Nuclear medicine staff exposure to ionising radiation in ¹⁸F-FDG PET/CT practice: a preliminary retrospective study

This retrospective study analysed staff irradiation from June to December 2018 recorded on thermoluminescent dosimeters (TLDs) resulting from exposure during routine diagnostic procedures with a Discovery 610 PET/CT scanner (General Electric, Chicago, IL, USA) and MEDRAD^? Intego PET Infusion System for automatic dispensing and injection of radiopharmaceuticals (Bayer HealthCare LLC, Whippany, NJ, USA). MEDRAD^® Intego can deliver ¹⁸F-FDG within ±10 % of the prescribed dose and within ±2 % of the measured dose (12). The only radiopharmaceutical in use during the observed period was ¹⁸F-FDG. It was delivered to the NM Department in a transport container in the form of single doses of fixed activity (3.70 GBq or, less frequently, 5.55 GBq each).

We followed six staff members, four of whom were certified radiological technologists (T1, T2, T3, T4) and two certified nurses (N1, N2). Typically, they were organised in shifts of two technologists and one nurse.

The age of the technologists was 38 (T1), 42 (T2), 37 (T3), and 31 years (T4) and of the nurses 37 (N1) and 42 years (N2). These six staff members were also occasionally involved in other nuclear medicine practices. However, from June to December 2018, they worked only on PET/ CT and performed 423 procedures (1–3 per week) in patients with an average weight of 80 kg (range 31–160 kg) and age 60 years (range 6–86 years). The number of procedures performed differed between staff members: T1 (142), T2 (53), T3 (102), T4 (126), N1 (189), and N2 (234).

Dosimetry

The NM staff was equipped with two passive TLDs each, one for the whole body and one for the hands (ring dosimeter). The whole-body TLD is worn on the chest, whereas the ring dosimeter is worn on the base of the index finger of the operator’s dominant hand (Figure 1).

Figure 1

Both whole-body and ring TLD readouts corresponded to one month of cumulative dose measurements, as stipulated by country regulation (11). TLDs are read by the Personal Dosimetry Laboratory at the Public Health Institute of the Republic of Srpska. The Laboratory holds an accreditation according to ISO/IEC I17025 standard for measurements. The TLDs used in this study are of the MTS-N (LiF, Ti) type and are read by a semi-automatic RADOS RE-200 TLD reader. The detection limit for this TLD type is 26 μSv for one-month measurements, and measurement uncertainties for the whole body and ring TLDs are 28 % and 33 %, respectively (4).

Personal dosimeters are routinely calibrated for Hp(10) and Hp(0.07) at the Secondary Standard Dosimetry Laboratory (SSDL) of Vinča Institute of Nuclear Sciences, Belgrade, Serbia, which has metrological traceability to the primary standard at the International Bureau of Weights and Measures (BIPM) through the International Atomic Energy Agency (IAEA) Dosimetry Laboratory. Furthermore, SSDL is accredited to the ISO/IEC 17025 standard.

NM department had in place all appropriate radiation protection procedures for PET/CT imaging. Lead-shielded transportation containers were used for the transport of the radiopharmaceutical according to international standards and national legislation on ionising radiation protection, which included 10 cm thick lead bricks (occasionally used as an additional tool to shield the radiopharmaceutical during its transport from the manufacturing facility to the NM Department). The NM staff also used 30 cm long grasping forceps for safer radiopharmaceutical manipulation and the MEDRAD^® Intego PET infusion system with the following radiation shielding profile: dose rates of 0.26 mSv/h up to 2.3 mSv/h for the vial shield, peak rate ≤0.014 mSv/h at 30.5 cm distance from any surface of the cart with 27.75 GBq in the vial, and peak rate <0.001 mSv/h at 30.5 cm distance from the surface in the operator position with 27.75 GBq in the vial (12). The injector itself is equipped with tungsten and lead shielding.

Personal protective garments such as lead aprons, gloves, and thyroid shields were also available, but not routinely used when handling ¹⁸F-FDG, as they have been found no to provide significant protection (14). Nonetheless, individual preferences whether to use them or not were always respected. Wearing protective garments may help to reduce exposure in rare cases of contamination with liquid spills. It is also worth mentioning that the whole-body dosimeter, worn at the chest level, may have occasionally been covered by the lead apron.

Latex gloves were routinely used for all types of operations. They do not protect against beta or gamma radiation but provide a physical barrier between the radiopharmaceutical and operator’s skin in case of a spill.

Shielding barriers also included walls, doors, and lead glazing of the PET/CT scanning and adjacent rooms. PET/ CT scans were performed from the control room shielded to block all X-ray and most of gamma radiation. This means that the positron-emitting radionuclide was the major contributor to the radiation dose to which the technologists and nurses were exposed.

In addition to these protective measures, the PET/CT facility was designed to ensure a smooth workflow (see subsection below) to avoid loss of radionuclide activity due to lag times and to minimise time spent in proximity of an injected patient. The negative pressure ventilation system was independent of the rest of the hospital to limit the circulation of radioactive particles (gasses and aerosols) and expel radioactive particles. The personnel had also received manufacturer-specific, hands-on training to work with the GE Discovery 610 PET/CT unit and the MEDRAD^® Intego PET infusion system ever since the PET/CT system was installed in March 2016.

Workflow

Figure 2 shows the floor plan design of the PET/CT facility. The technologists and nurses had separate responsibilities for the entire duration of patient scanning. Other professionals involved in PET/CT procedures (such as NM physicians) did not routinely perform any of the activities that might have resulted in measurable exposure to ionising radiation from radiopharmaceuticals used in PET/CT procedures. The workflow could roughly be divided into two phases: the first carried out by a nurse and the second by a technologist.

Figure 2

In phase 1, the nurse receives the container with ¹⁸F-FDG and places it into the automatic dispensing system. The system then calculates ¹⁸F-FDG activity to be dispensed and injected according to the EANM guidelines (10). Before ¹⁸F-FDG injection, all the scheduled patients are identified and registered. The nurse takes the patient from waiting room 1 to the patient prep room, where the patient’s blood glucose levels are measured and an intravenous cannula placed. If blood glucose levels do not fall within the reference interval even after a second measurement, the procedure is postponed to another day. Otherwise, the nurse takes the patient to the ¹⁸F-FDG injection room. The automated injection process takes about 30 seconds, and during that time, the nurse waits outside the injection room to avoid exposure. After injection, the patient is instructed to rest on the bed in one of the two patient uptake rooms for 60 minutes, covered with a quilt or two blankets.

In phase 2, at the end of the uptake time, the technologist sitting in the PET/CT control room instructs the patient via speakerphone to go to the patient bathroom to empty her/ his bladder, and escorts her/him to the scanning room, positions the patient on the scanner bench, and makes the scan. Scanning takes between 20 and 45 minutes, depending on the size of the region scanned and the protocol used (usually ‘whole body’ or ‘total body’). Whenever the total body scanning protocol is performed, the technologist must re-enter the scanning room and reposition the patient, which results in increased exposure. Then the technologist escorts the patient to waiting room 2, where the patient waits for the physician and/or radiologist to review the obtained PET/ CT images. If the images are good, the patient is released from the Department. Otherwise, the patient is immediately asked to do another round of imaging. The same procedure is repeated until all the scheduled patients are processed.

From 2016 to 2019, the annual number of PET/CT procedures at the NM Department increased from 208 to 876, which clearly confirms an increasing trend in PET/CT scans observed worldwide (5).

From the total number of procedures and injected activities over the seven months of our study (Table 1) we calculated the average activity of 258 MBq per procedure, which is slightly lower than activities per procedure reported in Lithuania (283 MBq) (7), Oman (298 MBq) (8), or India (308 MBq) (15) and mainly reflects differences in parameters between patients (15), use of different guidelines for radiopharmaceutical injection in terms of patient age and weight, and the uncertainties in injection dose measurements.

Figures 3 and 4 show normalised monthly whole-body and dominant-hand exposure doses for technologists. We can see that normalised doses are not uniformly distributed across the months and individual technologists, as they depended on the distribution of working days and workload of each technologist. Normalised monthly whole-body and dominant-hand exposure doses for nurses are shown in Figures 5 and 6.

Figure 3

Figure 4

Figure 5

Figure 6

Cumulative normalised doses for whole-body and dominant-hand exposure for the same period are shown in Figures 7a and b for the technologists and 8a and 8b for the nurses. These figures show that the nurses’ hands were more exposed to radiation than those of the technologists.

Figure 7

Table 2 compares the monthly normalised exposure doses for technologists and nurses. We found a strong correlation between normalised exposure doses Hp(10)/A and Hp(0.07)/A for technologists (r=0.61), Hp(10)/A and Hp(0.07)/A for nurses (r=0.85), and between Hp(10)/A for technologists and nurses (r=0.65). These results reveal that the compared pairs of normalised doses changed similarly over time. In contrast, the correlation between Hp(0.07)/A for technologists and nurses was weak (r=0.25). No significant differences between mean normalised doses were found.

Table 3 presents the Hp(10) and Hp(0.07) cumulative doses and activities per staff member for the observed seven months. Average cumulative Hp(10) and Hp(0.07) doses (with standard error of the mean, SEM) were (505±171) μSv and (462±91) μSv for the technologists and (850±258) μSv and (1385±353) μSv for the nurses, respectively. The average cumulative Hp(10) and Hp(0.07) doses for all staff members were (0.62±0.17) mSv and (0.76±0.21) mSv, respectively. As the cumulative dose is a linear function of time (5, 16), we extrapolated the average annual cumulative dose for the PET/CT staff to be (1.06±0.29) mSv for Hp(10) and (1.15±0.32) mSv for Hp(0.07), which is significantly lower than the respective limits of 20 mSv (or the dose constraint of 2 mSv) and 500 mSv (11).

Table 4 provides a comparative analysis of our data to those published in the literature. Hp(10)/procedure and Hp(10)/A of injected ¹⁸F-FDG established in our study are in good agreement with several other reports (5, 6, 8, 17) but not with some other studies (7, 15), probably due to different workflow between the compared centres. Hp(0.07)/procedure in our study was compared with only one study (18), and a weak agreement was found. In our opinion, this is mainly due to the different processes of dispersion and injection of the radiopharmaceutical. To our knowledge, there are no other data for Hp(0.07)/A of injected ¹⁸F-FDG in the literature that could be compared to our values.

Our study has several limitations. First, the dosimeters are supposed to be read in regular intervals (once a month), but retrospectively we found that readout times sometimes were longer than a month. Another issue with retrospective studies of this kind is that one cannot check whether dosimeters were worn correctly or whether additional exposure occurred from accidental ¹⁸F-FDG extravasation in patients with obstructed veins.

The doses received by our staff were well below the legal limits and are in good agreement with the doses reported in other available studies. Regardless of this, it is necessary to further optimise the procedure following the ALARA principle and to investigate in more detail the ergonomics and exposure potential of each step.

Future studies in this field should address staff eye doses for PET/CT procedures, as almost all studies made so far lack these data.

The limitations of this retrospective study also provide a helpful indication how to improve further research by prospectively monitoring staff at risk of exposure with real-time dosimeters in combination with TLDs and by tracking exposure from various radiopharmaceuticals.

Figure 8