Download Prognostic Factors for Survival in Human Immunodeficiency Virus

Prognostic Factors for Survival in Human
Immunodeficiency Virus–associated Pulmonary
Arterial Hypertension
Hilario Nunes, Marc Humbert, Olivier Sitbon, Jane H. Morse, Zemin Deng, James A. Knowles,
Catherine Le Gall, Florence Parent, Gilles Garcia, Philippe Hervé, Robyn J. Barst, and Gérald Simonneau
UPRES 2705, Centre des Maladies Vasculaires Pulmonaires, Service de Pneumologie et Réanimation Respiratoire, Hôpital Antoine Béclère,
Assistance Publique-Hôpitaux de Paris, Université Paris-Sud, Clamart, France; and the Departments of Medicine, Pediatrics, and Psychiatry,
Columbia University College of Physicians and Surgeons, The New York State Psychiatric Institute, New York, New York
We report a large monocentric case series of 82 patients with human
immunodeficiency virus–associated pulmonary arterial hypertension (PAH). No germline mutations of the PPH1 gene (bone morphogenetic protein receptor-II) were found in any of the 19 patients
tested. PAH was the direct cause of death in 72% of cases. Survival
rates of the overall population at 1, 2, and 3 years were 73, 60, and
47%, respectively. Survival was significantly poorer in patients in
New York Heart Association functional class III–IV at the time of
diagnosis, as compared with those in functional class I–II with respective rates of 60, 45, and 28% versus 100, 90, 84% at 1, 2, and
3 years (p ⬍ 0.0001). Subsequently, we analyzed prognostic factors
in patients in functional class III–IV. Univariate analysis indicated
that CD4 lymphocyte count of more than 212 cells mm⫺3, the use
of combination antiretroviral therapy (CART), and epoprostenol
infusion were related with a better survival. On multivariate analysis
only CD4 lymphocyte count was an independent predictor of survival, presumably because CART and epoprostenol infusion were
strongly linked in our patient population. These results suggest that
patients with severe human immunodeficiency virus–associated
PAH should be considered for long-term epoprostenol infusion in
association with CART.
Keywords: epoprostenol; combination antiretroviral therapy; human
immunodeficiency virus; pulmonary hypertension
Pulmonary arterial hypertension (PAH) is a severe condition
leading to progressive right-heart failure and ultimately death
(1). This disease results from chronic obstruction of small pulmonary arteries, which is due at least in part from endothelial
and vascular smooth muscle cell dysfunction and proliferation
(2). PAH is a well-recognized complication of human immunodeficiency virus (HIV) infection (3–9). Although, HIVassociated PAH occurs in every risk group for HIV infection
regardless of the degree of immunosuppression, only 0.5% of
HIV-infected individuals will develop PAH, suggesting that
displaying this condition requires an additional predisposition
(9). The recent discovery that a significant proportion of patients
with primary pulmonary hypertension is associated with germline mutations of the bone morphogenetic protein receptor-II
(BMPR2) gene suggests that dysfunctional transforming growth
factor-␤ signaling could lead to abnormal proliferation of pulmonary vascular cells (10–12). Similarly, it has been proposed
(Received in original form April 16, 2002; accepted in final form February 11, 2003)
Supported by grants from Legs Poix, Université Paris-Sud, INSERM, and AFM.
Correspondence and requests for reprints should be addressed to Dr. Marc
Humbert, M.D., Service de Pneumologie et Réanimation Respiratoire, Hôpital
Antoine Béclère, 157 rue de la Porte de Trivaux, 92140 Clamart, France. E-mail:
[email protected]
Am J Respir Crit Care Med Vol 167. pp 1433–1439, 2003
Originally Published in Press as DOI: 10.1164/rccm.200204-330OC on February 25, 2003
Internet address: www.atsjournals.org
that genetically predisposed individuals may have abnormal
growth factor signaling in response to HIV, which in turn
promotes abnormal cell proliferation.
Mortality of patients with HIV-associated PAH is usually
related to pulmonary hypertension itself rather than to another complication of HIV infection (3–9), and PAH is an
independent predictor of death in these patients (6). Since
the start of the epidemic, the management of HIV infection
has undergone considerable change as a result of better access
to care, improvements in prophylaxis against opportunistic
infections, and new antiretroviral therapeutic options particularly combination antiretroviral therapy (CART) including
a protease inhibitor (13–15). However, the effect of CART
in HIV-associated PAH remains unknown. Long-term intravenous epoprostenol therapy has been shown to improve hemodynamics, exercise tolerance, and survival in patients with
severe primary pulmonary hypertension (16, 17). Similarly,
this treatment might be effective in the setting of HIV-associated PAH on the basis of small case series (18).
In this retrospective study we report a large monocentric
case series of 82 HIV-infected patients with PAH, 20 of whom
have been treated with epoprostenol. We particularly aimed
to determine prognostic factors for survival in this population.
In addition, we analyzed whether BMPR2 germline mutations may play a part in this condition.
METHODS
Patients
From April 1986 to September 2000, 82 consecutive patients with HIVassociated PAH were admitted in our institution. Twenty patients were
reported previously (5). Each patient fulfilled the diagnostic criteria of
the National Institutes of Health Registry on PAH (19). Secondary
causes of PAH were excluded on the results of history, physical examination, chest radiography, pulmonary function testing, perfusion lung
scan and/or pulmonary angiography, and two-dimensional echocardiography. Patients with portal hypertension and/or cirrhosis were also
excluded. Hemodynamic evaluation was performed using the standard
techniques (20). Pulmonary vascular resistance index was defined as
mean pulmonary arterial pressure (Ppa)—pulmonary wedge pressure
divided by cardiac index (CI). Acute vasodilator response was tested
with a short-term intravenous infusion of epoprostenol until 1995 or
inhaled nitric oxide after 1995 (20). A positive response to acute vasodilator challenge was defined by a fall both in Ppa and pulmonary vascular
resistance index more than or equal to 20%. All the patients received
conventional medical therapy for PAH that could consist of warfarin,
diuretics, supplemental oxygen, and oral vasodilators. The choice of
concomitant antiviral therapy was left to referring physician judgment.
Molecular Methods
Nineteen consecutive patients with HIV-associated PAH were screened
for BMPR2 mutations, as described previously (10, 21, 22).
1434
AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 167 2003
Epoprostenol Therapy
Since 1995, epoprostenol therapy was offered to patients who were in
New York Heart Association (NYHA) functional class III or IV despite
optimal conventional medical therapy. All patients provided informed
consent. Epoprostenol was infused continuously, as described previously (23).
Statistical Analysis
Analysis was performed using the Statview version 5.0 statistical
package (SAS Institute, Cary, NC). All results are expressed as mean ⫾
SD. Groups were compared by unpaired Student’s t test and ␹2 test
when appropriate. For patients treated with epoprostenol, one-way
analysis of variance with repeated measures was performed for the
6-minute walk test and hemodynamic values obtained at baseline after
3 months of treatment and at the last scheduled visit. Multiple comparisons were made when the F value was statistically significant.
For the survival analysis of the overall population and of the patients
according to NYHA functional class, we used the date of PAH diagnosis
as the start point to estimate survival. For the survival analysis of the
patients in NYHA functional class III–IV, the start point was the time
they first became class III–IV. The probability of survival of each group
was estimated by the Kaplan–Meier method and compared by the logrank test. We determined the prognostic factors for survival in the
patients in NYHA functional class III–IV by performing univariate and
multivariable analysis. Univariate analysis based on the proportional
hazards model was used to examine the relationship between survival
and selected demographic, medical history, and hemodynamic variables
measured at initial right-heart catheterization. The results are expressed
as hazard ratios with 95% confidence intervals. For continuous variables, we chose to separate patients into two groups on both sides of
the median value. Multivariable analysis based on the Cox proportional
hazards regression model was used to examine the independent effect of
each variable on survival, controlling for possible confounding variables.
RESULTS
Demography, Biology and BMPR2 Germline
Mutations Screening
The main demographic features of the patients are shown in
Table 1. In our population, the major risk factor for HIV infection
was intravenous drug use. Among the 48 drug-addict patients, 10
had a history of injection of solutions derived from pills or tablets
intended for oral use. At the time of PAH diagnosis, blood CD4
lymphocyte count was 269 ⫾ 219 cells mm⫺3 (range, 7–1,073 cells
mm⫺3) and was under 200 cells mm⫺3 in 52% of the patients.
Viral load, which was available for 35 patients, ranged from an
undetectable value to 445,000 copies ml⫺1 that is from 0 to 5.6
log 10 with a mean of 3.2 log 10. The repartition of patients
according to CDC classification was Stage A (asymptomatic):
46%, Stage B (neither asymptomatic nor AIDS): 17%, Stage C
(AIDS): 37%. The time interval between the diagnosis of HIV
infection and PAH was 6.4 ⫾ 3.7 years (range, 0–14 years). In
nine patients, HIV positivity was discovered at the time of PAH
diagnosis. No BMPR2 mutations were found in any of the 19
patients with HIV-associated PAH who were tested.
tients (13%) were treated with calcium-channel blockers (diltiazem n ⫽ 7 and nifedipine n ⫽ 4).
Results of Epoprostenol Therapy
Twenty patients received continuous intravenous epoprostenol
infusion and were monitored during 17 ⫾ 13 months (range,
6–47 months) after initiation of treatment. The risk factor for
HIV infection was intravenous drug use in eight patients, homosexual contact in five, heterosexual contact in two, and transfusion in five. Antiretroviral therapy included CART in 16, zidovudine in 1, zidovudine plus lamivudine in 1, and lamivudine
plus stavudine in 2 patients. Epoprostenol was started within 3
months after PAH diagnosis in all the patients except two who
were NYHA functional class II at the time of PAH diagnosis
and received conventional therapy for 66 and 20 months before
they became NYHA functional class III–IV and were subsequently treated with epoprostenol.
Results at 3 months. The main results are shown in Table 2.
All the 20 patients were evaluated on NYHA functional class,
6-minute walk test, and right-heart catheterization. Doses of epoprostenol were 8 to 24 ng kg⫺1 minute⫺1 (15 ⫾ 4 ng kg⫺1 minute⫺1).
NYHA functional class improved in 19 patients and was unchanged in one. Exercise capacity improved in 19 out of the 20
patients, with a mean increase in the 6-minute walk test of 183 m
from baseline (p ⬍ 0.05). The results of right-heart catheterization
were also significantly better in comparison with baseline values
(Table 2). CD4 lymphocyte count (n ⫽ 13) and viral load (n ⫽
11) did not change significantly at 3 months in comparison with
baseline values (270 ⫾ 195 vs. 409 ⫾ 274 cells mm⫺3 and 2.5 ⫾
1.7 vs. 2.8 ⫾ 1.9 log 10, respectively).
Results at last scheduled visit. The main results are shown in
Table 2. Twelve patients were evaluated on NYHA functional
class, 6-minute walk test, and right-heart catheterization. The
following data only concern these 12 patients who had 3 months
as well as last follow-up evaluation. The last visit was achieved
on average 17 months after the initiation of epoprostenol therapy
(range 6–47 months) when doses of epoprostenol were 10 to 40
ng kg⫺1 minute⫺1 (22 ⫾ 9 ng kg⫺1 minute⫺1). Sustained improvement was demonstrated in NYHA functional class, 6-minute
walk test, as well as in hemodynamic measures (Table 2). CD4
lymphocyte count (n ⫽ 11) and viral load (n ⫽ 9) did not change
significantly at the last scheduled visit in comparison with baseline values (280 ⫾ 183 vs. 327 ⫾ 178 cells mm⫺3 and 2.3 ⫾ 1.2
vs. 2.6 ⫾ 2.0 log 10, respectively).
Complications. Minor complications due to the use of epoprostenol were frequent (jaw pain, diarrhea, flushing, headaches,
nausea, and vomiting). Additional problems included irritation
or infection at the catheter site requiring simple outpatient care.
Serious complications due to the drug-delivery system were not
uncommon but not fatal in this series of patients (three catheter
dysfunctions in three patients and seven catheter-related sepsis
in five patients).
Mortality and Survival
Clinical and Hemodynamic Findings
The baseline clinical and hemodynamic findings of the patients
are presented in Table 1. The time interval between onset of
symptoms and diagnosis of PAH was 9 ⫾ 13 months. The frequency of Raynaud’s phenomenon was 22% and that of antinuclear antibodies positivity was 14%. Lastly, 17% of the patients
had received amphetamine drugs. Eighty percent of the patients
received warfarin. In the other cases, anticoagulation therapy
was unadvisable because of thrombopenia (⬍ 75,000 per mm3;
n ⫽ 8) or poor medical compliance (n ⫽ 7); one trans-sexual patient was taking intramuscular feminine hormones. Although
there was no responder to acute vasodilator challenge, 11 pa-
Until the end of observation period, 42 patients were alive, 39
died, and 1 was lost to follow-up. Death was related to viral
infection in two cases (multifocal progressive leucoencephalitis
and lymphoma), whereas it was caused by PAH itself in 28
cases (right-ventricular failure, n ⫽ 23; sudden death, n ⫽ 5). The
other causes of mortality were suicide in two cases, Pseudomonas
aeruginosa pneumonia in one, hemoperitonitis in one and unknown in the 5 remaining cases. Seven autopsies were performed.
Histologic examination revealed the classic pattern of hypertensive arteriopathy with plexiform lesions in six cases and pulmonary
veno-occlusive disease in the remaining one. There was no evidence of intravascular foreign bodies in three drug-addict
Nunes, Humbert, Sitbon, et al.: HIV-associated Pulmonary Hypertension
1435
TABLE 1. BASELINE CHARACTERISTICS OF THE PATIENTS
Age, yr
Men/women
HIV risk factors
IV drug use
Homosexual
Heterosexual
Transfusion
Unknown
CD4 count, cells mm⫺3
Viral load, log 10†
Syncope or near syncope
Six-minute walk test, m
Hemodynamic measures
mRAP, mm Hg
Ppa, mm Hg
CI, L minute⫺1 m⫺2
PVRI, IU m⫺2
SvO2, %
Acute vasodilator response‡
Warfarine
Calcium-channel blockers
Epoprostenol infusion
CART
Overall
Population
(n ⫽ 82)
Patients in
NYHA Class I–II
(n ⫽ 24)
p Value*
Patients in
NYHA Class III–IV
(n ⫽ 58)
34 ⫾ 6
45/37
34 ⫾ 8
15/9
NS
NS
35 ⫾ 6
30/28
48 (59)
14 (17)
9 (11)
9 (11)
2 (2)
269 ⫾ 219
3.2 ⫾ 1.7
32 (39)
357 ⫾ 155
16
4
2
2
⫾5
⫾ 11
⫾ 0.7
⫾8
⫾ 10
0
66 (80)
11 (13)
20 (24)
39 (48)
9
51
2.6
18
59
(67)
(17)
(8)
(8)
0
242 ⫾ 191
3.3 ⫾ 1.1
5 (21)
429 ⫾ 90
NS
NS
⬍ 0.05
⬍ 0.03
⫾6
⫾ 10
⫾ 0.8
⫾6
⫾8
0
20 (83)
7 (29)
2 (8)§
12 (50)
NS
⬍ 0.03
⬍ 0.001
⬍ 0.005
⬍ 0.006
NS
NS
⬍ 0.02
⬍ 0.05
NS
32 (55)
10 (17)
7 (12)
7 (12)
2 (3)
280 ⫾ 230
3.1 ⫾ 1.9
27 (47)
328 ⫾ 164
NS
8
47
3.0
14
64
⫾4
⫾ 12
⫾ 0.6
⫾8
⫾ 10
0
46 (79)
4 (7)
18 (31)
27 (46)
10
53
2.5
20
57
Definition of abbreviations: CART ⫽ combination antiretroviral therapy refers to the combination of at least three antiretroviral
agents including a protease inhibitor; CI ⫽ cardiac index; HIV ⫽ human immunodeficiency virus; IV ⫽ intravenous; mRAP ⫽ mean
atrial pressure; NS ⫽ not significant; NYHA ⫽ New York Heart Association; Ppa ⫽ mean pulmonary arterial pressure; PVRI ⫽
pulmonary vascular resistance index; SvO2 ⫽ mixed venous oxygen saturation.
Normal ranges: CI: 2.8–4.2 L minute⫺1 m⫺2; mRAP ⫽ 0–8 mm Hg; Ppa ⫽ 10–22 mm Hg; PVRI ⫽ 0.95–2.55 UI m⫺2.
Data are expressed as the number (%) of patients or mean value ⫾ SD.
* Comparisons between the group of patients in NYHA functional class I–II and the group of patients in NYHA functional class
III–IV were made using the impaired Student’s t test or the ␹2 test when appropriate.
†
35 Patients were tested.
‡
Acute vasodilator response was defined by a decrease both in PVRI and Ppa ⭓ 20%.
§
These two patients were NYHA functional class II at the moment of PAH diagnosis, but they became NYHA functional class
IV during follow-up and were subsequently treated with epoprostenol.
patients. The probability of survival of the overall population at
1, 2, and 3 years was 73, 60, and 47%, respectively. Nevertheless it
appeared that survival was dramatically worse in the patients in
NYHA functional class III–IV in comparison with patients
in NYHA functional class I–II with respective rates of 60, 45,
and 28% versus 100, 90,, 84% at 1, 2, and 3 years (p ⬍ 0.0001)
(Figure 1).
Factors Associated with Mortality in Patients in NYHA
Functional Class III–IV
Fifty-eight subjects were NYHA functional class III–IV at the
time of PAH diagnosis and two more fell into NYHA functional
class III–IV during follow-up. Among these 60 patients, 20 received epoprostenol therapy. The 40 remaining patients did not
receive epoprostenol because this treatment was not available
TABLE 2. PATIENTS TREATED WITH EPOPROSTENOL: RESULTS AT 3 MONTHS AND LAST
SCHEDULED VISIT
Six-minute walk test, m
Hemodynamic measures
Ppa, mm Hg
CI, L minute⫺1 m⫺2
PVRI, IU m⫺2
SvO2, %
Baseline
(n ⫽ 20)
Three Months
(n ⫽ 20)
Baseline
(n ⫽ 12)
Three Months
(n ⫽ 12)
Last Scheduled Visit
(n ⫽ 12)
241 ⫾ 177
424 ⫾ 96*
213 ⫾ 195
434 ⫾ 86*
456 ⫾ 109†
⫾
⫾
⫾
⫾
46 ⫾ 10*
2.5 ⫾ 0.6*
16 ⫾ 6*
65 ⫾ 8*
45
3.1
14
66
57
2.0
25
55
⫾
⫾
⫾
⫾
11
0.4
6
9
48
2.7
16
65
⫾
⫾
⫾
⫾
9*
0.6*
6*
8*
55
1.9
25
54
9
0.4
6
10
⫾
⫾
⫾
⫾
9†
1.0†
6†
7†
Definition of abbreviations: CI ⫽ cardiac index; Ppa ⫽ mean pulmonary arterial pressure; PVRI ⫽ pulmonary vascular resistance
index; SvO2 ⫽ mixed venous oxygen saturation.
Data are expressed as mean value ⫾ SD.
The analysis of variance test was used for statistical analysis.
* p ⬍ 0.05 when comparing values at three months with those at baseline.
†
p ⬍ 0.05 when comparing values at last scheduled visit with those at baseline.
1436
AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 167 2003
Figure 1. Survival of the overall
population according to New
York Heart Association (NYHA)
functional class. Survival was calculated from the pulmonary arterial hypertension (PAH) diagnosis
until the end of the follow-up period. The probability of survival of
each group was estimated by the
Kaplan–Meier method and compared by the log-rank test.
before 1995 in this patient population and/or because they declined
it and/or because they had unstable psychosocial profiles incompatible with long-term infusion. However, they all benefited from
conventional PAH therapy and had appropriate specialized medical care for HIV infection and PAH.
Univariate analysis. The results of univariate analysis of the
relationship between survival and variables measured at baseline in patients in NYHA functional class III–IV are shown in
Table 3. Mortality was not associated with patient age, sex,
intravenous drug abuse, viral load, 6-minute walk test, hemodynamic parameters, treatment with warfarine, or calcium-channel
blockers. By contrast, a CD4 lymphocyte count of more than
212 cells mm⫺3, the use of CART, and treatment with epoprostenol were significantly related with a decreased risk of death.
Multivariate analysis. We selected the variables evidenced by
univariate analysis (CD4 lymphocyte count, CART, and epoprostenol therapy) and examined the independent effect of each
independent variable in the presence of others on mortality.
Only the CD4 lymphocyte count of more than 212 cells mm⫺3
remained independently associated with a better survival. Therefore neither CART nor epoprostenol therapy were independent
prognostic factors in HIV-PAH, presumably because they were
linked variables. In functional class III and IV patients, survival
was worse in the absence of epoprostenol therapy even in the
subgroup of patients receiving CART (Figures 2 and 3). It should
be emphasized, however, that the response of patients to antiretroviral therapy was not systematically evaluated and leaves open
the possibility that differences in the response to treatment with
antiretroviral drugs could have contributed to the differences
seen when comparing patient survival.
DISCUSSION
The present study provides the first comprehensive longitudinal
analysis of a large population of patients displaying HIV-associated PAH. It clearly shows that this complication is relevant both
in terms of impairment of quality of life and survival. It also
suggests that epoprostenol therapy and CART have significant
effects in HIV-associated PAH patients in NYHA functional class
III and IV.
The prognosis of HIV-associated PAH is mainly related to
pulmonary hypertension itself rather than to other complications of
HIV infection. Over the last 20 years, PAH prognosis has markedly
improved due to the development of novel treatments including
TABLE 3. UNIVARIATE ANALYSIS RELATING SURVIVAL TIME
TO SELECTED BASELINE VARIABLES IN THE PATIENTS IN
NEW YORK HEART ASSOCIATION FUNCTIONAL CLASS III–IV
Variables
Age ⬍ 34
Sex, women/men
IV drug use, yes/no
CD4 count ⬎ 212 cells mm⫺3
Viral load ⬎ 3.1 log
Six-minute walk test ⬍ 327 m
Baseline hemodynamic measures
mRAP ⬎ 10 mm Hg
Ppa ⬎ 50 mm Hg
CI ⬍ 2.3 L minute⫺1 m⫺2
PVRI ⬎ 19 IU m⫺2
SvO2 ⬍ 59%
Warfarine, yes/no
CART, yes/no
Epoprostenol infusion, yes/no
Hazard Ratio
(95% Confidence Interval )
p Value
1.219
1.296
0.678
0.339
0.811
0.993
(0.619–2.399)
(0.654–2.568)
(0.345–1.332)
(0.154–0.743)
(0.202–3.268)
(0.341–2.896)
0.57
0.46
0.26
0.007
0.77
0.99
0.824
0.545
0.902
0.517
1.192
0.662
0.470
0.365
(0.418–1.623)
(0.272–1.089)
(0.453–1.798)
(0.256–1.043)
(0.571–2.487)
(0.305–1.435)
(0.223–0.989)
(0.158–0.844)
0.58
0.09
0.77
0.07
0.64
0.30
0.047
0.019
Definition of abbreviations: CART ⫽ combination antiretroviral therapy refers to
the combination of at least three antiretroviral agents including a protease inhibitor; CI ⫽ cardiac index; IV ⫽ intravenous; mRAP ⫽ mean atrial pressure; Ppa ⫽
mean pulmonary arterial pressure; PVRI ⫽ pulmonary vascular resistance index;
SvO2 ⫽ mixed venous oxygen saturation.
Univariate analysis used the proportional hazards model.
Nunes, Humbert, Sitbon, et al.: HIV-associated Pulmonary Hypertension
high-dose calcium-channel blockers, long-term continuous intravenous epoprostenol, and lung transplantation (1). Unfortunately,
therapeutic options are more limited in HIV-associated PAH. Lung
transplantation is not advisable in this population. When possible,
oral anticoagulation should be recommended, but HIV-infected
patients often have hemostasis abnormalities that increase the risk
of bleeding (HIV-induced thrombocytopenia, chronic liver disease), and medical supervision may be difficult in drug addicts.
Benefit of calcium-channel blockers is proved only in the subset
of patients with a positive response to acute vasodilator challenge
(20, 24), which is very uncommon in patients with HIV-associated
PAH. Long-term intravenous epoprostenol therapy has been recently showed to improve functional status and hemodynamics in
a small series of patients displaying HIV-associated PAH (18).
However, only six patients were analyzed, and there was no control
group so that a definite effect on survival could not be ascertained.
The probability of survival of the overall population at 1, 2,
and 3 years was 73, 60, and 47%, respectively. Nevertheless it
appeared that survival was dramatically worse in patients in
NYHA functional class III–IV in comparison with patients in
NYHA functional class I–II, emphasizing the fact that NYHA
functional class is a major indicator of prognosis in HIV-PAH,
as demonstrated previously in primary pulmonary hypertension.
We therefore specifically analyzed predictors of survival in the
subgroup of patients in NYHA functional class III–IV. Univariate
analysis pointed that the CD4 lymphocytes count of more than
212 cells mm⫺3, the use of CART, and treatment with epoprostenol were significantly related with a decreased risk of death
in these patients. Multivariate analysis showed that only CD4
1437
lymphocyte count was an independent predictor of survival, presumably because CART and epoprostenol therapy were strongly
linked. Indeed, these two therapies were available in this indication in France approximately at the same time and 80% of patients
receiving epoprostenol also received CART, whereas only 30%
of patients on conventional therapy had concomitant CART
(p ⬍ 0.001) (Table 4). Although the respective benefits of CART
and epoprostenol therapy could not be determined in this cohort,
survival was worse in NYHA functional class III–IV patients
receiving CART in the absence of epoprostenol therapy as compared with patients receiving CART and epoprostenol (Figures
2 and 3) despite the fact that patients receiving epoprostenol had
a more severe hemodynamic impairment (Table 4). In addition,
epoprostenol infusion significantly improved surrogate markers
of PAH mortality (NYHA functional class, 6-minute walk test,
and hemodynamic parameters), whereas it remained unchanged
in patients receiving CART alone (data not shown).
To date, there is no trial evaluating properly the effect of
antiretroviral therapy in HIV-associated PAH. In a prospective
study, Opravil and colleagues (6) followed 13 patients with HIVassociated PAH by Doppler echocardiography for a mean period
of 1.3 years. They observed that four of five patients with increasing estimated pulmonary pressures did not receive any antiretroviral treatment whereas five of eight patients with stable
or decreasing estimated pulmonary pressures were treated with
antiretroviral agents (zidovudine or didanosine). However, the
number of patients studied was small, cardiac hemodynamics
were estimated by Doppler echocardiography and not measured
by catheterization, and antiretroviral monotherapies are not
Figure 2. Survival of the patients in NYHA functional classes III–IV. Fifty eight subjects were NYHA functional class III–IV at the time of PAH diagnosis
and two more fell into NYHA functional class III–IV during follow-up. Among these 60 patients, 20 received epoprostenol therapy. The start point
was the time when patients first became NYHA functional class III–IV. The probability of survival of each group was estimated by the Kaplan–Meier
method and compared by the log-rank test. Epoprostenol-treated patients had better survival than patients receiving conventional therapy.
1438
AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 167 2003
Figure 3. Survival of the NYHA functional class III–IV patients treated with combination antiretroviral therapy (CART). Twenty-eight patients in NYHA
functional III–IV received CART or epoprostenol plus CART. The probability of survival of each group was estimated by the Kaplan–Meier method and
compared by the log-rank test. Patients treated with epoprostenol plus CART (n ⫽ 16) had a better survival than those receiving conventional therapy
plus CART (n ⫽ 12).
usually very active on the course of HIV disease (13–15). To
study the impact of antiretroviral therapy on cardiac involvement,
Pugliese and colleagues (25) retrospectively analyzed a population
of 1,042 HIV-infected patients admitted to a division of infectious
diseases between 1989 and 1998. Surprisingly, they found that
PAH was significantly more frequent in patients treated with
highly active antiretroviral therapy than in those treated only
with nucleoside reverse transcriptase analogs. However, this difference may be due to a better knowledge of PAH as a complication of HIV infection in the most recent period when highly active
antiretroviral therapy was available.
The limitations of our study are as follows. (1) This is a retrospective analysis with possible biases; however, it would be impossible for ethical reasons to perform a prospective study designed
to compare novel therapies such as CART and epoprostenol with
less active treatments in a cohort of patients displaying HIVassociated PAH. (2) This is a monocenter study; however, this
center for pulmonary vascular diseases recruits more than 50%
of the HIV-associated PAH diagnosed in France and has the
unique possibility to monitor a large population of patients with
validated tools including the 6-minute walk test and right-heart
catheterization. (3) Our population includes some patients with
such a difficult psychosocial profile, particularly intravenous drug
abusers, that they did not receive epoprostenol therapy. Of course,
these patients are also less likely to use antiretroviral agents and
conventional therapy correctly, which may be responsible for a
selection bias in the analysis of survival. (4) The virologic effect
of CART was not evaluated systematically in our cohort so that
we cannot be sure that it was actually highly active. Nevertheless,
CART with at least three drugs including a protease inhibitor
usually allows good control of the viral infection (13–15).
The pathophysiology of HIV-PAH remains unclear. Because
this complication affects only a small minority of HIV-infected
patients, one could hypothesize that genetic factors play an important role in its development. Germline BMPR2 mutations
were recently identified as the cause of the disease in cases of
familial and sporadic primary pulmonary hypertension as well
as PAH occurring in patients exposed to fenfluramine derivatives
(10–12, 26). The absence of germline BMPR2 mutations observed in the subset of our patients tested displaying HIV-associated PAH suggests that other susceptibility factors are necessary
for developing PAH in the context of HIV infection (9). However, we cannot exclude the possibility of other types of mutations that would not be picked up by our sequencing method
such as promoter or intronic mutations.
We conclude that PAH is a complication of HIV infection that
independently burdens the prognosis of HIV-infected patients.
This report suggests that epoprostenol should be recommended,
in association with CART, in HIV-positive patients with severe
PAH in NYHA functional class III or IV despite optimal conventional therapy. Novel, less invasive vasodilator therapies including
oral, inhaled, or subcutaneous prostaglandin derivatives, endothelin-receptor antagonist, and phosphodiesterase type 5 inhibitor
should be evaluated in this patient population (27).
Nunes, Humbert, Sitbon, et al.: HIV-associated Pulmonary Hypertension
1439
TABLE 4. BASELINE CHARACTERISTICS OF PATIENTS IN
NEW YORK HEART ASSOCIATION FUNCTIONAL CLASS III–IV
Age, yr
Men/women
HIV risk factors
IV drug use
Homosexual
Heterosexual
Transfusion
Unknown
CD4 count, cells mm⫺3
Viral load, log 10†
Six-minute walk test, m
Hemodynamic measures
mRAP, mm Hg
Ppa, mm Hg
CI, L minute⫺1 m⫺2
PVRI, IU m⫺2
SvO2, %
Warfarine
Calcium-channel
blockers
CART
Conventional
Therapy
(n ⫽ 40)
p Value*
Epoprostenol
Infusion
(n ⫽ 20)
34 ⫾ 5
20/20
NS
NS
37 ⫾ 6
12/8
25 (62)
6 (15)
5 (13)
2 (5)
2 (5)
231 ⫾ 203
3.9 ⫾ 1.7
329 ⫾ 164
9⫾4
50 ⫾ 11
2.7 ⫾ 0.6
17 ⫾ 7
57 ⫾ 11
29 (72)
4 (10)
12 (30)
NS
⬍ 0.03
⬍ 0.02
NS
⬍
⬍
⬍
⬍
0.04
0.03
0.0001
0.0001
NS
NS
NS
⬍ 0.001
8
5
2
5
(40)
(25)
(10)
(25)
0
374 ⫾ 257
2.2 ⫾ 1.8
241 ⫾ 177
12 ⫾ 5
57 ⫾ 11
2.0 ⫾ 0.4
25 ⫾ 6
55 ⫾ 9
19 (95)
11.
12.
13.
14.
15.
1 (5)
16 (80)
Definition of abbreviations: CART ⫽ combination antiretroviral therapy refers to
the combination of at least three antiretroviral agents including a protease inhibitor; CI ⫽ cardiac index; HIV ⫽ human immunodeficiency virus; IV ⫽ intravenous;
mRAP ⫽ mean atrial pressure; NS ⫽ not significant; Ppa ⫽ mean pulmonary
arterial pressure; PVRI ⫽ pulmonary vascular resistance index; SvO2 ⫽ mixed venous
oxygen saturation.
Data are expressed as the number (%) of patients or mean value ⫾ SD.
* Comparisons between the group of patients treated with conventional therapy
and the group of patients treated with epoprostenol infusion were made using
the impaired Student’s t test or the ␹2 test when appropriate.
†
25 Patients were tested.
16.
17.
18.
19.
References
1. Rubin LJ. Primary pulmonary hypertension. N Engl J Med 1997;336:111–
117.
2. Archer S, Rich S. Primary pulmonary hypertension: a vascular biology
and translational research “Work in progress”. Circulation 2000;102:
2781–2791.
3. Kim KK, Factor SM. Membranoproliferative glomerulonephritis and
plexogenic pulmonary arteriopathy in a homosexual man with acquired immunodeficiency syndrome. Hum Pathol 1987;18:1293–1296.
4. Speich R, Jenni R, Opravil M, Pfab M, Russi EW. Primary pulmonary
hypertension in HIV infection. Chest 1991;100:1268–1271.
5. Petitpretz P, Brenot F, Azarian R, Parent F, Rain B, Hervé P, Simonneau
G. Pulmonary hypertension in patients with human immunodeficiency
virus infection: comparison with primary pulmonary hypertension. Circulation 1994;89:2722–2727.
6. Opravil M, Pechère M, Speich R, Joller-Jemelka HI, Jenni R, Russi EW,
Hirschel B, Lüthy R. HIV-associated primary pulmonary hypertension: a case control study: Swiss HIV cohort study. Am J Respir Crit
Care Med 1997;155:990–995.
7. Mesa RA, Edell ES, Dunn WF, Edwards WD. Human immunodeficiency
virus infection and pulmonary hypertension: two new cases and a
review of 86 reported cases. Mayo Clin Proc 1998;73:37–45.
8. Mehta NJ, Khan IA, Mehta RN, Sepkowitz DA. HIV-related pulmonary
hypertension: analytic review of 131 cases. Chest 2000;118:1133–1141.
9. Humbert M, Nunes H, Sitbon O, Parent F, Hervé P, Simonneau G. Risk
factors for pulmonary arterial hypertension. Clin Chest Med 2001;22:
459–475.
10. Deng Z, Morse JH, Slager SL, Cuervo N, Moore KJ, Venetos G,
20.
21.
22.
23.
24.
25.
26.
27.
Kalachikov S, Cayanis E, Flischer SG, Barst RJ, et al. Familial primary
pulmonary hypertension (gene PPH1) is caused by mutations in the
bone morphogenetic protein receptor-II gene. Am J Hum Genet 2000;
67:737–744.
Lane KB, Machado RD, Pauciulo MW, Thomson JR, Philipps JA III,
Loyd JE, Nichols WC, Trembath RC. Heterozygous germline mutations in a TGF-␤ receptor, BMPR2, are the cause of familial primary
pulmonary hypertension: The International PPH consortium. Nat Genet
2000;26:81–84.
Thomson JR, Machado RD, Pauciulo MW, Morgan NV, Humbert M,
Elliott GC, Ward K, Yacoub M, Mikhail G, Rogers P, et al. Sporadic
primary pulmonary hypertension id associated with germline mutations of the gene encoding BMPR-II, a receptor member of the TGF-␤
family. J Med Genet 2000;37:741–745.
Carpenter CCJ, Fischl MA, Hammer SM, Hirsch MS, Jacobsen DM,
Kazenstein DA, Montaner JSG, Richman DD, Saag MS, Schooley RT,
et al. Antiretroviral therapy for HIV infection in 1998: updated recommendations of the international AIDS society-USA Panel. JAMA 1998;
280:78–86.
Palella FJ, Delaney KM, Moorman AC, Loveless MO, Furher J, Satten
GA, Aschman DJ, Holmberg SD. Declining morbidity and mortality
among patients with advanced human immunodeficiency virus infection: HIV outpatient study investigators. N Engl J Med 1998;338:853–
860.
Mocroft A, Vella S, Benfield TL, Chiesi A, Miller V, Gargalianos P,
d’Arminio Monforte A, Yust I, Bruun JN, Phillips AN, et al. Changing
patterns of mortality across Europe in patients infected with HIV-1.
Lancet 1998;352:1725–1730.
Barst RJ, Rubin LJ, Long WA, McGoon MD, Rich S, Badesch DB,
Groves BM, Tapson VF, Bourge RC, Brundage BH, et al. A comparison of continuous intravenous epoprostenol (prostacyclin) with conventional therapy for primary pulmonary hypertension. N Engl J Med
1996;334:296–301.
McLaughlin VV, Genthner DE, Panella RN, Rich S. Reduction in pulmonary vascular resistance with long-term epoprostenol (prostacyclin)
therapy in primary pulmonary hypertension. N Engl J Med 1998;
338:273–277.
Aguilar RV, Farber HW. Epoprostenol (prostacyclin) therapy in HIVassociated pulmonary hypertension. Am J Respir Crit Care Med 2000;
162:1846–1850.
Rich S, Dantzker DR, Ayres SM, Bergofsky EH, Brundage BH, Detre
M, Fishman AP, Goldring RM, Groves BM, Koerner SK, et al. Primary
pulmonary hypertension: a national prospective study. Ann Intern Med
1987;107:216–223.
Sitbon O, Humbert M, Jagot J-L, Taravella O, Fartoukh M, Parent F,
Hervé P, Simonneau G. Inhaled nitric oxide as a screening agent to
safely identify responders to oral calcium-channel blockers in primary
pulmonary hypertension. Eur Respir J 1998;12:265–270.
Underhill PA, Jin L, Lin AA. Detection of numerous Y chromosome
biallelic polymorphisms by denaturing high-performance liquid chromatography. Genome Res 1997;7:996–1005.
O’Donovan MC, Oefner PJ, Roberts SC. Blind analysis of denaturing
high-performance liquid chromatography as a tool for mutation detection. Genomics 1998;52:44–49.
Sitbon O, Humbert M, Nunes H, Parent F, Garcia G, Hervé P, Rainisio
M, Simonneau G. Long-term intravenous epoprostenol infusion in
primary pulmonary hypertension: prognostic factors and survival.
J Am Coll Cardiol 2002;40:780–788.
Rich S, Kaufmann E, Levy PS. The effect of high doses of calciumchannel blockers on survival in primary pulmonary hypertension.
N Engl J Med 1992;327:76–81.
Pugliese A, Isnardi D, Saini A, Scarabelli T, Raddino R, Torre D. Impact
of highly active antiretroviral therapy in HIV-positive patients with
cardiac involvement. J Infect 2000;40:282–284.
Humbert M, Deng Z, Simonneau G, Barst RJ, Sitbon O, Wolf M, Cuervo
N, Moore KJ, Hodge SE, Knowles JA, et al. BMPR2 germline mutations
in pulmonary hypertension associated with fenfluramine derivatives.
Eur Respir J 2002;20:518–523.
Hoeper MM, Galiè N, Simonneau G, Rubin L. New treatments for
pulmonary arterial hypertension. Am J Respir Crit Care Med 2002;165:
1209–1216.