CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 3, April 2012
144
AFRICA
performed because the patients had only mild or moderate
pericardial effusions and there was no sign of compression of
the heart.
CEA, CA 15-3 and CA 19-9 levels were measured in blood
samples using an electrochemiluminescence immunoassay on
a Roche Modular E170. CA 125 levels were measured with a
quantitative immunoassay technique. All results were evaluated
by a biochemist who did not know the patients.
Echocardiographic examinations were performed by an
experienced cardiologist who was blinded to the tests. They
were performed using standard protocol and a standard device
(Vivid 7, GE Medical Systems, Horten, Norway). Measurement
of the cardiac walls was done according to the American Society
of Echocardiography guidelines. Since the number of patients
with PE was not large enough to establish quantitative amounts
of fluid, these were arbitrarily established by echocardiography
using the sum of the epicardial and pericardial separations in
the anterior and posterior spaces.
19
The effusions were graded as
mild (less than 10 mm), moderate (10–20 mm) and severe (more
than 20 mm).
19
Statistical analysis
SPSS Inc for Windows, standard version 11.0 was used for
statistical analysis. All data are given as mean and standard
deviation. Comparison between patients with malignant and
other aetiologies was done using the Student’s
t
-test, and
for unequally distributed variables, the Mann-Whitney
U
-test.
Logistic regression analysis adjusting for CA 125, CA 15-3,
CA 19-9, alpha-fetoprotein (AFP), prostate membrane antigen
(PSA) and CEA was done to evaluate the relationship between
malignancies and tumour markers. Threshold values of tumour
markers were established by ROC analysis.
Results
A total of 69 patients (32 women and 37 men with an average
age of 58
±
17 years) were included in the study. Aetiological
diagnosis was done on clinical evaluation, imaging techniques
and biochemical markers. The patients were grouped into
categories depending on the aetiology of the PE (group 1: viral/
idiopathic; group 2: bacterial; group 3: tuberculosis; group 4:
malignancy; group 5: other).
Twenty-one patients were found to have malignancies and
48 had benign aetiologies after the first examination. During
the follow-up period, cancer (three lymphoma, one thymoma,
one lung cancer, one gastrointestinal malignancy) developed
in six patients who had been considered idiopathic at the first
examination. Finally, we had 42 patients with benign and 27 with
malignant aetiologies (11 lymphoma, six breast cancer, five lung
cancer, three thymoma, one ovarian cancer, one gastrointestinal
malignancy). The aetiology in 27 patients could not be identified
and they were considered as the viral/idiopathic PE group.
Aortic dissection (one patient), tuberculosis (four patients),
bacterial microorganism (two patients), autoimmune or
rheumatic disease (three patients), chronic renal disease (four
patients) and coronary bypass graft operation (one patient) were
benign underlying causes of PE in 15 patients (Fig. 1). The
characteristics of the patients with malignancies and those with
other aetiologies of PE are shown in Table 1. The clinical and
echocardiographic characteristics of the patients are shown in
Table 2. Fibrin bands were detected most commonly in patients
with tuberculosis (2: 50%), however, this was statistically
insignificant.
Levels of CA 125, CA 15-3 and CEA in the group with
malignancies were significantly higher than in the group with no
malignancies (Table 3). No significant differences were detected
between the levels of CA 19.9, PSA and AFP. The relationship
between malignant aetiology and tumour markers was evaluated
with multivariate analysis, adjusting for CA 125, CA 15-3 and
27
27
Fig. 1. Diagnostic groups of the patients.
patients
Malignancy
Viral/idiopathic
Tuberculosis
Bacterial
Others
4
2
9
0 5 10 15 20 25 30
TABLE 1. CHARACTERISTICS OF PATIENTSWITH PE
OF MALIGNANTAND BENIGNAETIOLOGIES
Malignant PE Benign PE
p
Patient (M/F)
27 (14/13)
42 (23/19)
NS
Age (years)
57.7
±
17.8
52.1
±
13.9 NS
Glucose (mg/dl)
123
±
50.2
121.7
±
46.3 NS
Urea (mg/dl)
48
±
22.9
39.2
±
22.9 NS
Creatinine (mg/dl)
1.1
±
0.4
1.3
±
0.7
NS
Sedimentation (mm/h)
83.2
±
40.4
67.6
±
37.9 NS
WBC
×
1 000
18.4
±
19.3
20.6
±
21.9 NS
CRP (mg/dl)
14.5
±
11.2
10.5
±
9.1
NS
Duration of
hospitalisation (days)
28.3
±
11.4
17.3
±
9.7
NS
Treatment type
NSAID
14
20
Corticosteroid
NSAI
+
colchicine
3
3
10
19
NSAID; non-steroid anti-inflammatory drug, WBC; white blood cell,
M; male, F; female, NS; not significant.
TABLE 2. ECHOCARDIOGRAPHIC PARAMETERS OF
PATIENTSWITH PE OF MALIGNANTAND BENIGN
AETIOLOGIES
Malignant PE
(
n
=
27)
Non-malignant PE
(
n
=
42)
p
LVEDD
48.1
±
6.1
47.4
±
5.3
NS
LVESD
12.1
±
1.7
11.6
±
2.1
NS
PWD
11.3
±
2.3
11.1
±
1.5
NS
EF
64.5
±
15.6
63.1
±
17.2
NS
LAD
39.6
±
5.9
36.2
±
7.3
NS
E
1.2
±
0.9
1.2
±
0.7
NS
RA
1
±
0.8
1.1
±
0.6
NS
RV
24.7
±
2.9
24.1
±
4.1
NS
LVEDD; left ventricular end-diastolic dimension, LVESD; left ventricu-
lar end-systolic dimension, PWD; posterior wall dimension, EF; ejection
fraction, LAD; left atrial diameter, E; mitral early flow, RA; right atrium,
RV; right ventricle; NS; not significant.
