The evaluation was performed based on the guidelines of Fleischner Society which consider both density and size of ground glass nodules.
These can be the expression of benign conditions (organizing pneumonia,
focal fibrosis,
areas of inflammation or bleeding),
pre-invasive lesions,
such as atypical adenomatous hyperplasia (AAH) or adenocarcinoma in situ and,
less frequently,
pulmonary metastasis (1,2,3). According to these guidelines,
no follow up is required for pure ground glass nodules of size < 6 mm,
while for pure ground glass nodules of size ≥ 6 mm,
a CT scan has to be performed at 6-12 months to confirm its persistence,
followed by CT at 3-5 years.
No follow up is required for partially solid ground glass nodules of size < 6 mm,
while for partially solid ground glass nodules of size ≥ 6 mm,
a CT scan is performed at 3-6 months to confirm persistence,
followed by annual CT for five years (4).
The equipment we used is Delfinum 8000,
GE (General Electric,
Milwaukee,
WI,
USA).
The workstation has a specialized computer and an image database.
The X-ray tube has a high thermal capacity and a focal spot of about 1 cm.
The GE X-ray digital detector uses amorphous silicon technology that converts incident radiation into typed data.
The data is then processed into high quality images. The volumeRAD (3D X-ray) provides a series of low dose exposures during a single scan with the stationary steady-state detection system and the X-ray tube which is within a limited angular range (5).
After the acquisition,
the digital system reconstructs the data of the volume of interest. In this way the multiple planes parallel to the detector are displayed.
This process is based on pre-established reconstruction parameters,
in relation to the thickness of the patient,
chosen in the acquisition protocol.
Literature studies report effective dose values for chest Digital Tomosynthesis between 0.1 and 0.2 mSv,
with standard settings of different equipment (6,7).
In our study was used an anthropomorphic phantom produced by the Japanese company KYOTO KAGAKU (Chest phantom N1 “LUNGMAN”) in which the pulmonary structures are very similar to the anatomy and density of the human chest (Fig.1).
The phantom set includes the main body that consists of synthetic bones in epoxy resin and calcium carbonate,
the mediastinum,
the soft tissues and vessels in polyurethane and a total amount of fifteen synthetic nodules with spherical shape different in size (3,5,8,10,12 mm) and Hounsfield units (-800,
-630,
100 HU) (Fig.2).
First four DTS acquisitions of the nodule-free phantom were carried out,
keeping the acquisition parameters daily used:
- 120 KVp
- 0.25 mAs and 125 mA
- Dose Rateo 10
- 0.1 Cu filter
To verify the possible variation of the effective dose,
in excess or in defect,
we have moved on to another 12 acquisitions,
4 for each available Cu plate (0.0,
0.1,
0.2,
0.3).
Compared to our clinical practice in which we use the 0.1 Cu filter,
the 0.2 Cu filter allowed a lower effective dose with a good diagnostic confidence.
All the five dimensions provided by the set of simulated nodules,
with different densities,
were used:
1) -630 HU to simulate intermediate density nodules
2) -800 HU to simulate pure ground glass nodules
Starting from the right lung,
we chose to place the lesions in four different areas of the chest: apical (Fig.3),
hilar (Fig.4),
basal (Fig.5) and margino-costal region (Fig.6).
For each of them we performed a CT scan to confirm the position of the nodules.
Then we moved on to acquiring Digital Tomosynthesis.
For each of them,
7 DTS scans were carried out,
for a total of 35 acquisitions.
Initially only the Dose Rateo value was changed while the standard parameters of the equipment were kept fixed; mAs and mA values were recorded (Tab.1).
Subsequently,
the values of kVp were modified,
keeping the value of Dose Rateo fixed at 10 and also for these we recorded the parameters that remained unchanged in all acquisitions (Tab.2).
Two radiologists expressed in consensus a visual score for both DTS and CT images,
evaluating the visibility and the position of every single lesion:
Score 4: unequivocal identifiable nodules
Score 3: good identifiable nodules
Score 2: poor identifiable nodules
Score 1: no identifiable nodules
The Mann Whitney U test was used to confirm the results.
The significance level was set at P < 0.05.