12 / 70
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 30, No 5, September/October 2019
254
AFRICA
levels between HIV-negative and -positive patients. Of note,
alkaline phosphatase (146.0
±
67.7 vs 201.0
±
108.8 U/l;
p
=
0.005) and gamma glutamyl transferase (172.96
±
104.76 vs
370
±
300 U/l;
p
≤
0.001) levels were significantly elevated in
HIV-positive patients.
Pre-operative mortality rate
Of the initial study cohort of 83 patients with constrictive
pericarditis, 31 (37.3%) patients did not undergo immediate
pericardiectomy. Of these 31 subjects, four died in hospital
shortly after admission (all HIV negative) from a low-cardiac-
output state, and the remaining 27 who were offered surgery
did not return for the operation. Survival status of those lost to
follow up was established telephonically as well as by checking
the national registry of deaths. In this way it was established
that a further 10 had died out of hospital (HIV positive:
n
=
4),
yielding a total pre-operative mortality rate of 16.7% (14/83)
(95% CI: 9.5–26.6%).
Bivariate logistic regression analysis identified seven predictors
of pre-operative mortality (Table 2
).
These were age (OR 1.11;
95% CI: 1.04–1.18;
p
≤
0.001), levels of haemoglobin (OR 0.67;
95% CI: 0.45–0.99;
p
=
0.031), albumin (OR 0.90; 95% CI: 0.82–
0.99;
p
=
0.019) and aspartate aminotransferase (OR 0.91; 95%
CI: 0.85–0.98;
p
=
0.003), and pulmonary artery pressure (OR
1.13; 95% CI: 1.05–1.22;
p
≤
0.001). HIV status had no influence
on the pre-operative mortality rate (
p
=
0.693).
On multivariable analysis, age (OR 1.17; 95% CI: 1.03–1.34;
p
=
0.02), serum albumin level (OR 0.63; 95% CI: 0.43–0.92;
p
=
0.016), gamma glutamyl transferase level (OR 0.97; 95% CI:
0.94–0.1.0;
p
=
0.034) and pulmonary artery pressure (OR 1.49;
95% CI: 1.07–2.08;
p
=
0.018) emerged as independent predictors
of pre-operative mortality rate.
Operative outcome of patients undergoing peri-
cardiectomy
A total of 52 patients (62.7%) underwent pericardiectomy,
which included 32 HIV-negative (61.54%) and 20 HIV-positive
patients (38.5%). Of the 20 HIV-positive patients, 15 (75%) were
on antiretroviral therapy with successful viral load suppression
(
<
1 000 copies/ml). Pericardial biopsy specimens taken at the
time of surgery showed histological evidence of tuberculosis in
the form of granulomas and/or acid-fast bacilli in 12/49 (24.5%)
patients.
Complete pericardiectomy was achieved in 38 patients (73.1%)
and there was no significant difference between HIV-positive and
-negative patients (26%; 81.3 vs 12; 60%;
p
=
0.093). There were
three in-hospital peri-operative deaths, yielding a peri-operative
mortality rate of 5.7% (95% CI: 9.5–26.7%). One patient (HIV
positive) died of intra-operative haemorrhage in theatre and
two (HIV negative), who were both severely symptomatic
pre-operatively (NYHA IV) with impaired ejection fraction,
died in the intensive care unit (ICU) as a result of a low-cardiac-
output state in the ICU. There was no significant difference
in the length of ICU stay between HIV-negative and -positive
patients (4.28
±
2.74 vs 5.11
±
2.84 days;
p
=
0.321).
Postoperative complications occurred in seven patients (9.6%),
three of whom had also suffered intra-operative complications.
These postoperative complications were: sternal wound sepsis
(one), re-intubation for respiratory failure and tachyarrhythmia
(one), thoracotomy for postoperative haemorrhage (one),
postoperative renal impairment (one) and low-output cardiac
failure (three). In total, peri-operative (intra- and post-operative)
complications occurred more frequently in HIV-positive patients
(HIV positive: 9, 45% vs HIV negative: 6, 17.6%;
p
=
0.030). The
higher complication rate in HIV-positive patients could not be
explained by left ventricular function since the left ventricular
function was similarly preserved in both groups (HIV negative
53.33
±
6.7% vs HIV positive 53.93
±
6.79%;
p
=
0.783).
Of the 49 patients who were discharged (three died in
hospital) after undergoing pericardiectomy, 41 (26 HIV positive)
returned for the six-week postoperative follow up at our hospital.
Six patients were followed up at their referral hospital and two
were lost to follow up. Most patients improved their NYHA class
by one or two levels (
p
<
0.001) (Fig. 2). The majority of patients
had improved from NYHA class II to class I (
n
=
21, 50%) and
NYHA class III to class I (
n
=
10, 23.8%). Eight patients showed
no improvement in functional class. There was no significant
difference in symptoms of dyspnoea (
p
=
1.000) or ejection
fraction (
p
=
0.785) between HIV-positive and -negative patients.
Discussion
This study shows a relatively high rate of HIV infection
(32/83, 38.6%) among patients with constrictive pericarditis
compared to the 14.6% reported by Mutyaba
et al
.
2
in a recent
South African study, but less than the 12/19 (63%) reported by
Abubaker and colleagues
17
in a Nigerian study. These data for
developing countries are in contrast to the very low rate reported
by Gopaldas
et al
.
18
in the USA, who found only 10 HIV-positive
Table 2. Bivariate logistic regression model
of associated pre-operative mortality
Characteristics
Alive (
n
=
69)
Pre-operative
death
(
n
=
14)
Odds ratio
(95% CI)
p-
value
Gender
Female
33 (47.8)
7 (50.0)
0.92 (0.29–2.89) 0.882
Male
36 (52.2)
7 (50.0)
HIV positive
0.693
CD4 > 200 cells/mm
3
21 (30.4)
3 (21.4)
0.59 (0.15–2.36)
CD4
<
200 cells/mm
3
7 (10.1)
1 (7.1)
0.59 (0.65–5.32)
NYHA class
69 (100)
14 (100)
1.50 (0.65–3.48) 0.351
Haemoglobin (g/dl)
12.96
±
1.70 11.91
±
1.78 0.67 (0.45–0.99) 0.031
White cell count (10
9
cells/l)
5.17
±
1.45 4.99
±
1.58 0.91 (0.61–1.37) 0.660
Platelets (10
12
cells/l)
257
±
89.01 224.64
±
49.96 0.99 (0.99–1.00) 0.160
Sodium (mmol/l)
137
±
3.33 136
±
3.28 0.91 (0.77–1.07) 0.243
Urea (mmol/l)
6.37
±
2.17 7.6
±
3.96 1.17 (0.96–1.42) 0.131
Creatinine (umol/l)
80.37
±
20.87 88.21
±
17.94 1.02 (0.99–1.04) 0.192
Albumin (g/l)
38.35
±
6.29 33.93
±
5.37 0.90 (0.82–0.99) 0.019
AST (U/l)
41.16
±
13.71 31.36
±
9.97 0.91 (0.85–0.98) 0.003
ALT (U/l)
25.87
±
16.88 22
±
17.52 0.98 (0.94–1.03) 0.403
Alkaline PO
4
(U/l)
175.94
±
93.06 125.29
±
54.11 0.99 (0.98–1.00) 0.061
Gamma GT (U/l)
269.39
±
235.30 149.43
±
119.43 1.00 (0.99–1.00) 0.071
Ejection fraction (%)
51.97
±
6.75 53.29
±
6.06 1.03 (0.94–1.13) 0.491
PA pressure (mmHg)
32.80
±
6.88 43
±
11.19 1.13 (1.05–1.22)
<
0.001
Data presented as mean
±
standard deviation for continuous variables and
n
(%)
for categorical variables. NYHA, New York Heart Association; CI, confidence
interval; AST aspartate aminotransferase; ALT alanine aminotransferase; alka-
line PO
4
, alkaline phosphatase; gamma GT, gamma glutamyl transferase; PA
pressure, pulmonary artery pressure.