CARDIOVASCULAR JOURNAL OF AFRICA • Vol 24, No 7, August 2013
252
AFRICA
surgery and/or revision surgery due to other reasons, valve
protection, aortic root expansion, presence of valve calcification
and infective endocarditis, pre-operative and postoperative
rhythm pattern, brand name of prosthesis, distance of the
patient’s house from a cardiac surgery centre, and concomitant
non-cardiac systemic diseases. The factors potentially leading
to re-operation included valvular thrombus formation, pannus
formation and perivalvular leak.
The study protocol was approved by the local ethics committee.
Statistical analysis
This was performed using Windows SPSS v13.0 software.
Data were expressed as percentage. A
p
-value of
<
0.05 was
considered significant. Categorical data were expressed as
number and percentage, while numerical data were expressed as
mean
±
standard deviation. Pearson’s chi-square test and Fisher’s
exact test were used for non-parametric variables, while the
Mann-Whitney
U
-test was performed for parametric variables.
A linear regression analysis was also performed for significant
parameters. A ROC curve was drawn to determine the sensitivity
and specificity of these parameters.
Results
The mean age of patients was 49.3 years (range 19–78 years);
51% (
n
=
22) were males and 49% (
n
=
21) were females.
Twenty-nine patients underwent mitral valve replacement, while
12 underwent aortic valve replacement. A tricuspid valve was
implanted in two patients.
There was no statistically significant difference in baseline
demographic characteristics of the patients. Age and gender
were not a determining factor for re-operation. Demographic
characteristics of the patients and distribution of indications for
re-operation are summarised in Tables 1 and 2, respectively.
The incidence of thrombus formation in mechanical prosthetic
valves was statistically significantly higher in patients with valve
calcification (
p
<
0.05) and left atrial thrombus (
p
=
0.007)
during the primary surgery. Pearson’s chi-square test revealed
that the incidence of perivalvular leak was higher in patients with
left atrial thrombus during the primary surgery (
p
<
0.05). The
incidence of perivalvular leak was statistically higher in patients
with valvular thrombus and pannus formation (
p
<
0.05). In
addition, the incidence of perivalvular leak was higher in patients
with infective endocarditis compared with those without the
disease (
p
<
0.05).
The re-operation mortality rate was 11.6%. A total of 67.4%
(
n
=
29) of patients had mitral valve disease, while 27.9% (
n
=
12) had aortic valve disease. A mitral valve was implanted
anatomically and extra-anatomically in 22 and seven patients,
respectively. It was observed that valve type and implantation
position were not risk factors for re-operation. A St Jude (St Jude
Medical Inc, Minnesota, USA) prosthetic valve was implanted
in 81.4% of patients, a Carbomedics valve (SuzerCarbomedics
Inc, Austin, Texas, USA) was implanted in 7% of patients,
and a Medtronic valve (Medtronic Inc, Minnesota, USA) was
implanted in 7% of patients. The brand of valve prosthesis was
not a risk factor for re-operation (Fig. 1).
The mean platelet volume was higher and statistically
significant in patients with valvular thrombus during re-operation
(
p
<
0.001). A linear regression analysis was performed of
parameters that were statistically significantly related to
valvular thrombus, including left atrial thrombus, MPV, valve
calcification, and perivalvular leak. It was observed that there
was a statistically significant impact of these four parameters
on valvular thrombus formation (
R
=
0.60). However, MPV was
an independent risk factor (
p
<
0.001). A ROC curve showed
a higher percentage of sensitivity (85%) and specificity (87%)
(Fig. 2).
Discussion
Although surgical modalities and myocardial protection
techniques have been improved recently, the mortality rate of
heart valve re-operation varies between 10 and 20%.
7
This leads
to increased cost of care and work load of surgical centres. Delay
in re-operation also results in increased morbidity and mortality,
particularly in developing countries. Such undesired outcomes
may be prevented by defining the factors that lead to re-operation
and designing a preventive healthcare policy.
Overall complications observed with prosthetic heart
valves are divided into six main categories: structural valvular
deterioration, non-structural dysfunction, valve thrombosis,
embolism, bleeding and endocarditis. While leaflet calcification
and leaflet tearing are more commonly encountered with
TABLE 1. DEMOGRAPHIC CHARACTERISTICS
OF THE PATIENTS
Variables
Number (n)
Percentage (%)
Gender
Male
22
51
Female
21
49
Valve replacement
Mitral
29
67.4
Aortic
12
27.9
Tricuspid
2
4.6
Implantation position
Anatomical
33
76.7
Extra-anatomical
10
23.2
Additional cardiac intervention
10
23.2
Concomitant non-cardiac disease
3
6.9
Concomitant cardiac disease
3
6.9
Infective endocarditis
13
30.2
Perivalvular leak
23
53.4
Valve calcification
17
39.5
Pannus formation
19
44.1
TABLE 2. DISTRIBUTION OF INDICATIONS FOR
RE-OPERATIONAMONG PATIENTS
Indications*
Number of patients
Perivalvular leakage
23
Thrombus formation
21
Pannus formation
19
Valvular calcification
17
Infective endocarditis
13
Additional cardiac intervention
10
Left atrial thrombus
6
Accompanying cardiac disease
3
Left atrial aneurism
2
*There were patients with more than one indication for re-operation.
