Cardiovascular Journal of Africa: Vol 27 No 3 (May/June 2016)

50 / 102

CARDIOVASCULAR JOURNAL OF AFRICA • Volume 27, No 3, May/June 2016 172 AFRICA
for mortality were identified during the mid-1990s. These risk
factors were age (younger than 70 years, and 70 years or older
were arbitrarily selected by the surgeon), gender, re-operation,
left main stem, low ejection fraction (40% was applied as the
cut-off point), urgency, and number of coronary artery systems
involved. Data on the last factor are not available for this series.
From Table 4 it is clear that of these factors, the female gender
was not significantly associated with operative mortality rate. As
expected, patients with chronic obstructive pulmonary disease
(COPD) were at higher risk (223 patients with 5.8% mortality
rate, compared to 2.6% for the other patients) (p = 0.0090).
About one-fifth (21.7%) of the patients had renal impairment
(CKD III) based on the calculated GFR. Their mortality rate
was 6.7% compared to 2.1% for the rest (p < 0.0001). Other
potential risk factors that were not significantly associated with
operative mortality included hypertension, diabetes mellitus and
body mass index (BMI ≥ 30 kg/m2), not even in combination, as
in the metabolic syndrome.
A perception that the patients of some referring cardiologists
are at higher risk was tested (Table 5). Six cardiologists are or
were involved at the Mediclinic, Bloemfontein, at some point.
One of the six ‘cardiologists’ actually represents a number of
cardiologists, each with only a small number of cases. Cardiologist
B had a higher risk score and therefore also a higher mortality rate
than A, D and E. Cardiologists B and C (the latter had only 58
cases) did not differ significantly regarding mortality rate.
Stepwise logistic regression
All the significant risk factors were used to establish a model
of factors that are significantly associated with mortality in
a multivariate logistic regression. Four risk factors could be
considered independent risk factors: urgency (intra-aortic balloon
pump; IABP), renal impairment (CKD III), re-exploration, and
an additional procedure (Table 6).
Morbidity
Besides the 53 deaths, 115 patients had a major complication
during their initial stay in hospital; however, 31 of these patients
also had a major complication after the operation and then died
(22 patients died without an official major complication). A
further seven patients were included as morbidity after they had
been discharged, but were re-admitted with sternal dehiscence. The
combined mortality and major morbidities involved 144 patients
(8.2%). Table 7 illustrates the prevalence of major morbidities.
Some patients (23 + 5) had more than one major complication
and their mortality rate was 50%. The low mortality rate (1.0%)
associated with no major complication is obvious (Table 8).
Gastrointestinal complications are not often regarded as
major complications, yet some are very serious. Twenty-one such
patients (1.2%) were identified, of whom11 had a gastroscopically
diagnosed peptic ulcer; four presented with active bleeding and
six had laparotomies. This finding was despite the routine use of
a proton pump inhibitor. The laparotomies were done for bowel
ischaemia (three), bowel obstruction (two) and ulcer perforation
(one). There were four (19.0%) deaths among these 21 patients,
two with bleeding and two with ischaemia.
Table 3. Reason for death
Reason for death Number of deaths (%)
Cardiac (death in theatre) 5 (9.4)
Surgical bleed 1 (1.9)
Sudden and unexplainable 5 (9.4)
Stroke 2 (3.8)
Brain dead 1 (1.9)
Inflammatory conditions (DIC, IE, sepsis, SIRS) 7 (13.2)
Gastrointestinal 3 (5.7)
Organ failure (cardiac, respiratory, renal, MOF) 28 (52.8)
Pining away 1 (1.9)
DIC, diffuse intravascular coagulopathy; IE, infective endocarditis; MOF, multi-
organ failure; SIRS, systemic inflammatory response syndrome.
Table 4. Core risk factors25
Risk factor Mortality/total Mortality (%) p-value
< 70 years 31/1 352 2.3
≥ 70 years 22/398 5.5 0.0009
Male 46/1 344 3.4
Female 7/406 1.7 0.0801
1st sternotomy 35/1 554 2.3
2nd sternotomy 14/179 7.8
3rd sternotomy 4/15 26.7
4th sternotomy 0/2 0.0 < 0.0001
No left main stem 34/1 420 2.4
Left main stem 19/330 5.8 0.0013
LVEF ≥ 40% 41/1 652 2.5
LVEF < 40% 12/98 12.2 < 0.0001
Home/ward 14/570 2.5
CCU 21/868 2.4
IABP 12/299 4.0
Ventilator/lab 6/13 46.2 < 0.0001
CCU, coronary care unit; IABP, intra-aortic balloon pump; lab, catheter labora-
tory; LVEF, left ventricular ejection fraction.
Table 5. Cardiologist’s mortality/risk and contribution to the cohort
Cardiologist
Mortality/
total
Mortality
(%) EuroSCORE
Contribution to
the total (%)
A 12/484 2.5 3.57 27.6
B 19/335 5.7 4.76 19.1
C 0/58 0.0 4.36 3.1
D 5/309 1.6 3.69 17.7
E 11/431 2.6 3.55 24.6
F 6/133 4.5 4.17 7.6
Table 6. Significant risk factors in logistic regression with odds ratio
Risk factor OR 95% CI
Urgency (IABP) 2.21 1.13–4.32
Renal impairment (CKD III) 2.58 1.44–4.65
Re-operation 4.31 2.32–8.00
Additional procedure 7.14 3.60–14.18
CI, confidence interval; CKD, chronic kidney disease; OR, odds ratio; IABP,
intra-aortic balloon pump.
Table 7. Major morbidities
Complication Number of patients (%)
Re-exploration 32 (1.8)*
Prolonged ventilation 36 (2.1)
Renal failure 32 (1.8)
Permanent stroke 20 (1.1)
Sternal dehiscence 17 (1.0)
The percentage was calculated from 1 745 patients as five died in theatre.
*31 for bleeding and one instrument that was left behind.
28 patients had more than one major complication (Table 8).