Residual tracer activity in venous access devices rarely disrupts PET and SPECT scans

When patients arrive for a PET or SPECT scan, the radiopharmaceutical that will light up their organs has to get into the bloodstream somehow. In routine practice, nuclear medicine departments usually place a fresh peripheral intravenous catheter for the injection. But many patients, particularly those already undergoing treatment, arrive with a semi-permanent venous access device already in place: a central venous catheter, a peripherally inserted central catheter known as a PICC, or an implanted venous port sitting under the skin. A new observational study from Skåne University Hospital in Sweden now offers the most detailed real-world picture yet of how much injected tracer stays behind in these devices, and whether that leftover activity matters for the images doctors rely on.

The answer, in short, is that it usually does not matter much at all. Across 167 patient administrations performed between June 2023 and January 2026, covering 13 different radiopharmaceuticals and 214 individual measurements, residual activity was generally low. Only 3.3 percent of measurements, seven out of 214, exceeded 5 percent of the administered activity. The highest absolute residual activity recorded was 31 megabecquerels of [99mTc]pertechnetate left in a peripheral intravenous catheter, which corresponded to just 4.5 percent of the injected dose. The highest proportional retention was 17.3 percent, seen in a single administration of [99mTc]Tc-MAA, a lung-scintigraphy agent, through a midline catheter; that amounted to 19 megabecquerels.

Crucially for patients and clinicians, none of this leftover activity interfered with diagnosis. In the 84 cases where the venous access device fell within the clinical field of view, residual activity was visually detectable in 30 of them, either inside tubing within the patient or in parts of the injection system outside the body. Yet in not a single case did the study team find that device-related activity impaired interpretation or reduced diagnostic confidence, whether at the original clinical read or during a dedicated study review. The findings were published in EJNMMI Reports as an open-access original article.

The study matters because the literature on this question has been remarkably thin. Radiopharmaceuticals are drugs containing radionuclides, and intravenous administration is the most common route in diagnostic nuclear medicine. Guidelines on venous access generally do not address radiopharmaceutical injection specifically, and a systematic literature search by the researchers, spanning PubMed, Embase and INSPEC, yielded only seven relevant articles from an initial pool of 240, including two case reports and three conference abstracts. For marketed radiopharmaceuticals, the summaries of product characteristics sometimes give preparation and administration instructions, but only a single product, the lung agent Pulmocis, explicitly advises against administration via implanted venous ports.

Quantifying residual activity required two complementary approaches. For removable peripheral intravenous catheters, the researchers measured activity directly in a calibrated activity meter after the catheter had been removed from the patient, correcting for background radiation and radioactive decay. For devices that stayed in place, such as ports and PICCs, they drew regions or volumes of interest over the access site on reconstructed SPECT, planar or PET images and converted measured counts to activity using system-specific calibration factors. The residual activity fraction was calculated as the residual activity divided by the administered activity, with the administered dose itself corrected for activity retained in the syringe. In parallel, an experimental setup tested 45 venous access devices and an equal number of stopcocks under controlled conditions, injecting radiopharmaceutical into a lead-shielded glass vial and flushing with saline to mirror clinical routine.

The numbers by tracer class tell a reassuring story. For the four PET radiopharmaceuticals, residual activity fractions were vanishingly small: at most 0.3 percent for [18F]FDG across 30 measurements, at most 0.2 percent for [18F]fluorocholine across 40, and at most 0.1 percent for both [68Ga]Ga-DOTA-TOC and [68Ga]Ga-PSMA-11. Among the SPECT agents, [99mTc]Tc-DPD and [99mTc]Tc-HDP never exceeded 1 percent, and for [123I]MIBG, [99mTc]Tc-MAG3, [99mTc]Tc-tetrofosmin, [99mTc]pertechnetate, [99mTc]Tc-MBrIDA and [99mTc]Tc-DMSA the highest fraction stayed below 7 percent. Only one measurement, the [99mTc]Tc-MAA midline case, crossed the 10 percent threshold that is often treated as a pragmatic cutoff for potentially clinically relevant loss of delivered activity. In the experimental arm, only three of 90 measurements exceeded 5 percent, all involving [99mTc]Tc-tetrofosmin retained in stopcocks, with a maximum of 11.1 percent and 22 megabecquerels.

Why does retention vary at all? The researchers point to mechanistic properties of the tracers: hydrophilic small molecules such as [18F]FDG and [99mTc]Tc-HDP clear quickly, while larger or charged molecules, including peptides, antibodies and nanoparticles, may linger if not flushed efficiently. Injection technique, including flush speed and the volume of 0.9 percent sodium chloride used, also modifies the result. At the Swedish hospital, devices were flushed with saline before and after administration per local clinical standard, and during some [18F]FDG and [68Ga]Ga-DOTA-TOC examinations with contrast-enhanced CT, devices were flushed with up to 245 millilitres of saline and contrast agent combined. The adhesive nature of [99mTc]Tc-MAA, which contains macroaggregated albumin particles, is well known in clinical practice, which is why it is typically deferred from injection through filtered lines such as PICCs.

The cohort itself was broad and unusually inclusive of children: 40 percent of the 167 administrations, 67 in total, involved patients under 18 years of age. Repeated examinations from the same patient were treated as independent observations, with 23 patients contributing more than one scan. Of the total, 130 administrations were evaluated by direct activity measurement of removed catheters and 84 by image-based analysis. The study also had an ethical advantage: for PET patients the injection site was already visible on clinical scans, and for planar or SPECT examinations outside the field of view a static image was acquired, so no patient received any extra radiation exposure for participation. Sixty-eight administrations were identified retrospectively through the hospital picture archiving system under an ethics board waiver.

The authors are careful about the limits of what their data can show. The study was conducted at a single two-site institution, which may limit external validity, and no patient-level clinical characteristics were recorded, so variables such as venous access history and comorbidities could not be analysed. Data on device age, number of lumens, prior complications and catheter-related adverse events were not systematically collected. Potential technical explanations for the isolated high-retention case, such as catheter malposition, partial obstruction or injection resistance, were not evaluated. Static imaging has limited sensitivity for very low residual fractions, attenuation within the patient may cause underestimation, and the study-specific image review was performed by a single non-blinded reader. The effect of residual PET activity on standardized uptake value quantification was not formally evaluated, though the authors consider any impact negligible given the very low fractions observed. Therapeutic radiopharmaceuticals were out of scope entirely, and conclusions are limited to the agents studied.

Even with those caveats, the practical implications are considerable. The findings suggest that flexible intravenous access strategies are feasible in routine practice, including for adult and pediatric patients with difficult veins, without risking diagnostic reliability, and they do not support routinely administering extra activity to compensate for what stays behind. The widespread availability of SPECT/CT has also reduced the clinical impact of unexpected focal uptake, since improved anatomical localization allows such findings to be identified and dismissed. The descriptive results support using existing devices, potentially sparing patients the pain, hematoma, thrombophlebitis and infection risks of a new catheter placement, and easing workflow for wards and imaging departments alike. One notable exception stands out: in hospitalized patients undergoing [18F]FDG PET for suspected or confirmed infection, reusing a pre-existing device can be problematic, because local inflammation or thrombophlebitis near the catheter can produce focal FDG uptake that mimics residual tracer, so a new catheter with clear documentation is recommended in that setting. The authors also urge some caution around [99mTc]Tc-MAA until larger studies confirm the pattern, and they call for prospective studies across a broader range of tracers and devices with quantitative endpoints to fully define the clinical implications of residual activity.

Subject of Research: Residual radiopharmaceutical activity in venous access devices during PET and SPECT imaging

Article Title: Residual radiopharmaceutical activity in venous access devices during PET and SPECT imaging: a real-world clinical observational cohort study

Article References: Solem, R., Al-Amiri, N., & Skarping, I. (2026). Residual radiopharmaceutical activity in venous access devices during PET and SPECT imaging: a real-world clinical observational cohort study. EJNMMI Reports, 10(1), Article 33. https://doi.org/10.1186/s41824-026-00317-4

Image Credits: AI Generated

DOI: 10.1186/s41824-026-00317-4

Keywords: nuclear medicine, PET, SPECT, radiopharmaceuticals, venous access devices, residual activity, image interpretation, radiation safety, pediatric imaging, observational cohort, PICC, venous port