Our results with regards to bacterial pneumonia may be at odds with those reported by Sullivan et al.8 They found that community-acquired pneumonia has become less common in patients who receive HAART. Because our population included cases of hospital-acquired pneumonia as well as community-acquired pneumonia and represented a select group of patients who were referred to a Pulmonary and Critical Care Medicine Service, the two studies may not be directly comparable. Our results suggest that in a group of patients with pulmonary complaints or abnormal chest imaging, the proportion of patients with bacterial pneumonia is higher than it once was. In these patients, HAART is a risk factor for bacterial pneumonia. This most likely reflects a decreased risk for opportunistic infections rather than a direct increase in risk for bacterial pneumonia. Our finding that a prior episode of PCP increases the risk for bacterial pneumonia is in agreement with Sullivan et al.8
The largest difference in our population of patients between the two eras was the increase of NHL. Others have reported either no change or a slight decrease in the occurrence of NHL. In our study, the receipt of HAART was associated with a 15-fold increase in the likelihood of developing NHL.
Duration of HIV positivity and more profound immunosuppression have been believed to be risk factors for the development of NHL.28 33 34 Our analysis demonstrates that the association between HAART and NHL is independent of duration of HIV positivity and CD4 cell count. Before the advent of HAART, longer duration of HIV positivity and level of immunosuppression (CD4 count) were inter-related. In patients responding to HAART, these two factors no longer may be so intimately associated. Long-term B-cell stimulation is also suggested as a risk factor for the development of HIV-related NHL.35 36 It is possible that B-cell stimulation persists or increases in patients receiving HAART, placing them at increased risk for NHL.
Perhaps the development of lymphoma is somehow triggered by the therapy itself. There have been reports of "immune reconstitution syndromes," wherein patients responding to HAART had recrudescence of diseases that had become dormant or developed new inflammatory-type disease.37 38 39 40 41 Report of sarcoidosis had been uncommon in the setting of HIV infection; however, there have been reports of patients developing sarcoid-like lesions after the institution of HAART.42 43 44 It is possible that the newly reconstituted immune system still has uncharacterized deficiencies and that these deficiencies may place a patient at risk for the development of NHL. Further studies will be needed to elucidate the relationship between HAART and NHL.
We also report that NHL development was more common in patients receiving HAART who have a history of CMV-related disease. It seems unlikely that the CMV itself causes malignant transformation of lymphocytes. Prior CMV infection may merely be a marker for relatively more severe immunosuppression before the institution of HAART. Perhaps patients with more profound immunosuppression before the institution of HAART are more likely to have aberrant or incomplete reconstitution of their immune systems and, thus, may be at higher risk for the development of NHL.
The purpose of our study was to determine whether there had been a change in the spectrum of HIV-related complications in the population of patients referred to our Pulmonary and Critical Care Medicine Service. This is a select population and is not necessarily representative of patients with HIV infection in general. However, because virtually all patients with HIV infection at our institution who have pulmonary complaints or abnormalities on chest imaging are referred to our service, it seems unlikely that a selection bias is entirely responsible for the changes that we have noted. In fact, it seems more likely that there have been significant changes in the complications of HIV infection in the era of HAART.
Retrospective studies, by their nature, are subject to bias. By requiring histologic proof of PCP, CMVP, and NHL, we hope to have eliminated some of these potential biases. Duration of documented HIV positivity is an imperfect surrogate for duration of HIV infection, but it was the best available to us.
The relatively small number of patients receiving HAART (n = 51) also limits our study. This may account for our inability to document differences between patients receiving HAART who develop PCP, bacterial pneumonia, or NHL and patients who do not develop PCP, bacterial pneumonia, or NHL (Table 5) . For example, our data suggest that among patients receiving HAART, CD4 cell count is not significantly associated with the development of PCP. This seems unlikely to be true. Certainly, the number of patients for whom viral load measurement was available compromised our ability to find associations with that variable. Perhaps we would be able to find such associations if our population of patients receiving HAART was larger.
We have shown that, in the select population of patients referred to our service, PCP has become less common, whereas bacterial pneumonia and NHL have become more common with the use of HAART. Furthermore, we have identified possible risk factors for the development of such HIV-related diseases. It is critical that, in the face of changes in the natural history of HIV infection, we maintain a high index of suspicion for complications that once may have been unusual but that may be more common in the era of HAART.
| Footnotes |
|---|
Received for publication March 22, 2001. Accepted for publication June 14, 2001.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
H. Khouli, A. Afrasiabi, M. Shibli, R. Hajal, C. R. Barrett, and P. Homel Outcome of Critically Ill Human Immunodeficiency Virus-Infected Patients in the Era of Highly Active Antiretroviral Therapy J Intensive Care Med, December 1, 2005; 20(6): 279 - 285. [Abstract] [PDF] |
||||
![]() |
C. R. Uphold, M. Deloria-Knoll, F. J. Palella Jr, J. P. Parada, J. S. Chmiel, L. Phan, and C. L. Bennett US Hospital Care for Patients With HIV Infection and Pneumonia: The Role of Public, Private, and Veterans Affairs Hospitals in the Early Highly Active Antiretroviral Therapy Era Chest, February 1, 2004; 125(2): 548 - 556. [Abstract] [Full Text] [PDF] |
||||
![]() |
S H Allen, P Brennan-Benson, M Nelson, D Asboe, M Bower, B Azadian, B Gazzard, and J Stebbing Pneumonia due to antibiotic resistant Streptococcus pneumoniae and Pseudomonas aeruginosa in the HAART era Postgrad. Med. J., December 1, 2003; 79(938): 691 - 694. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. W. Brecher, G. Aviram, and P. M. Boiselle CT and Radiography of Bacterial Respiratory Infections in AIDS Patients Am. J. Roentgenol., May 1, 2003; 180(5): 1203 - 1209. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |