Blood sampling in PET studies
Absolute quantification of PET image data requires measurement of input function. Input function (delivery function) describes the concentration of the unchanged (non-metabolized) radiotracer in arterial plasma (or blood, depending on the tracer) as a function of time (PTAC or BTAC).
Blood radioactivity concentration can be measured from the PET image, if a large artery or LV heart cavity is visible in the image. Even then a few blood samples may be needed for
- scaling image-derived blood data to the correct level, if data are biased because of partial volume effect,
- scaling of population mean or model based blood data to the correct level,
- measuring plasma-to-blood ratio, if concentration in blood needs to be corrected to concentration in plasma,
- measurement of the fraction of unchanged tracer in plasma, if metabolite correction is needed,
- measurement of hematocrit, if it is needed in blood-to-plasma or plasma-to-blood transformation,
- measurement of glucose, FFA, or drug concentration, or oxygen saturation, in case that is needed in the quantitative analysis.
Before the PET study arterial catheterization or intravenous cannulation is performed. While placing an arterial line is safe and reliable, it is burdensome. Venous injection catheter can be used for venous sampling (Hoekstra et al., 2000; Ponto et al., 2002). For collection of “arterialized” venous blood, the hand is heated with hot air or water.
For withdrawal of blood samples, catheter is connected to saline drip via three-way cock. After tracer inhalation or injection via another vein and catheter, blood samples are collected by a syringe. A 2 mL waste sample (containing mainly saline) is taken before collecting the actual blood sample. Blood is transferred from syringe to a heparinized tube, avoiding hemolysis. Heparin and blood is mixed by turning the tube upside-down. Tube is then immediately placed on ice to slow down further metabolism and transport from plasma to blood cells, which could introduce significant errors (Bungay et al., 1994). The exact time of blood withdrawal is recorded.
Blood sample tubes are centrifuged (at +4 °C) to separate plasma. A certain volume of plasma is then moved to another tube, and its radioactivity is measured using gamma counter. Events are collected until a predefined count- or time-limit is reached, assuring that the measurement noise does not increase with lower radioactivity concentration, unless the concentration is extremely low and measurement would otherwise last too long time.
If also concentration in total blood needs to be measured, then the rest of the sample tube is also measured using gamma counter, and the tube is weighed. Concentration in blood can then be calculated from the sum of sample radioactivities and sum of weights, taking into account the physical decay between the measurements. Finally, all radioactivity concentrations are calibrated to kBq/mL and decay corrected to the injection time of the radiotracer.
Separate blood samples are collected for metabolite analysis. These tubes may contain inhibitor, preventing further metabolism of the tracer in vitro. Blood tube must still be immediately placed on ice to further slow down the metabolism and transport from plasma to blood cells (Bungay et al., 1994). Plasma is separated by centrifugation at +4 °C. Plasma proteins are precipitated, usually with acetonitrile followed by centrifugation. Protein free plasma supernatant is then analysed by chromatographic methods.
The total radioactivity of blood, plasma and urine samples is obtained by automated gamma counters (1480 Wizard 3”, Wallac, Turku).
Arterial whole blood radioactivity concentration can be measured from arterial line with an automatic blood sampling system (ABSS, “blood pump” with on-line detector) to record the curve precisely after radiotracer bolus infusion. In long PET studies ABSS is used only during the first 3-5 minutes p.i. and manual blood sampling is continued after that.
ABSS provides more precise results than manual blood sampling when the dynamics of the tracer in the blood is very fast, especially in oxygen-15 studies. However, longer tubing is needed in ABSS than with manual sampling, which may lead to higher dispersion error. ABSS measures only activity concentration in whole blood, and conversion to concentration in plasma may be error prone with some tracers. Arterial catheterization is required with ABSS.
Separate blood samples for metabolite analysis and measurement of plasma-to-blood ratio can be collected while the ABSS is still working. These samples can be taken from the end of the ABSS tubing, so that it does not any disturbance in the on-line data. For this purpose, ABSS can have a carousel system holding the blood tubes (Espagnet et al., 2017).
Internal carotid artery is close to the skin, and therefore positrons from the arterial blood can penetrate into a transcutaneous detector. Positron signal can be separated from γ-ray background, but arterial signal cannot be fully separated from the venous signal; yet, the peaks of arterial and venous curves can be detected, and this kind of system could be used as a complimentary method for determining the circulation time through the brain (Litton & Eriksson, 1990; Watabe et al., 1995). In radiowater studies, an external detector monitoring γ-rays from the superior lobe of the right lung provided BTAC equivalent to the blood sampling (Nelson et al., 1993).
Intra-arterial probe, combined with external detector which monitors the background, could be used to measure AIF (Pain et al., 2004; Lee et al., 2008), but the method may still suffer from background activity from tissues and vein close to the artery, especially when measuring AIF in small animals with small arteries. Radiosensitive probes can also be used to measure tissue radioactivity in small animals without PET scanner (Pain et al., 2002; Weber et al., 2003).
Arterial catheter is a short thin tube that is placed into an artery of the wrist, arm, or leg, to collect arterial blood samples (Clark & Kruse, 1992; Stafford, 2003). The most common site of cannulation is the radial artery. Local anaesthesia is applied to the puncture site. Absolute contraindications for arterial catheterization are rare, including thromboangiitis obliterans (an inflammatory vasculopathy) and Raynaud syndrome. Relative contraindications are common, including atherosclerosis, anticoagulation therapy, and inadequate collateral circulation.
Arterial cannulation is safe (Slogoff et al., 1983; Wilkins, 1985; Weiss & Gattiker, 1986; Jons et al., 1997; Scheer et al., 2002; Brzezinski et al., 2009), and success rate is high when guided by ultrasound (Benedik, 2014; White et al., 2016; Gottlieb & Bailitz, 2016). Complications do occur (Mandel & Dauchot, 1977; Wallach, 2004; Chim et al., 2015; Simon et al., 2017), but mainly when arterial lines are kept for days during hospital treatment in otherwise ill patients. Possible complications of arterial catheterization include pseudoaneurysm, permanent ischemic damage, thrombosis, nerve injury, and compartment syndrome. Short-term arterial cannulation in physically healthy subjects is safe; Everett et al (2009) reported that in their 924 PET research subjects, with 1132 arterial cannulations, only one symptomatic thrombotic occlusion occurred, without ischemic damage, and condition was was resolved without intervention.
Despite being safe in healthy subjects, the tolerance for invasive research procedures can be lower in ailing and ageing populations, especially in case of repeated arterial cannulations. This may delay study subject recruitment or increase drop-out rate in longitudinal studies (Kang et al., 2018).
Arterial catheterization is feasible in rats (Graham & Lewellen, 1993; Weber et al., 2002; Sharp et al., 2005; Wu et al., 2007; Ravasi et al., 2012; Roehrbacher et al., 2015; Huang et al., 2016; Sijbesma et al., 2016). Loss of blood could easily affect physiological stability of small animals, and therefore on-line measurement of BTAC from (femoral) arteriovenous shunt that returns blood to the circulation is preferred (Weber et al., 2002; Warnock et al., 2011). Intra-arterial or -venous beta probe is another option, but, like arteriovenous shunt, requires microsurgery (Laforest et al, 2005). Automatic blood sampling systems for rats and mice have been developed and validated (Convert et al., 2007; Ose et al., 2012; Alf et al., 2013; Roehrbacher et al., 2015; Napieczynska et al., 2018).
Blood sampling from mouse tail yields a mixture of arterial and venous blood. Cardiac puncture can yield venous or arterial blood, or their mixture (Hoff, 2000).
In intravenous (IV) cannulation a cannula is placed inside a vein to provide venous access for administration of radiotracer, fluids, and medications, and/or sampling of venous or arterialized venous blood. There are no absolute contraindications for IV cannulation. Diabetes is a risk factor for difficult venous access; ultrasound guidance or transillumination should be used when veins are not easily palpable. Local anaesthetic cream should be applied topically ∼30 minutes prior to IV insertion, or topical anaesthetic should be injected intradermally just before cannulation to reduce the pain.
The veins of the arms and hands are preferred to those of the legs for intravenous cannulation, because cannulation of a leg vein hinders walking and poses a higher risk for blood clots and venous inflammation (thrombophlebitis). Vein valves impede the penetration of the catheter; valves are less common in the straight portions of veins and in the upper extremities.
Possible complications include accidental arterial puncture (Kang et al., 2014), thrombophlebitis, peripheral nerve palsy, and compartment syndrome.
Blood sampling may fail due to vein collapse or venospasm, or because needle hub is in contact with venous valve or penetrates the posterior wall of the vein.
- Input function preprocessing
- Metabolite correction
- Input function
- ABSS data
- Image-derived input function
- Population-based input function
Espagnet R, Frezza A, Martin J-P, Hamel L-A, Lechippey L, Beauregard J-M, Després P. A CZT-based blood counter for quantitative molecular imaging. EJNMMI Phys. 2017; 4:18. doi: 10.1186/s40658-017-0184-5.
Frayn KN, Macdonald IA. Methodological considerations in arterialization of venous blood. Clin Chem. 1992; 38(2): 316-317.
Hoff J. Methods of blood collection in the mouse. Lab Animal. 2000; 29(10): 47-53.
Jons PH, Ernst M, Hankerson J, Hardy K, Zametkin AJ. Follow-up of radial arterial catheterization for positron emission tomography studies. Hum Brain Mapp. 1997; 5(2): 119-123.
Lee K, Fox PT, Lancaster JL, Jerabek PA. A positron-probe system for arterial input function quantification for positron emission tomography in humans. Rev Sci Instrum. 2008; 79(6): 064301. doi: 10.1063/1.2936880.
Litton J-E, Eriksson L. Transcutaneous measurement of the arterial input function in positron emission tomography. IEEE Trans Nucl Sci. 1990; 37: 627-628. doi: 10.1109/23.106688.
Nelson AD, Miraldi F, Muzic RF Jr, Leisure GP, Semple WE. Noninvasive arterial monitor for quantitative oxygen-15-water blood flow studies. J Nucl Med. 1993; 34(6): 1000-1006.
Pain F, Lanièce P, Mastrippolito R, Gervais P, Hantraye P, Besret L. Arterial input function measurement without blood sampling using a β-microprobe in rats. J Nucl Med. 2004; 45(9): 1477-1582.
Ponto LLB, Graham MM, Richmond JC, Ward CA, Clermont DA, Schmitt BA, Clark J, Conklin A, Weldon L, Watkins GL, Madsen MT, Hichwa RD. Contamination levels in blood samples drawn from the injection intravenous line. Mol Imaging Biol. 2002; 4(6): 410-414. doi: 10.1016/S1536-1632(02)00121-X.
Updated at: 2018-01-12
Created at: 2006-05-23
Written by: Vesa Oikonen, Pauliina Virsu, Tuula Tolvanen, Anne Roivainen