This study reviewed IAC procedure to provide a feedback to treating neuroradiologists and to check the adherence to the principles of radiation safety and ALARA.
Notably,
optimization of neurointerventional procedures is the most important step to take for radiological teams in order to reduce doses and maximize results as recommended by the European Commission (14).
This study was also carried out as a part of a multidisciplinary project aimed to establish a diagnostic reference level (DRL) for this paediatric neurointerventional procedure,
as recently recommended by the European Commission (15).
It is well known the relevance that IAC has gained worldwide (including our University Hospital with one of the largest experience in Europe) in the treatment of intraocular retinoblastoma.
Concerns,
however,
have been raised over the level of ionizing radiations to which children might be cumulatively exposed after several sessions.
It is therefore important to establish as objectively as possible whether children are at high risk of developing long term side-effects of radiations.
We measured the main radiation parameters including fluoroscopic time,
DAP as recorded by the angiographic system,
and ESD as assessed by a dosimeter placed along the X-ray beam entrance pathway.
Our investigation included cases where drug administration was performed by direct catheterization of the OA and by alternative routes through the ECA in children with unfavorable patterns of OA artery flow.
In our experience IAC has reached a good level of standardization.
The mean fluoroscopic time was 9±6.7 min,
and DAP resulted 595±445 cGy·cm2.
Whereas fluoroscopic times are in line with those previously reported by others (16,18),
DAP values appear higher than previously reported (16) though still largely acceptable.
Our results confirm that fluoroscopic time is an important variable that affect the radiation dose.
The length of the fluoroscopic time may reflect several variables including operator experience,
microcatheter navigation through the arterial tree to reach the OA as well as the modality to get access to its ostium (16).
Notably,
the blood flow within the OA may derive entirely from the ICA,
from the ECA or it may come from both arteries (9).
In such cases,
the fluoroscopy time needed to search and find the appropriate pathway of drug delivery may vary significatively (8.4±5.5 min for OA catheterization vs.
16.62±12.2 min for ECA branches catheterization).
In addition,
the hemodynamic pattern within a single IAC session may result stable or unstable.
In the latter instance,
the pattern can switch from an ICA dominance towards an ECA dominance or vice-versa (17) increasing the fluoroscopy time.
This is demonstrated by the fact that the main contribution to the overall irradiation dose derives from phase 4 (in lateral projection and using a small FOV) during which the possible alternative pathways of drug delivery are searched.
DAP values parallel fluoroscopic times.
When the OA is catheterized,
DAP is significatively lower than when catheterization of ECA branches is required (526±328 cGy·cm2 vs.
1257±934 cGy·cm2).
Describing DAP and fluoroscopy times in terms of mean ± SD,
as was previously done (8,
16,
17,
18) in our opinion is misleading as the distribution of these parameters plotted as a function of their frequency is not a Gaussian one; with the exception of phase 4,
fluoroscopy times and DAP of all other steps were more frequently confined to the left of the graphs.
The main value of this work is in providing a method that ultimately allow the estimation of the absorbed doses of internal organs starting from DAP value for prospective radiation tracking and risk association. Measuring the exposition of internal organs by repeated irradiations on an anthropomorphic phantom with the same settings adopted in vivo we could establish correlations between: 1) DAP and the dose recorded by a single sTLD placed along the X-ray beam entrance pathway; 2) the dose recorded by the sTLD placed along the X-ray beam entrance pathway and the dose absorbed by internal organs.
Our results show that the radiation doses absorbed by the thyroid is lower than toxic levels in most instances.
After six sessions,
the cumulative dose is 30 mGy at the 75th percentile and 62 mGy only at the 95th percentile.
However,
these values are obtained multiplying for 6 the dose absorbed on a very long session of IAC.
Indeed,
a long session of IAC is usually due to a complex vascular anatomy and/or hemodynamic which requires a careful study.
This is usually accomplished in the first sessions.
The following sessions,
therefore,
are almost always shorter and the cumulative dose should be considered far lower than the value reported at the 95th percentile.
As far as the lens is concerned,
ICRP revised the eye dose threshold for cataract induction to 0.5 Gy (19).
After 6 sessions,
we determined that the right and left lens absorbs respectively 47.8 mGy and 49 mGy at the 75th percentile and 98.9 mGy and 133.8 mGy at the 95th percentile,
values far below the cataractogenic levels.
In addition,
even in this case,
the value of 133.8 mGy has been obtained as if a child could undergo 6 very long sessions of IAC and this is highly unlikely.
This study has some limitations:
1) The sTLD5/DAP ratio can be used in any angiographic room equipped with the same equipment and operative settings employed in our Center.
2) The correlation between dose recorded by the sTLD located within the X-ray beam entrance pathway and the dose absorbed by internal organs is not as constant as the sTLD5/DAP ratio depending on several parameters (i.e FOV size and irradiation geometries).
3)The size of the child head influences the amount of ionizing radiations reaching internal organs.
A correction factor can be devised for each child (i.e using MRI scan) to assess the depth of the organ (or point) for which one wants to calculate the absorbed dose.
In any case,
based on the experiments carried out on the anthropomorphic phantom,
we could determine an attenuation curve of the dose absorbed by tissues within the head as a function of their depth.
In conclusion,
this work demonstrates that,
in experienced hands,
the dose ranges during IAC are lower than thresholds for deterministic and stochastic risk.