Chest Email Content Delivery
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Guest Access | Sign In via User Name/Password
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Taggart, S.
Right arrow Articles by Lipman, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Taggart, S.
Right arrow Articles by Lipman, M.
(Chest. 2002;122:878-885.)
© 2002 American College of Chest Physicians

The Changing Pattern of Bronchoscopy in an HIV-Infected Population*

Simon Taggart, MD; Ronan Breen, MD; Neil Goldsack, MD; Caroline Sabin, PhD; Margaret Johnson, MD and Marc Lipman, MD

* From the Royal Free Hospital HIV Centre, Hampstead, London, UK.

Correspondence to: Simon C. O. Taggart, BSc, MRCP, MD, Consultant Respiratory Physician, Trafford General Healthcare NHS Trust Hospital, Moorside Rd, Davyhulme, Manchester M41 5SL, UK; e-mail: scotaggart{at}hotmail.com


   Abstract
TOP
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Study objectives: Little information exists on the impact of antiretroviral therapies (ARTs) on HIV-related bronchoscopic activity. This study was performed to identify any changes to our pattern of use of bronchoscopy over the last decade, and how this might relate to the introduction of more effective ARTs to our center in 1996.

Design: Retrospective data analysis.

Setting: Academic medical center.

Patients: HIV-positive patients attending the clinic.

Methods: Basic demographic details and bronchoscopy status were collected and compared for all patients with HIV attending our center between 1989 and 1998. Poisson regression analysis was performed to more formally identify the risk factors for bronchoscopy. Individual case notes and bronchoscopic findings were also examined for all patients undergoing bronchoscopy in 1990, 1995, and 1998.

Results: From 1996 to 1998, bronchoscopic rates fell dramatically by 60% (p < 0.0001) despite a linear increase in patients receiving follow-up. Prior use of protease inhibitor (PI)/nonnucleoside reverse transcriptase inhibitor (NNRTI) combinations was significantly associated with a decreased risk of bronchoscopy even after adjusting for CD4 counts. Indications for bronchoscopy and diagnostic yield remained relatively stable in 1990, 1995, and 1998, although rates of pulmonary infection (Pneumocystis carinii pneumonia, bacteria, and virus) requiring bronchoscopy among our HIV population fell significantly from 1990 to 1998.

Conclusion: Improvements in HIV health care are having a dramatic impact on the rates of certain pulmonary infections requiring bronchoscopy. Of these, the introduction of more effective ARTs to our service in 1996 seems most closely related to the temporal decline in bronchoscopy. PI/NNRTI combinations may have additional protective effects to their recognized action on CD4 counts.

Key Words: antiretroviral therapy • bronchoscopy • HIV


   Introduction
TOP
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
HIV-infected individuals with low blood CD4 T-cell counts (< 200 cells/µL) are at increased risk of opportunistic infection and bacterial pneumonias,1 which may be difficult to distinguish solely on clinical grounds. Since 1989, our center has used bronchoscopy and BAL as our first-line investigation of possible opportunistic infection, particularly Pneumocystis carinii pneumonia (PCP), in patients presenting with respiratory symptoms, low CD4 T-cell counts, abnormal chest radiograph (CXR) findings, and abnormal exercise pulse oximetry results.

Following the widespread use of effective prophylaxis in 1989, the proportion of patients acquiring PCP has fallen very significantly, as has the number of patients dying from the disease.2 PCP does, however, remain a significant problem in those individuals who present for the first time with relatively advanced disease or, alternatively, in those who experience difficulty adhering to their medication or who chose not to take it.

Mortality and morbidity from HIV have further declined following the introduction of highly active antiretroviral therapy (HAART) combining inhibitors of HIV-1 protease and reverse transcriptase.3 4 Using these treatments, it is now possible to lower viral loads to undetectable levels and boost blood CD4 T-cell counts,5 6 leading to improved host immunity and protection from infection.

This study was performed to investigate the pattern of use of bronchoscopy in our HIV-positive population over the last decade (from 1989 to 1998). In particular, we were interested to identify the demographic factors that are associated with patients undergoing bronchoscopy in addition to observing any temporal change in bronchoscopic rates that may have arisen as a result of the introduction of more effective antiretroviral therapy (ART) to our center in 1996. To further explore any relationship that might exist between the introduction of more effective ARTs and rates of pulmonary infections requiring bronchoscopy in this population, we selectively analyzed our indications for bronchoscopy, our bronchoscopic diagnostic yield, and our overall bronchoscopic activity for each of 3 years: 1990, 1995, and 1998.


   Materials and Methods
TOP
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Study Setting
The study was performed at a large academic medical center in London (Royal Free Hospital, Hampstead NHS Trust, London, UK), where our HIV clinic (Ian Charleson Day Center) is based. Since its establishment in 1989, the HIV service has expanded almost linearly, with the number of patients receiving regular follow-up increasing from 162 in 1989 to 1,141 in 1998. All patients receiving active follow-up at our HIV clinic between 1989 and 1998 were reviewed.

Study Design and Data Collection
The study was performed in two separate parts, and is based on a retrospective analysis of data collected from patients attending our HIV clinic from 1989 to 1998. Part 1 of the study involved detailed interrogation of our HIV-specific database to characterize the demographic details that are associated with patients undergoing and not undergoing bronchoscopy in each calendar year. Selected demographics (HIV exposure, age at start of year, gender, ethnic group) and clinical factors at the start of the year (first CD4 T-cell count measured in each year, whether the patient had an AIDS diagnosis on January first each year, whether the patient had received ART before January first each year, and whether it was the first year of follow-up for each patient) were recorded. The HIV database was then used to identify which of these factors are independently associated with an increased risk of undergoing bronchoscopy.

In Part 2 of the study, individual case notes and bronchoscopy records of all patients undergoing bronchoscopy in 1990, 1995 (before introduction of more effective ART), and 1998 (after introduction of more effective ART) were examined by hand. This information was used to identify whether any temporal trends were apparent in our pattern of use of bronchoscopy in this population. Information was therefore sought on the following: the indications for bronchoscopy; CXR appearance (normal or abnormal); use of either ART, PCP prophylaxis, or both for at least 3 months prior to bronchoscopy; whether bronchoscopy was performed as a day case or inpatient; whether symptoms had been present for > 1 week or not; outcome of bronchoscopy (whether diagnostic or not); and actual bronchoscopic diagnosis (once results from culture/viral studies were available).

Statistical Analysis
Part 1: Within a given year, the characteristics of those subjects who underwent bronchoscopy were compared to those subjects who did not. Each year was treated as a separate entry; thus, individuals contributed as many entries as years of follow-up, and a patient’s status could change over time, if they underwent more than one bronchoscopy in different years. Comparisons for illustrative purposes were made using {chi}2 tests, t tests, and Mann-Whitney U tests as appropriate, and the level of significance was set at 5%.

To more formally identify the factors that were associated with a patient undergoing bronchoscopy, and to assess whether there had been a reduction in the number of bronchoscopies performed over time, Poisson regression was performed. At each visit, information on whether or not the patient had undergone bronchoscopy since their last clinic visit was obtained. The patient’s most recent CD4 T-cell count and presence of an AIDS diagnosis were also recorded. In addition, prior use of any ART—either nucleoside reverse transcriptase inhibitors (NRTIs), nonnucleoside reverse transcriptase inhibitors (NNRTIs), or protease inhibitors (PIs)—was also recorded. Furthermore, because of a suspicion that patients were more likely to undergo bronchoscopy at the start of follow-up, rather than once their care had been established, each visit was stratified according to whether it occurred within 3 months of their first visit to the clinic. In addition, their gender, risk group, ethnicity, and age at the time when they first visited the clinic were recorded. These latter variables retained the same value at each visit.

Each visit was further stratified according to whether it occurred before or after January 1, 1996 (period 1 and period 2, respectively) when more effective ART was introduced to our center. Poisson regression was performed using PROC GENMOD in the statistical package (SAS Institute; Cary, NC). As each patient contributed multiple entries in the data set, and as it was possible for each patient to have more than one bronchoscopy, generalized estimating equations were used to adjust for the nonindependence of the data. Factors that were significant in univariate analyses were included in a backwards-selection multiple regression analysis, and the level of significance was again set at 5%.

Part 2: {chi}2 and Fisher exact tests were used when appropriate to explore whether there were any differences in the various chosen factors between those patients undergoing bronchoscopy in 1990, 1995, and 1998. As some patients underwent bronchoscopy on more than one occasion, each bronchoscopy was treated as a separate new episode and therefore contributed to the whole data set. Since a diagnostic bronchoscopy was defined as one in which either evidence of pulmonary infection or other pathologic disease was detected (eg, Kaposi sarcoma [KS] or pulmonary hemorrhage), those bronchoscopies associated with detection of more than one pathogen or pathology were treated as single bronchoscopies when calculating diagnostic yield. Logistic regression analysis was used to determine the factors associated with patients undergoing a diagnostic bronchoscopy. The level of significance was set at 5%.


   Results
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
From from 1989 to 1998, a total of 2,722 bronchoscopies were performed in our department (excluding those relating to research protocols), of which 634 cases (23.3%) were HIV related. As a percentage of total bronchoscopic activity, the number of HIV-related bronchoscopies varied significantly ({chi}2 test, p < 0.001) between 1990 (40 of 144 bronchoscopies, 27.8%), 1995 (128 of 355 bronchoscopies, 36.1%), and 1998 (57 of 334 bronchoscopies, 17.1%). The proportion of HIV-related bronchoscopies performed as a day case also varied significantly (p < 0.001) between 1990 (5.1%), 1995 (40.8%), and 1998 (31.6%).

Part 1, Changing Characteristics of Those Undergoing Bronchoscopy
HIV-positive patients undergoing bronchoscopy (n = 390) underwent between one bronchoscopy and six bronchoscopies (median, one bronchoscopy) during follow-up. The number of bronchoscopies performed and the number of patients receiving follow-up at our center in each year are shown in Table 1 . The percentage of patients undergoing bronchoscopy in each year remained relatively stable between 1989 and 1995. However, from 1996 onwards, there was a sharp decline in the percentage of patients undergoing bronchoscopy each year despite a linear increase in the number of patients receiving follow-up.


View this table:
[in this window]
[in a new window]

  		Table 1.. No. of HIV-infected Patients Undergoing Bronchoscopy and No. of Patients Receiving Follow-up in Each Year

 
Selected demographics and clinical details of HIV-infected patients undergoing or not undergoing bronchoscopy for the whole 10-year period are presented in Table 2 . Overall, the population under study consisted of predominantly young homosexual men of white origin at various stages of disease progression. As expected, patients undergoing bronchoscopy had significantly lower CD4 T-cell counts than their counterparts who did not undergo bronchoscopy. They were also more likely to have had a previous AIDS diagnosis and to have received ART, although a third of patients who underwent bronchoscopy were presenting for the first time with presumably relatively advanced disease.


View this table:
[in this window]
[in a new window]

  		Table 2.. Selected Demographics and Clinical Data for Patients Who Did and Did Not Undergo Bronchoscopy From 1989 Through 1998*

 
Selected clinical data for patients who did or did not undergo bronchoscopy in each calendar year are presented in Table 3 . When each year was examined in greater detail, it was apparent that the proportion of all patients being seen for the first time gradually decreased, from approximately 59% in 1989 to 20% in 1998, and that this fall was most obvious in those patients not undergoing bronchoscopy. Median CD4 T-cell counts remained relatively stable in the population as a whole up until 1997, when they were seen to increase only in the patients who underwent bronchoscopy (range, 25 to 72 cells/µL from 1989 to 1996; range, 140 to 155 cells/µL from 1997 to 1998). A notable increase in the use of ART and minor rise in the prevalence of AIDS diagnoses among the population who did not undergo bronchoscopy was observed with time (14% and 16% in 1989, and 53% and 22% in 1998, respectively), perhaps reflecting the growing experience of this particular group of patients.


View this table:
[in this window]
[in a new window]

  		Table 3.. Selected Clinical Data for Patients Who Did or Did Not Undergo Bronchoscopy in Each Calendar Year

 
Poisson Regression Analysis: Univariate results from the Poisson regression analysis are shown in the first panel of Table 4 . A significant 60% reduction in the proportion of patients undergoing bronchoscopy from 1996 onwards (period 2) compared with earlier years (period 1) was detected (p < 0.0001). High CD4 T-cell counts, prior use of PI/NNRTI therapy, and the presence of a previous AIDS diagnosis were all associated with a significantly decreased risk of undergoing bronchoscopy. In contrast, older age, being seen within 3 months of first visit, and prior use of NRTI therapy were all significantly associated with an increased risk of undergoing the procedure. No associations were found for gender, risk group, or ethnic origin.


View this table:
[in this window]
[in a new window]

  		Table 4.. Relative Event Rates, 95% CIs, and p Values for Undergoing Bronchoscopy: Unadjusted and Adjusted Results From Poisson Regression

 
In general, these associations remained significant in multivariate analyses (second panel of Table 4 ) except for prior NRTI therapy and prior AIDS diagnoses, which showed no independent associations. It is noteworthy that, even after allowing for the changing nature of the patients receiving follow-up and their changing CD4 counts, there was still a 45% reduction in the proportion of bronchoscopies performed between 1996 and 1998, compared to earlier calendar years.

Part 2, Analysis of HIV-Related Bronchoscopies Performed in 1990, 1995, and 1998
A total of 225 HIV-related bronchoscopies were performed in 181 patients. Records from 216 bronchoscopies (174 patients) were complete and were therefore included in the analysis. For these 174 patients, rates of PCP prophylaxis were significantly higher in 1995 (61.7%, p = 0.006 by {chi}2 test) compared with 1990 and 1998 (46.4% and 34.6%, respectively). In contrast, rates of ART use did not differ significantly between the years (39.3%, 25.5%, and 32.7% for 1990, 1995, and 1998, respectively; p = 0.33, {chi}2 test).

Indications for Bronchoscopy: Five clinical indications for bronchoscopy were identified: suspected new episode of opportunistic infection, repeat bronchoscopy for PCP treatment failure, repeat bronchoscopy for slow-to-resolve PCP (previously diagnosed by bronchoscopy), suspected KS/known KS under surveillance, and "other" indications, comprising a total of three other indications (pulmonary hemorrhage, pulmonary fibrosis, and left hilar mass).

The results of how these five indications changed with time are illustrated in Table 5 . From this it can be seen that the clinical indications for bronchoscopy appeared to change significantly with time (Fisher exact test, p = 0.001), although these changes seemed to be limited to a relatively small number of bronchoscopies performed for indications other than to exclude a new episode of suspected opportunistic infection (which accounted for almost 86% of all bronchoscopies). In particular, the number of bronchoscopies performed for PCP treatment failure fell from 5 of 39 bronchoscopies (12.8%) in 1990, to 0 of 57 bronchoscopies in 1998.


View this table:
[in this window]
[in a new window]

  		Table 5.. Indications for Patients Undergoing Bronchoscopy in 1990, 1995, and 1995*

 
CXR Appearance: Overall, 68 of 216 bronchoscopies (31.5%) were performed despite the presence of a normal CXR appearance. No significant difference in the proportion of bronchoscopies with a normal CXR appearance was detected between the 3 years ({chi}2 test, p = 0.67): 10 of 39 bronchoscopies (25.6%) in 1990, 40 of 120 bronchoscopies (33.3%) in 1995, and 18 of 57 bronchoscopies (31.6%) in 1998.

Diagnostic Yield: Information relating to diagnostic yield is displayed in Table 6 . This illustrates that there was a tendency for our diagnostic yield to fall with time, from 59% in 1990 to 45.6% in 1998, although this change did not reach statistical significance ({chi}2 test, p = 0.42). PCP was the most common pathologic diagnosis, accounting for approximately 23% of all bronchoscopy findings, and this proportion remained broadly similar between the years. In contrast, the proportion of viral and bacterial infections diagnosed decreased from 1990 to 1998 (23.1 to 5.3% and 20.5 to 7%, respectively).


View this table:
[in this window]
[in a new window]

  		Table 6.. Bronchoscopic Diagnostic Yield for Various Categories of Pulmonary Infection, KS, and Other Conditions in 1990, 1995, and 1998*

 
In order to observe for any temporal change in diagnostic yield that may have occurred with time, the last four diagnostic categories were combined (mycobacteria, KS, fungi, and others) and compared with the first three categories (PCP, bacteria, and virus) to ease computation of the {chi}2 statistic. This revealed a significant change in diagnostic yield between the years ({chi}2 test, p = 0.05).

To further investigate whether the number of pulmonary infections requiring bronchoscopy within this HIV-infected population had changed with time, we expressed the number of diagnostic bronchoscopies for each category of infection as a percentage of the total number of patients receiving regular follow-up each year. The results of these analyses are illustrated in Table 7 and reveal a significant downward trend in the number of PCP, bacterial, and viral (but not mycobacterial) infections. It is noteworthy that the observed reductions in pulmonary infections were already present by 1995, but continued to fall by > 60% when measured again in 1998.


View this table:
[in this window]
[in a new window]

  		Table 7.. Pulmonary Infections Diagnosed at Bronchoscopy*

 
Comparison of Diagnostic Yield Between Patients Receiving and Not Receiving ART: The total diagnostic yield for all 3 years (1990, 1995, 1998) was not significantly different in those patients receiving ART compared to those who were not, although there was a trend for the yield to be higher in the latter group (41.9% and 53.2%, respectively; p = 0.18). Similarly, no significant change in diagnostic yield was observed with time between the two groups (p = 0.46).

A noticeable and significant doubling in the rate of PCP infection was seen in patients not receiving ART compared with those receiving ART (27.3% vs 12.9%, respectively; p = 0.04). Over the 3 years of study, PCP was detected in 50 bronchoscopies, of which 6% were associated with prior use of ART, 10% with ART plus PCP prophylaxis, 20% with PCP prophylaxis, and 64% with no treatment. No other significant differences were detected for diagnostic bronchoscopies between the two groups, although there was a tendency for the yield of bacteria, viruses, mycobacteria, and KS to be higher in those not receiving ART compared to those receiving ART: 12.3% vs 4.8% (p = 0.16), 10.4% vs 8.1% (p = 0.79), 9.1% vs 6.5% (p = 0.72), and 18.5% vs 9.7% (p = 0.60%), respectively.

Risk Factors for Undergoing a Diagnostic Bronchoscopy: The main risk factor associated with undergoing a diagnostic bronchoscopy was the presence of an abnormal CXR appearance (p = 0.0001). In addition, borderline relationships were detected for prior use of ART (p = 0.07), PCP prophylaxis (0.07), symptom duration > 1 week (p = 0.07), and calendar year (1995 compared to 1990, p = 0.03; 1998 compared to 1990, p = 0.02). However, in multivariate analysis, an abnormal CXR appearance was the only factor that remained significant (odds ratio, 12.36; 95% confidence interval [CI], 5.65 to 27.03; p = 0.0001).


   Discussion
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
This study has demonstrated a highly significant and abrupt 60% decline in our use of fiberoptic bronchoscopy in HIV-positive patients between 1996 and 1998 compared to previous years, despite a linear increase in the number of patients receiving regular follow-up. This finding is remarkably similar to the 59% gradual decline in HIV-related bronchoscopic examinations that occurred at a large academic medical center in the United States between 1991 and 1997, which is believed to have resulted from changes in the insurance details of patients and the introduction of a managed care pathway for the empiric treatment of PCP.7 The odds of undergoing bronchoscopy have also been shown to reduce significantly and gradually from 1996 onwards in a previous European study conducted by Murri et al.8 Our study has also shown that bronchoscopic rates remained relatively stable from 1989 to 1995, in keeping with an earlier British investigation involving 580 patients with HIV between 1983 and 1993.9

We believe that improvements in HIV health care are likely to underlie this decline in HIV bronchoscopy, which was temporally associated with the introduction of more effective ART to our center in 1996. Prior to this, the use of ART had been largely restricted to NRTI monotherapy for patients with relatively advanced disease, in whom an earlier study had shown improved survival.10 Paradoxically, prior use of NRTI therapy in our study was associated with an apparent increased risk of bronchoscopy. We suspect that those subjects with experience of NRTI had been receiving ART for some time, and that any (probably relatively minor) benefit had now disappeared. Hence, this marker was in fact a surrogate for patients with advanced disease who would be at high risk of pneumonia. This is supported by the fact that the relationship between NRTI and increased risk of bronchoscopy is lost after adjustment is made for CD4 T-cell count.

Following the publication of the Delta Trial in September 1995,11 standard zidovudine monotherapy was replaced by dual nucleoside therapy in 1996, which in turn has been superseded by HAART combinations of NRTI/NNRTI and PI therapy in 1997 and 1998. In this study, prior use of PI/NNRTI therapy was associated with a significantly decreased risk of undergoing bronchoscopy. Of interest, is the observation that this association occurred independently of changes in CD4 T-cell counts. Limited evidence from an in vitro investigation suggests that treatment with HAART leads to improved neutrophil and macrophage function,12 which is normally depressed in HIV infection.13 It is therefore possible that PI/NNRTI-related improvements in the pulmonary host defense protect against both primary viral and secondary bacterial infection, and therefore reduce the need for bronchoscopy to exclude opportunistic infection. This hypothesis is supported by the finding that rates of both viral and bacterial bronchoscopic diagnoses (in addition to PCP) among patients receiving regular follow-up fell by approximately 60% in 1998 compared with 1995.

In contrast to PI/NNRTI therapy, PCP prophylaxis appears to protect against PCP and bacterial pneumonia by virtue of its direct antimicrobial properties.14 It is interesting to note that rates of PCP and bacterial pulmonary infection fell in 1995 compared with 1990, possibly as a result of higher rates of PCP prophylaxis among the patients who underwent bronchoscopy. We do, however, remain unclear as to why rates of viral infection appear to have fallen before the introduction of more effective ART to our center in 1996. We are aware of a small number of patients at this time who were using dual-NRTI therapy, but whether this may have had an influence on rates of viral infection diagnosed by bronchoscopy remains open to question.

We do not believe that the observed decline in the use of bronchoscopy resulted from a change in our bronchoscopic practice per se, as the indications for patients undergoing the procedure have remained broadly similar over the 10-year period of study, with up to 86% of procedures being performed to exclude a new episode of suspected opportunistic infection. When taken together, we have in fact demonstrated a significant change in the pattern of clinical indications for bronchoscopy with time, but suspect that this is largely because of a fall in the small number of patients failing PCP treatment who have in the past required bronchoscopy to exclude the presence of a copathogen. We stress that the move toward performing more bronchoscopies as outpatient day cases is unrelated to any economic factors (Balfe and Mohsenifar7 ), as access to all HIV health care including bronchoscopy is free to all patients in the National Health Service from the point of entry. We suspect that the significant rise in day-case procedures most probably relates to the increased patient awareness that has inevitably accompanied the growing experience of those receiving follow-up, and recognition of the need for earlier presentation.

Consistency in our bronchoscopic practice is also evidenced by the results of our diagnostic yield, which also does not appear to have changed significantly with time, and is likely therefore to reflect a degree of constancy in our clinical acumen at targeting those patients in need of a bronchoscopy. However, we do appear to operate a relatively low threshold for bronchoscopy despite the presence of a normal CXR appearance in up to one third of our patients with symptoms and worrying signs, although this proportion has also remained stable over the study period.

Unfortunately, information on the use of PCP prophylaxis was only available for analysis in those patients undergoing bronchoscopy in 1990, 1995, and 1998. This may be important, as long-term treatment with cotrimoxazole is associated with lower rates of bacterial infection,15 which often require bronchoscopy to exclude PCP. However, we suspect that overall rates of PCP prophylaxis have changed little during the study period, as evidenced by the relative stability of median CD4 T-cell counts in the population who did not undergo bronchoscopy. Furthermore, use of PCP prophylaxis does not appear to have greatly influenced our decision to perform bronchoscopy, as evidenced by the relatively high rates of PCP prophylaxis among the population who underwent bronchoscopy. PCP prophylaxis is, however, associated with a significant failure rate (18% when used as secondary prophylaxis over a 3-year period15 ), as exemplified in our study in which PCP was actually detected in approximately one third of patients receiving regular prophylaxis.

In conclusion, we believe that recent advancements in HIV health care have had a dramatic impact on our rates of bronchoscopic examination. Of these, the introduction of more effective ART regimens to our center in 1996 appears most closely related to the observed temporal decline in bronchoscopy. We have shown that this dramatic fall in bronchoscopic numbers occurred as a result of a significant reduction in the number of pulmonary infections requiring bronchoscopy to exclude opportunistic infection and not as a result of a change in our bronchoscopic practice per se. The apparent additional protective properties of ART regimens containing PI/NNRTI combinations outside of their known effects on CD4 T cells merits further investigation.


   Footnotes
 
Abbreviations: ART = antiretroviral therapy; CI = confidence interval; CXR = chest radiograph; HAART = highly active antiretroviral therapy; KS = Kaposi sarcoma; NNRTI = nonnucleoside reverse transcriptase inhibitor; NRTI = nucleoside reverse transcriptase inhibitor; PCP = Pneumocystis carinii pneumonia; PI = protease inhibitors

Received for publication May 15, 2001. Accepted for publication April 23, 2002.


   References
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Wallace, JM, Rao, AV, Glassroth, J, et al (1993) Respiratory illness in persons with HIV infection. Am Rev Respir Dis 148,1523-1529[ISI][Medline]
  2. Peters, BS, Beck, EJ, Coleman, DG, et al Changing disease patterns in patients with AIDS in a referral centre in the United Kingdom: the changing face of AIDS. BMJ 1991;302,203-206[ISI][Medline]
  3. Hogg, RS, O’Shaughnessy, MV, Gataric, N, et al Decline in deaths from AIDS due to new antiretrovirals [letter]. Lancet 1997;349,1294[ISI][Medline]
  4. Reed, JB, Schwab, IR, Gordan, JR, et al Regression of cytomegalovirus retinitis associated with protease-inhibitor treatment in patients with AIDS. Am J Opthalmol 1997;124,199-205[Medline]
  5. Hammer, SM, Squires, KE, Hughes, MD, et al A controlled trial of two nucleoside analogues plus indinavir in patients with human immunodeficiency virus infection and CD4 counts of 200 per cubic millimetre or less. N Engl J Med 1997;337,725-733[Abstract/Free Full Text]
  6. Collier, AC, Coombs, RW, Schonfield, DA, et al Treatment of human immunodeficiency virus infection with saquinavir, zidovudine and zalcitabine. N Engl J Med 1996;334,1011-1017[Abstract/Free Full Text]
  7. Balfe, D, Mohsenifar, Z Downward trend in bronchoscopies performed between 1991 and 1997. Chest 1999;116,238-242[Abstract/Free Full Text]
  8. Murri, R, Cingolani, A, Limongelli, P, et al Reduction of bronchoscopy performances should be included in the advantages of highly active antiretroviral therapy (HAART). J Acquir Immune Defic Syndr 2000;25,286-287[Medline]
  9. Taylor, IK, Coke, RJ, Clarke, J, et al Pulmonary complications of HIV disease: 10 years retrospective evaluation of yields from bronchoalveolar lavage, 1983–93. Thorax 1995;50,1240-1245[Abstract/Free Full Text]
  10. Fischl, M, Richman, D, Grieco, M, et al The efficacy of AZT in the treatment of patients with AIDS and ARC. N Engl J Med 1987;317,185-191[Abstract]
  11. Choo, V Combination superior to zidovudine in Delta Trial [letter]. Lancet 1995;346,895[Medline]
  12. Mastroianni, CM, Lichtner, M, Mengoni, F, et al Improvement in neutrophil and monocyte function during highly active antiretroviral treatment of HIV-infected patients. AIDS 1999;13,883-890[CrossRef][ISI][Medline]
  13. Lipman, MCI, Johnson, MA, Poulter, LW Functionally relevant changes occur in HIV-infected individuals alveolar macrophages prior to the onset of respiratory disease. AIDS 1997;11,765-772[CrossRef][ISI][Medline]
  14. Bozette, SA, Finkelstein, DM, Spector, SA, et al A randomized trial of three anti-Pneumocystis agents in patients with advanced human immunodeficiency virus infection. N Engl J Med 1995;332,693-699[Abstract/Free Full Text]
  15. Hardy, WD, Feinberg, J, Finkelstein, DM, et al A controlled trial of trimethoprim-sulfamethoxazole or aerosolized pentamidine for secondary prophylaxis of Pneumocystis carinii pneumonia in patients with the acquired immunodeficiency syndrome. N Engl J Med 1992;327,1842-1848[Abstract]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Taggart, S.
Right arrow Articles by Lipman, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Taggart, S.
Right arrow Articles by Lipman, M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS