Radiation dose to patients and staff during angiography of the lower limbs. Derivation of local dose reference levels

Investigation procedure 

DSA of the lower limbs was performed according to a standard protocol, by two experienced radiologists and four training radiologists. The procedure was performed utilizing Philips Integris 3000, an integrated X-ray system for all interventional procedures.

The area selected for the puncturing was primarily the right femoral artery, but when this was impossible, the left artery was chosen. Initially, the catheter, fluoroscopically guided, was placed at the level of the renal arteries and after the administration of contrast medium digital images were acquired usually with a frame rate of two images per second (cine mode). Rarely, and if deemed necessary, a higher frame rate was selected (four images per second). Then, the X-ray tube was moved towards the feet in order to visualize all the arteries, usually acquiring one image per second. Whenever necessary, extra images were acquired with a tilt of the X-ray tube over the region of interest (two cases) or images were acquired for one of the two feet.

The TLDs were calibrated under reproducible reference conditions using the same X-ray machine against an ionisation chamber model 9060/10X5-60 connected to a Radiation Monitor Controller model 9010 (Radcal Corporation, Monrovia, CA, USA). Both the chamber and the electrometer were calibrated for the energy range of 30–120kV at the National Standard Laboratory (National Metrology Institute of the Greek Atomic Energy Commission). Calibration cycle of the TLDs was carried out every month in order to extract individual calibration factors.

Patient dose measurements 

A total of 30 patients participated in this study. Indications included diagnosis and follow up of vascular diseases. The study was approved by ethics and research committee at UHL.

Groups of four TLDs were packed into sachets. Eight sachets were positioned corresponding to five different exposure areas. An overall statistical uncertainty associated with the measurements of the TLDs was estimated to be 8%. All sachets were attached at the back of the patient in order to measure ESD. One sachet was placed at the level of aorta bifurcation, one at the pelvic area and 15cm lower from the first one, one at the middle of the each femur, one at the back side of each knee and one at the back side of each foot. These positions were selected as they correspond to the positions where dynamic frames were acquired. ESD was measured as a total dose from both fluoroscopic and radiographic exposures. For each patient all the exposure parameters (i.e. radiographic data: kV, mAs, number of frames and region, fluoroscopic data: kV, mA and total screening time) and patient data (age, weight, height, clinical indication, and radiologist name) were recorded.

Patient effective dose 

The ESD was used to calculate the effective dose. Estimation of the effective dose was done by the use of conversion factors provided by the National Radiological Protection Board (NRPB-SR 262) [13], [14]. The projections that were selected are: kidney PA for the TLDs that were at the level of the aorta bifurcation, pelvis PA for the TLDs that were at the level of the pelvis area, and shoulder PA for the rest [12]. The total effective dose was calculated by the addition of effective dose for each projection.

Occupational dose measurements 

One sachet containing four TLDs, was attached outside the lead apron of 0.25-mm Pb equivalent thickness, at the chest. That position was selected, since the chest is the most common body part for dose evaluation using appropriate conversation factors [15].

Protective eyeglasses and thyroid collar (0.5-mm Pb equivalent thickness) were always worn. The staff radiation dose in diagnostic X-ray departments is routinely monitored by the official Greek authorities (Greek Atomic Energy Commission), with a dosimeter worn at the chest level, above the lead apron.

Patient exposure factors 

The mean fluoroscopic time during the procedure was 96s (1.6min) and the mean number of captured images was 103.7. Comparing these results with the ones from previous studies the screening time agrees with reported values, but the number of frames is quite large (Table 3).

Thwaites et al. [8] reported that, on average, fluoroscopy contributed to 10% of the total radiation dose. Ruiz Cruces et al. [7] estimated the fluoroscopy contribution to be 36%. Bor et al. [9] calculated the contribution of fluoroscopy to be 21%. Mini et al. [16] reported that the contribution of the fluoroscopy to the DAP was around 25%. Struelens et al. [11] noted that fluoroscopy does not seem to be a substantial irradiating part of the procedure and that the dose to the patient depends mainly on the number of frames, the exposure parameters and the position of the radiation field. Hoskins et al. [17] estimated that digital spot film exposure contributed to 88% of the total effective dose.

Thwaites et al. [8] reported the mean values of kV at the same areas of measurement during this study. The agreement, between corresponding areas, is good with a small deviation, of 10% maximum with the present study (Table 1). Ruiz Cruces et al. [7] reported average kV during radiography to be 66kV. Struelens et al. [11] reported the results for eight different centres. The mean value of the results obtained during this study for the regions of abdomen and pelvis (71.25kV) is very similar to the values of the present study with two exceptions, namely centre E and F1. In two centres, although they used similar exposure factors and number of frames, the effective dose is different (11.1–16.8mSv for centre E and 7.7–8.3mSv for centre F1). This difference may be due to the different fluoroscopy mode (different X-ray machine) or to the time of fluoroscopy. Bor et al. [9] reported average tube voltage 69kV, and from a survey at four different centres [18] the tube voltage ranges from 67 to 88kV.

The above values demonstrate that fluoroscopy time gives a low contribution to the total dose of the patient and that the effective dose does not strongly depend on the tube voltage. Thus, it can be concluded that the calculated dose for the present study is relatively high, despite the short duration of fluoroscopy, mainly due to the large number of frames acquired.

Patient effective doses, risk estimation 

The effective dose was derived from ESD using NRPB software [13], [14]. These factors were derived for a standard 70kg anthropomorphic fantom irradiated with standard field sizes, and their use introduces an error in the estimation of the effective dose. The data for the present study (weight, field sizes) differ from those of the NRPB phantom, and doses to real patients may deviate significantly from the NRPB data even when the ESDs are equal. The effective doses derived in the present work are thus only indicative, compelled the research team for this study to consider this error of minor significance.

The total effective dose is the sum of the effective dose at all the five regions. The regions contributing the most to the effective dose are the pelvic and the abdomen regions, while the smallest contributions stem from the femur, knees and feet. Struelens et al. [11] stated that 85% of the measured DAP stems from the exposure of the abdomen and pelvis. Bor et al. [9] assumed that 50% of the total DAP was contributed from the pelvic region. Hoskins et al. [17] estimated the contribution of torso exposure to be 98% of the total effective dose. From the results obtained during this study, it was calculated that the mean contribution of the abdomen and the pelvic regions to the effective dose to be 94.4%, (range 89.7–98.7%) of the total.

The mean effective dose per procedure is 9.8mSv, which is slightly higher than the mean effective dose reported in other studies. This can be explained by the large number of frames acquired, as the number of frames acquired contribute most to the effective dose. We must notice that the effective dose can be different than the reported one. This is not only due to the TLDs inaccuracy (8%) but also because there are places in the body that receiving dose and due to the luck of a dosimeter the dose is not recorded. Since the above happen mostly at the legs and their contribution at the total effective dose is minor we assume that the recorder dose is the total dose that the patient receives. It should be noticed that effective dose is not an additive quantity. The purpose of addition of each projection result is to be able to make simple comparison with the literature data which are usually given for the whole examination.

Radiation risks estimation for fatal cancer per procedure was found to be 5.4×10−4 (540 per million) using the fatal cancer coefficient of 0.055 per Sv [19].

Staff doses 

Occupational dose is principally caused by scattered radiation and leakage from the X-ray tube housing. During the examination, the radiologist performing the procedure was protected by a movable ceiling mounted shield. No table mounted lead curtain exists in the department. The mean ESD at the chest level was recorded to be 0.15mGy.

The dose to the staff dose is very low under protective shield, since the radiologist wore lead apron of 0.25mm lead equivalent with transmission varying from 4.3 to 10.2% at 70kVp and thyroid collar of 0.5mm lead equivalent with transmission varying from 0.6 to 1.6% at 70kVp [20].

Taking into account the ESD at the chest level and assuming a percentage of 10% transmission the effective dose was calculated to be 0.023mSv per procedure.

Using the conversion factors obtained from International Commission of Radiological Protection (ICRP) [6], the eye lens dose and thyroid dose was estimated to be 14.92, and 0.75μSv, respectively. Wearing thyroid collar and eyeglass reduces the dose to the radiologist to a negligible level.

In general, the radiation dose of the staff is proportional to the dose to the patient. Scatter radiation could be decreased by decreasing the fluoroscopy time and the field size but also with the installation of a curtain shielding that would decrease the radiation that reach doctors extremities.

Diagnostic reference level 

The European Union in 1997 [5] introduced the use of diagnostic reference levels (DRLs) as an efficient standard for optimizing the radiation protection of patients. DRL is not a dose limit, but should not be exceeded in normal practice without adequate justification in order to achieve an acceptable image quality or adequate diagnostic information, consistent with the medical imaging task.

Several quantities can be used as DRLs. Effective dose, ESD (mGy), DAP (Gycm2). Zoetelief et al. [22] proposed as DRL the 75% of the effective dose, which in the present report was 7.0mSv. Wall et al. [23] used the DAP value to propose DRL for femoral angiograph. He proposed a DAP value per exam of 33Gycm2 and 5min of fluoroscopy. For lower limbs angiography, Cruces et al. [7] calculated the 75% of the DAP value to be 36Gycm2 while Francis R. Verdun [24] found the 75% of the DAP value to be 226Gycm2.

Following the recommendations of ICRP 60 [6] and Euratom 97/43 [5], the research team of this study proposes a local DRL for DSA of the lower limbs that corresponds to 75% of the registered effective doses, which is an entrance skin dose of 96.4mGy in the pelvic region, as presented in Fig. 1.