TOPLINE

Second-line multiphase CT (four-dimensional [4D] CT) provided high specificity and reliable diagnostic accuracy for localising and lateralising parathyroid adenomas in patients with primary hyperparathyroidism (P-HPT) who had negative or discordant first-line imaging with technetium 99m (99mTc) bound to six methoxyisobutylisonitrile (MIBI) single-photon emission CT (SPECT)/CT and ultrasound (US) yields, a retrospective study showed.

METHODOLOGY

• Researchers conducted a retrospective cohort study including 73 patients with P-HPT (mean age, 56.2 years; 79.5% women) who underwent multiphase contrast-enhanced CT neck imaging between 2019 and 2023 following negative or discordant dual-tracer 99mTc-sestamibi SPECT/CT and US results.
• Patients were divided into two cohorts: a negative cohort (19 patients with no adenoma localised on either US or other scans) and a discordant cohort (54 patients with one negative study, contradictory results, or equivocal findings including a low suspicion for a lesion).
• Multiphase CT acquisition included a three-phase protocol with an initial non-contrast scan, followed by an arterial phase at 30 seconds and a delayed phase at 80 seconds after the injection of 75 mL of iodinated intravenous contrast at a rate of 4 mL/sec.
• Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, and positive and negative likelihood ratios were calculated for preoperative adenoma lateralisation (correct side) and localisation (correct quadrant) across four imaging quadrants.
• Diagnostic performance metrics for side lateralisation and quadrant localisation were calculated against surgical and histopathologic findings as ground truth.

TAKEAWAY

• Across the whole cohort, 4D CT demonstrated a sensitivity of 70.4% and a specificity of 86.4% for localising adenomas to the correct quadrant, with a corresponding PPV of 77.0% and NPV of 81.9%; for lateralisation, performance was numerically higher, with a sensitivity of 82.6% and a specificity of 86.2% (overall accuracy, 84.0%).
• Positive likelihood ratios for localisation and lateralisation were 5.26 and 5.99, respectively, indicating a moderate rule-in value for the presence of an adenoma, and negative likelihood ratios were 0.34 and 0.20, respectively, suggesting limited rule-out capability when CT is negative.
• In the discordant cohort, sensitivity and specificity were 76.0% and 84.0% for localisation and 88.0% and 85.7% for lateralisation, respectively; however, in the negative cohort, sensitivity was lower (57.1% and 68.4%) with notably higher specificity (93.6% and 87.5%) for localisation and lateralisation, respectively.
• The sensitivity for lateralisation was not significantly different between the discordant and negative cohorts, with no significant differences observed in the sensitivity or specificity for localisation.

IN PRACTICE

"Multiphase CT provided reliable diagnostic accuracy for localisation and lateralisation of parathyroid adenomas in the context of both discordant and negative first-line imaging (Tc-MIBI SPECT/CT + US). Its high specificity and strong positive likelihood ratios make it a valuable rule-in tool that enables targeted surgical approaches," the authors wrote.

"In ~70% of the cohort, four-quadrant exploration was avoided following second line multiphase CT in favour of a more minimally invasive approach," they added.

SOURCE

The study was led by Jake E. Cowen, BMBS, Queen Alexandra Hospital, Portsmouth Hospitals University NHS Trust, Portsmouth, England. It was published online on August 12, 2026, in the British Journal of Radiology.

LIMITATIONS

The study was limited by relatively small sample sizes in subgroup analyses, a single-centre retrospective design with potential selection and verification biases, reliance on intraoperative documentation and radiologic interpretation for quadrant-level localisation, and an assumption of independence between observations in quadrant- and side-based analyses.

DISCLOSURES

This study did not receive any funding. The authors reported having no conflicts of interest.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.