Health Technology Assessment (Mar 2022)

Dynamic contrast-enhanced CT compared with positron emission tomography CT to characterise solitary pulmonary nodules: the SPUtNIk diagnostic accuracy study and economic modelling

  • Fiona J Gilbert,
  • Scott Harris,
  • Kenneth A Miles,
  • Jonathan R Weir-McCall,
  • Nagmi R Qureshi,
  • Robert C Rintoul,
  • Sabina Dizdarevic,
  • Lucy Pike,
  • Donald Sinclair,
  • Andrew Shah,
  • Rosemary Eaton,
  • Andrew Clegg,
  • Valerio Benedetto,
  • James E Hill,
  • Andrew Cook,
  • Dimitrios Tzelis,
  • Luke Vale,
  • Lucy Brindle,
  • Jackie Madden,
  • Kelly Cozens,
  • Louisa A Little,
  • Kathrin Eichhorst,
  • Patricia Moate,
  • Chris McClement,
  • Charles Peebles,
  • Anindo Banerjee,
  • Sai Han,
  • Fat Wui Poon,
  • Ashley M Groves,
  • Lutfi Kurban,
  • Anthony J Frew,
  • Matthew E Callister,
  • Philip Crosbie,
  • Fergus V Gleeson,
  • Kavitasagary Karunasaagarar,
  • Osei Kankam,
  • Steve George

DOI
https://doi.org/10.3310/WCEI8321
Journal volume & issue
Vol. 26, no. 17

Abstract

Read online

Background: Current pathways recommend positron emission tomography–computerised tomography for the characterisation of solitary pulmonary nodules. Dynamic contrast-enhanced computerised tomography may be a more cost-effective approach. Objectives: To determine the diagnostic performances of dynamic contrast-enhanced computerised tomography and positron emission tomography–computerised tomography in the NHS for solitary pulmonary nodules. Systematic reviews and a health economic evaluation contributed to the decision-analytic modelling to assess the likely costs and health outcomes resulting from incorporation of dynamic contrast-enhanced computerised tomography into management strategies. Design: Multicentre comparative accuracy trial. Setting: Secondary or tertiary outpatient settings at 16 hospitals in the UK. Participants: Participants with solitary pulmonary nodules of ≥ 8 mm and of ≤ 30 mm in size with no malignancy in the previous 2 years were included. Interventions: Baseline positron emission tomography–computerised tomography and dynamic contrast-enhanced computer tomography with 2 years’ follow-up. Main outcome measures: Primary outcome measures were sensitivity, specificity and diagnostic accuracy for positron emission tomography–computerised tomography and dynamic contrast-enhanced computerised tomography. Incremental cost-effectiveness ratios compared management strategies that used dynamic contrast-enhanced computerised tomography with management strategies that did not use dynamic contrast-enhanced computerised tomography. Results: A total of 380 patients were recruited (median age 69 years). Of 312 patients with matched dynamic contrast-enhanced computer tomography and positron emission tomography–computerised tomography examinations, 191 (61%) were cancer patients. The sensitivity, specificity and diagnostic accuracy for positron emission tomography–computerised tomography and dynamic contrast-enhanced computer tomography were 72.8% (95% confidence interval 66.1% to 78.6%), 81.8% (95% confidence interval 74.0% to 87.7%), 76.3% (95% confidence interval 71.3% to 80.7%) and 95.3% (95% confidence interval 91.3% to 97.5%), 29.8% (95% confidence interval 22.3% to 38.4%) and 69.9% (95% confidence interval 64.6% to 74.7%), respectively. Exploratory modelling showed that maximum standardised uptake values had the best diagnostic accuracy, with an area under the curve of 0.87, which increased to 0.90 if combined with dynamic contrast-enhanced computerised tomography peak enhancement. The economic analysis showed that, over 24 months, dynamic contrast-enhanced computerised tomography was less costly (£3305, 95% confidence interval £2952 to £3746) than positron emission tomography–computerised tomography (£4013, 95% confidence interval £3673 to £4498) or a strategy combining the two tests (£4058, 95% confidence interval £3702 to £4547). Positron emission tomography–computerised tomography led to more patients with malignant nodules being correctly managed, 0.44 on average (95% confidence interval 0.39 to 0.49), compared with 0.40 (95% confidence interval 0.35 to 0.45); using both tests further increased this (0.47, 95% confidence interval 0.42 to 0.51). Limitations: The high prevalence of malignancy in nodules observed in this trial, compared with that observed in nodules identified within screening programmes, limits the generalisation of the current results to nodules identified by screening. Conclusions: Findings from this research indicate that positron emission tomography–computerised tomography is more accurate than dynamic contrast-enhanced computerised tomography for the characterisation of solitary pulmonary nodules. A combination of maximum standardised uptake value and peak enhancement had the highest accuracy with a small increase in costs. Findings from this research also indicate that a combined positron emission tomography–dynamic contrast-enhanced computerised tomography approach with a slightly higher willingness to pay to avoid missing small cancers or to avoid a ‘watch and wait’ policy may be an approach to consider. Future work: Integration of the dynamic contrast-enhanced component into the positron emission tomography–computerised tomography examination and the feasibility of dynamic contrast-enhanced computerised tomography at lung screening for the characterisation of solitary pulmonary nodules should be explored, together with a lower radiation dose protocol. Study registration: This study is registered as PROSPERO CRD42018112215 and CRD42019124299, and the trial is registered as ISRCTN30784948 and ClinicalTrials.gov NCT02013063. Funding: This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 26, No. 17. See the NIHR Journals Library website for further project information.

Keywords