52 / 74
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 29, No 6, November/December 2018
384
AFRICA
surgery were not identified. It was also difficult to compare
results with the literature because of vast differences within the
patient population.
Similar to data in a study by Jobes
et al
.,
21
our documentation
of amount of blood products transfused by both the anaesthetists
and perfusionists were in units and not in millilitres.
In our study, patients were categorised according to the
RACHS category, as it signifies complexity of the congenital
cardiac lesion and surgery.
20
There was an increasing progression
of transfused blood products across the RACHS categories.
Patients in RACHS category 4 had a higher transfusion rate for
RPC, platelets and cryoprecipitate than the other groups. These
patients were younger and had prolonged CPB and AOX times.
Studies by Kipps
et al.
9
and Szekely
et al
.
6
also demonstrated this
association with increased rates of transfusion of blood products.
No FFP was transfused in RACH category 4 patients, with
a preference being shown for cryoprecipitate, which contains
concentrated clotting factors in a small volume.
5
This allowed
for haemostasis to be achieved without the complication of fluid
overload in these patients.
Bloodless paediatric cardiac surgery has been performed
successfully with the use of miniature circuits.
13,14,18,22
At our
institution, miniature CPB circuits were not utilised at the time
of the audit. Perhaps this was the reason for only 1.9% of cardiac
surgical cases having bloodless surgery in our study. Miniature
circuits have also resulted in a decrease in volume of RPC
transfusion without an increase in morbidity or mortality rates.
23
Studies by Redlin
et al.
24
and Miyaji
et al
.
18
showed higher
rates of bloodless cardiac surgery (48%) on patients below 16
kg in weight, and 64% on patients less than 7 kg. A one-year
retrospective study by Durandy
et al
.
25
reported 61% of patients
between 6 and 15 kg having undergone bloodless cardiac surgery.
The blood products mostly transfused were RPC and FFP,
with perfusionists mainly transfusing RPC and anaesthetists
transfusing RPC, FFP, cryoprecipitate and platelets. The reason
for this might be that the perfusionists primed the CPB circuit
with RPC to prevent haemodilution associated with large-
volume CPB circuits.
22
As a result, the haemoglobin levels
were not significantly different across the RACHS categories
throughout the peri-operative period.
There was an overall decrease in haemoglobin from
pre-operative levels on initiation of CPB, and an increase to
>
10 g/dl on arrival in ICU. A similar peri-operative haemoglobin
pattern was seen in the blood transfusion group of the cohort
study by Redlin
et al.
24
This demonstrates the appropriate use of
RPC in the maintenance of haemoglobin levels.
The anaesthetists empirically transfused FFP, cryoprecipitate
and platelets at the end of CPB to manage bleeding. Only
activated clotting time was used as a point-of-care test, which
did not give information about the state of the other components
of the coagulation system. A recent cohort by Machovec
et al
.
26
in infant cardiac surgery showed decreased transfusion exposure
with the use of a haemostasis-management system compared to
using activated clotting time.
It has become common practice to use intra-operative point-
of-care tests to assess the coagulation status of patients before
transfusion of blood products, as well as blood-management
programmes in paediatric cardiac surgery.
23,27
This practice has
resulted in decreased units of blood products transfused when
used in conjunction with algorithms.
26-29
Testing of fibrinogen
and platelet function during rewarming resulted in decreased
cryoprecipitate transfusion post CPB in a cohort by Machovec
et al
.
29
Although there were no blood-usage protocols at our
institution, the mean haemoglobin level on CPB of 8.3 (1.5) g/dl
was attained, which is comparable with haemoglobin trigger
levels of 7–8 g/dl observed in other studies.
24,25
Even though
transfusion triggers are set, transfusion of blood products should
be individualised to the clinical condition of the patient.
4,30
In this study, an expected statistically significant difference
was shown in CPB and AOX times across RACHS categories.
There was a statistically significant correlation between FFP
units transfused with CPB time, and RPC units transfused.
Jenkins
et al.
20
had concluded that a high RACHS category,
signifying the complexity of the surgery, was associated with
long CPB time, which increased the risk for transfusion of
blood products. A study by Redlin
et al
.
24
revealed a significant
association between transfusion amount and CPB time (OR
=
1.02, 95% CI
=
1.01–1.03,
p
≤
0.0001). Another study by Salvin
et al
.
31
showed that patients with increased CPB and AOX times
were transfused more blood products postoperatively.
In the current study, when secondary analysis was conducted
by categorising data according to body weight, there was no
significant difference with median FFP transfusion between the
weight categories. There was a statistically significant difference
in transfusion of RPC, cryoprecipitate and platelets between
the weight groups. Studies have shown that low body weight
is associated with an increased transfusion amount of blood
products.
9,32
There are no protocols on conservation strategies at our
institution, but the majority of patients utilised some form
of blood-conservation strategy, singly or in combination. The
Table 10.
Post hoc
Dunn’s test of use of blood products
between weight categories
Products
RPC
Cryoprecipitate
Platelets
Weight categories
<
6 kg 6–15 kg
<
6 kg 6–15 kg
<
6 kg 6–15 kg
6–15 kg
0.096
0.049*
0.019*
>
15 kg
0.002* 0.002* 0.004* 0.043* 0.099 0.097
*
p
<
0.05; RPC, red packed cells.
Table 8. Peri-operative haemoglobins across the RACHS categories
Haemoglobin (g/dl) RACHS 1 RACHS 2 RACHS 3 RACHS 4
p
-value
Pre-operative
12.6 (1.3) 14.7 (3.6) 14.3 (3.6) 11.1 (2.6)
0.16
Initial CPB
8.5 (1.4)
9.1 (1.8)
9.3 (1.5)
8.9 (2.6)
0.61
Lowest CPB
8.3 (1.4)
8.3 (1.6)
8.5 (1.2)
7.2 (1.2)
0.42
Last CPB
8.9 (1.3)
9.4 (1.3)
9.7 (1.2) 10.3 (1.4)
0.33
Initial ICU
12.5 (1.3) 12.1 (2.1) 11.8 (1.9) 11.2 (1.8)
0.72
CPB, cardiopulmonary bypass; ICU, intensive care unit; RACHS, risk-adjusted
classification for congenital heart surgery.
Table 9. Median units of blood products transfused, by body weight
Products
Weight
<
6 kg
(
n
=
11)
Weight 6
–
15 kg
(
n
=
52)
Weight
>
15 kg
(
n
=
42)
Kruskal
–
Wallis
p
-value
RPC
1 (1–2)
1 (1–1)
1 (0–1)
0.001*
FFP
0 (0–0)
0 (0–1)
1 (0–1)
0.087
Platelets
1 (0–1)
0 (0–0)
0 (0–1)
0.038*
Cryoprecipitate 0 (0–1)
0 (0–0)
0 (0–0)
0.009*
*
p
<
0.05; RPC, red packed cells; FFP, fresh frozen plasma.