For CTO procedures,
the dose metric ranges (Fig. 1) and local dose reference levels (Fig. 2),
(RPAK = 4945 mGy and DAP=307 Gycm2),
compare favourably with values published by Maccia et al (2),
however they are higher than those suggested by Siiskonen et al (3).
Comparisons with published data are complicated by the fact that different centers use different classifications of CTO complexity.
The local method of classification,
currently based on national exam codes,
ideally should facilitate the distinction between different levels of complexity within CTO procedures,
e.g.
antegrade vs retrograde procedures and/or different types of arterial access.
The local dose reference levels of RPAK (2265 mGy) and DAP (117 Gycm2) (Fig. 2) for PCI (and PCI MV) procedures are significantly higher than published values (2),
(3). The wide dose metric ranges for PCI and PCI MV in Fig. 1 suggests that incorrect classification of some PCI procedures may skew the local dose reference levels upwards,
e.g.
the procedure is registered as a PCI but ends up being a PCI MV or PCI CTO and this is not retrospectively corrected in the dose tracking software.
During interventional procedures,
the cardiologist is alerted by the interventional equipment at RPAK values above 3000mGy,
at 1000mGy intervals.
All patients are subsequently followed up in outpatients 6-12 weeks after the procedure.
Maccia et al report no skin injuries with RPAK values just above 5000mGy and consequently raised their trigger level for a follow-up to RPAK≥ 7000mGy. The experience in this center is similar and there are no reported skin injuries to date for the ranges of dose metrics in Fig. 1 . This is an interesting finding given that the three patients in the dataset with the highest dose metrics respectively have an estimated PSD value of 7050mGy,
7600mGy and 7683mGy.
These correspond to RPAK values of 8846mGy,
9942mGy and 11096mGy respectively.
Further investigation of PSD calculation accuracy needs to be carried out for lengthy high dose procedures,
where large source to table distances and/or acquisition runs at large primary angles may contribute up to 50% error in the PSD value (6).
Based on the results in Fig. 3 and Fig. 4, for all three categories of PCI,
the Q2 (median) and Q3 RPAK and DAP values are >40% higher for male patients compared to female patients.
However,
for MV and CTO procedures,
this does not translate into higher PSD where the values are comparable or,
in the case of MV procedures,
lower for males.
The latter may be due to variability caused by the low number [4] of female MV procedures for which the PSD percentiles were calculated.
For PCI (single vessel) procedures all dose metrics are approximately twice as high for males compared to females. This may be due to the weight difference between males (mean =84kg) and females (mean=70kg) for identical BMI (mean=28).
Fig. 5 illustrates how patient age,
height,
weight,
and BMI are correlated with dose metrics derived for males and females in each of the three categories of PCI procedure.
In general,
the dose metrics are positively correlated with patient weight for males and females.
For CTO procedures on females,
all the dose metrics are weakly negatively correlated with patient age (i.e.
younger women tend to receive lower radiation dose for CTO procedures).
The same negative correlation is much weaker for CTO procedures on males,
despite the wider age range for males in the dataset. There is a strong positive correlation (+0.96,
+0.97) of PSD with patient weight and BMI for female PCI MV but the correlation is only moderate for male PCI MV (+0.47,
+0.46).
Further investigation of the discrepancy between male and female PCI dose metrics will be carried out.
Also,
further work needs to be done to correctly classify procedures.
This includes creating exam codes that capture the differing degree of complexity in PCI procedures. Additionally,
correct use of existing exam codes could be facilitated by use of an HL7 feed from the cardiology information system to the dose tracking system to update procedures where the registered exam is changed on the table but is not retrospectively corrected on the acquisition or dose tracking systems.