CARDIOVASCULAR JOURNAL OF AFRICA • Vol 24, No 6, July 2013
220
AFRICA
appropriate. Statistical analysis of quantitative image analysis
data was performed on pooled individual data points collected
for each patient within the evaluated groups. All tests were
two-sided and
p
<
0.05 was considered significant. Statistical
analysis was performed using Statistica (version 7) and EpiInfo
software.
Results
Control endomyocardial biopsy samples were obtained from
HIV-negative individuals who were undergoing routine
endomyocardial biopsy in the Cardiac Catheterisation Laboratory
at Groote Schuur Hospital, either for the investigation of
idiopathic dilated cardiomyopathy or for monitoring of rejection
in heart transplant recipients. The heart transplant recipients
served as positive controls for immunosuppression (which
is also present in HIV-associated cardiomyopathy) and their
likelihood of having a high frequency of myocarditis resulting
from varying degrees of rejection, whereas the idiopathic dilated
cardiomyopathy cohort acted as negative controls for absence
of immunosuppression and a likely low prevalence of active
myocarditis.
During the study period, a total of 40 participants (15 with
HIV-associated cardiomyopathy, 10with idiopathic dilated cardio-
myopathy and 15 heart transplant recipients) met the inclusion
criteria and were subjected to the endomyocardial biopsy
study. Of these, seven were excluded due to inadequate
amounts of endomyocardial biopsy tissue for the histological,
immunohistochemical and virological analysis. The clinical
characteristics of participants who were analysed are presented
in Table 1. There was no significant difference in age and
gender between the cases with HIV-associated cardiomyopathy,
idiopathic dilated cardiomyopathy and heart transplant recipients.
Fifteen patients with HIV-associated cardiomyopathy
underwent endomyocardial biopsy, but one patient with
inadequate biopsy material was excluded from the analysis,
resulting in 14 patients being entered into the analysis. All cases
were of black African descent (mean age, 40.5
±
10.7). A low
CD
4
+
T-cell count (mean cell count 246
±
193 cells/mm
3
; normal
range, 600–1000 cells/mm
3
) and left ventricular ejection fraction
(mean range 27.9
±
6.92%) were found in the patients with
HIV-associated cardiomyopathy, as expected.
6
Ten consecutive patients with idiopathic dilated
cardiomyopathy were recruited and used for comparison with
findings in the patients with HIV-associated cardiomyopathy.
Two patients were excluded from the analysis due to inadequacy
of endomyocardial biopsies, resulting in eight patients being
entered in the comparative analysis (mean age 43
±
7.35 years;
mean left ventricular ejection fraction 26.2
±
10.39%).
Fifteen consecutive heart transplant recipients who were
undergoing routine endomyocardial biopsy were used as controls
for comparison with findings in the patients with HIV-associated
cardiomyopathy. Due to tissue inadequacy, four patients were
excluded from the analysis, resulting in 11 patients being
entered into the analysis (mean age 47.1
±
14.1 years; mean left
ventricular ejection fraction 52.8
±
7.7%).
Immunohistochemical analysis of the mononuclear infiltrates
in the myocardium demonstrated a lymphocyte population
expressing the pan T-cell marker CD
3
, CD
4
+
T lymphocytes
(helper/inducer), CD
8
+
T lymphocytes (cytotoxic/suppressor),
and monocytes and macrophages (CD
68
+
cells). Using CD
3
and CD
68
immune cell counts, we determined that six of 14
(42.8%) of the HIV-associated cardiomyopathy cohort met the
World Heart Federation criteria for myocarditis, displaying
>
14 leukocytes/mm
2
; three cases of 14 had acute myocarditis
(21.4%), while three of 14 had chronic myocarditis (21.4%) (Fig.
1A). The remaining eight cases did not display any myocarditis.
Two of the eight idiopathic dilated cardiomyopathy cases
(25%) were classified as having chronic myocarditis; the other
six samples were classified as having no myocarditis (Fig. 1B).
None of the idiopathic dilated cardiomyopathy cases met the
criteria for acute myocarditis.
Four heart transplant recipients met the criteria for myocarditis:
two of 11 cases (18%) had acute myocarditis and two of 11
(18%) displayed chronic myocarditis (Fig. 1C). There was no
significant difference in the prevalence of chronic myocarditis
between the groups (
p
=
0.48).
All cases of HIV-associated cardiomyopathy were infected
with at least one cardiotropic virus based on results of
testing for viral genomes by PCR. No virus was detected by
immunohistochemical methods, suggesting low levels of viral
particles in the endomyocardial biopsies. The number of viruses
per case ranged from one to four with a mean viral burden of 2.5
viruses per case irrespective of whether patients were classified
as having myocarditis or no myocarditis. The biopsy tissues from
patients with HIV-associated cardiomyopathy were dominated
by the Epstein-Barr virus (64%) and herpes simplex virus
(50%) with parvovirus B19 (14%) and cytomegalovirus (7%)
having the lowest frequency. HIV-1 was detected in 29% of the
HIV-associated cardiomyopathy cases (Fig. 2).
Similarly, all cases of idiopathic dilated cardiomyopathy were
infected with at least one cardiotropic virus, with a mean viral
burden of 1.1 viruses per case. Enterovirus was the commonest
virus (56%), followed by Epstein-Barr virus (29%). Herpes
TABLE 1. CLINICAL CHARACTERISTIC OF PATIENTSWITH HIV-ASSOCIATED CARDIOMYOPATHY (HIVAC),
IDIOPATHIC DILATED CARDIOMYOPATHY (IDCM) AND HEART TRANSPLANT RECIPIENTS (HTX)
WHOWERE STUDIEDWITH ENDOMYOCARDIAL BIOPSY
HIVAC (
n
=
14)
iDCM (
n
=
8)
HTx (
n
=
11)
p
-value
‡
Median age in years (range)
40 (27–63)
43.5 (31–54)
50 (26–69)
NS*
Male (%)
58
75
82
0.44
Median CD
4
+
T-cell count (cells/mm
3
)
253 (200–344)
not applicable
not applicable
Median HIV load, (copies/ml)
155 000 (28 000-265 000)
not applicable
not applicable
Median LVEF (%)
30.6 (23–35)
27 (13.5–29)
56.5 (46.8–59.0)
<
0.0001
Median duration of illness (weeks)
18 (14.5–26)
14 (7–18)
18.5 (6–21.8)
0.3
‡
p
-value: Fisher’s exact test for categorical variables, Kruskal-Wallis test for continuous variables. LVEF, left ventricular ejection fraction, NS, not significant.
