Answer to Question #12775 Submitted to "Ask the Experts"

Is it appropriate to use a Ludlum 44-9 probe (thin-window Geiger-Mueller (GM) as a calibrated survey meter for a positron emission tomography-computed tomography (PET-CT) clinic?

Specifically, I'm worried about the fact that most calibration labs use a ¹³⁷Cs source to calibrate meters. In this case, the ¹³⁷Cs calibration would be appropriate for the 511 keV annihilation photons coming off the positron emitters but the thin window of the 44-9 probe would also let in positrons which would cause massive over-response (depending, of course, on distance from source). This high sensitivity to positrons is why a lot of PET clinics use the 44-9 probes for contamination meters.

The type of detector you described is commonly used in nuclear medicine facilities and often used in situations involving PET radionuclides. Your concerns are justified in some aspects of use. I shall attempt to address some conditions of use, possible methods of dealing with the sensitivity to positrons, and interpretation of readings.

As you have inferred, it is reasonable to use the detector you describe for measuring the photon emission intensity from the PET radionuclide(s) since the 511 keV energy is close enough to the 662 keV calibration energy to be acceptable. Some of the scattered photon radiation that is degraded in energy may yield some overresponse, but it would likely be acceptable. As you have also implied, if the PET radionuclide source is in an unshielded configuration such that positron radiation can escape, the reading on the instrument may be heavily weighted by the positron radiation, and a meaningful estimate of dose rate may not be feasible. The intrinsic GM detection efficiency for the positrons is more than 100 times greater than that for the photons, and this accounts for the much greater sensitivity to incident positron radiation compared to photon radiation.

If the instrument is being used to look for surface contamination associated with the preparation and use of PET radionuclides, it would be acceptable for purposes of identifying contamination, but the quantitative interpretation of the reading may be difficult. In such instances, if you require a quantitative assessment of the contamination level (e.g., Bq or Bq cm^-2), shielding the detector face with sufficient thickness of plastic (e.g., Lucite) to shield all of the positrons would allow measurement of only the 511 keV and possible scattered photons. If the reading is high enough you may be able to estimate the activity contamination level if you have previously obtained a calibration factor for similar surface contamination of known amount, using the same positron shield and measuring at the same distance from the contamination. Increased sensitivity may be obtained using the positron reading obtained by subtracting the shielded reading from the unshielded reading. Again, a calibration would have to have been done to establish a counting efficiency or other reading-to-activity area concentration (or simple activity if the contamination area is small compared to the areal dimensions of the detector). The manufacturer may well have already determined counting efficiencies for conventional beta emitters of different maximum beta energies that may be suitable for interpreting contamination measurements associated with positron radiation. For ¹⁸F, the maximum positron energy is less than 650 keV, and essentially all of the positrons can be stopped by a lucite thickness of about 3 mm. If you are using ¹⁵O, the maximum positron energy is about 1.75 MeV, and about 8 mm of Lucite would be required to stop the positrons. Similar rationale would apply to other possible radionuclides.

Bare plastic syringes are typically not of sufficient wall thickness to stop all of the positrons emitted. Syringe shields are available and should be used to minimize dose, especially from the positrons. Measuring the photon radiation from syringes that are not adequately shielded against positron radiation or during parts of radionuclide preparation and/or administration when complete positron shielding may not be practical can be done by using an appropriate thickness of Lucite over the face of the detector. The dose rates from positrons emitted in such situations may be of concern from the point of view of dose to the lens of the eye and possibly dose to the live skin. Readings on the unshielded detector may show the influence of the positrons on the instrument reading, but proper interpretation of the dose may be difficult without further calibration of the instrument or specific calculations that may be used to estimate positron fluence rate and from that dose rate.

As a brief example, suppose that a small spot of ¹⁸F contamination was detected on a surface and that we measure a net count rate (unshielded minus shielded count rates) of 8.8 × 10⁴ cpm with the 44-9 probe with the detector window centered approximately 1 cm above the contamination spot. The manufacturer's specifications for the detector show a likely 4π detection efficiency, ε, of about 0.20 counts per emitted beta particle for beta radiation of about the same maximum energy as the ¹⁸F positron (about 634 keV) when the detector is centered about 1 cm above the small source. The expected positron emission rate, S, from the source would then be S = 8.8 × 10⁴ cpm/0.20 counts per β⁺ = 4.4 × 10⁵ β⁺ min^-1. Since the yield from ¹⁸F decay is 0.967 β⁺ per disintegration, the estimated activity, A, in the spot of contamination would then be A = 4.4 × 10⁵ β⁺ min^-1  per 0.967 β⁺ d^-1= 4.6 × 10⁵ dpm = 7.6 × 10³ Bq.

We should also recognize that once we have determined the positron emission rate from the source at a given location we can use that number to estimate positron fluence rate at the detector location. From that we could calculate the tissue dose rate near the surface of the body, which could be helpful for estimating skin dose or eye dose. The active facial area of the Ludlum 44-9 detector is 15.5 cm², implying an effective radius of 2.22 cm. It is easy to show that the average fluence rate over the active facial area of the detector is given by

Φ = (S/4) ln[(R² + H²)/H²]/ πR² (1),

where S is the total (4π) positron emission rate from the contamination spot, R is the effective radius of the active facial area, and H is the height of the detector above the contamination. For our example R = 2.22 cm, and H = 1.0 cm. The fluence rate may than be calculated from equation 1 as Φ = (4.4 × 10⁵  β⁺ min^-1/4) ln[(2.22² + 1²)/1²/15.5 cm²= 1.26 × 10⁴  β⁺ cm^-2 min^-1. The soft tissue dose rate at the same point would be calculated from the product of the fluence rate and the mass collision stopping power for the positrons. For estimative purposes we may use the value of the collision mass stopping power for soft tissue for the average energy positrons emitted by ¹⁸F (250 keV), a value of about 2.6 Mev-cm² g^-1. We then obtain an estimated absorbed dose rate, D, in units of cGy h^-1 of

D = (1.26 × 10⁴ cm^-2 min^-1)(2.6 MeV cm² g^-1)(1.6 × 10^-6 erg Mev^-1)(1 cGy per 100 ergs g^-1) (60 min h^-1) = 0.032 cGy h^-1.

We may not be interested in the tissue dose rate at such a close distance from the source, but the same approach could be used at other locations for which we could calculate the fluence rate of positrons.

Because the detection efficiency for the positrons is high compared to photon detection efficiency, even relatively modest amounts of activity may produce excessive count rates such that dead time problems may have to be considered. In the above example, the measured count rate was about half of the value at which dead time losses would start to become significant for the Ludlum 44-9 probe used with a typical rate meter. In cases where activities are higher it may be necessary to make measurements at greater distances from the contamination. There are also calculational methods that would allow estimation of activities and fluence rates at these greater distances. If you require more on this, please contact me.

After the radionuclide has been administered to the patient, the positron radiation will be completely attenuated by the body tissues, and readings in the vicinity of the patient will be associated only with the photon radiation, assuming that no exposed contamination is present on the patient or in the area.

I hope some of this may be useful to you.

George Chabot, PhD, CHP