Scholarlink

Purpose <p>Dual-phase amyloid positron emission tomography (PET) combines early-phase perfusion assessment with late-phase amyloid deposition evaluation, but the optimal timing for early-phase acquisition remains debated. This study aimed to compare semiquantitative and qualitative image metrics between 5-minute and 10-minute early-phase acquisitions using a within-patient, within-acquisition design.</p> Methods <p>Nine patients underwent dual-phase [¹⁸F]-flutemetamol PET/CT with list-mode acquisition, allowing retrospective reconstruction of both timeframes from the same dataset. Semiquantitative analysis included regional absolute standardised uptake value mean (SUVmean), cerebellar-normalised standardised uptake value ratio (SUVr), and Z-scores calculated from cerebellar-normalised SUVr values using a fluorodeoxyglucose normative database as reference, across 6 clinically relevant brain regions analysed bilaterally. Image quality assessment included signal-to-noise ratio (SNR) and coefficient of variation (CV). In borderline cases, two independent nuclear medicine readers performed a blinded qualitative concordance evaluation.</p> Results <p>SUVmean was higher at 10&#xa0;min (+ 3–5%), reflecting continued tracer uptake, with excellent proportional consistency (ICC &gt; 0.97). SUVr showed no significant differences in any region (all <i>p</i> &gt; 0.05), confirming complete compensation by cerebellar normalisation. Z-scores were more negative at 5&#xa0;min in 4 out of 6 regions, though ICC remained near-perfect (≥ 0.98), indicating full preservation of hypoperfusion patterns. Image quality metrics were statistically consistent between acquisitions (all False Discovery Rate (FDR)-corrected <i>p</i> &gt; 0.05). The Linear Mixed-Effects Model (LMM) analysis yielded consistent results, confirming that the effect of acquisition duration was small in magnitude and uniform. Z-score interpretation was identical in 93.8% of regions; in the 10 borderline cases, inter-reader concordance was 80%.</p> Conclusion <p>The 5-minute early-phase protocol provided non-inferior diagnostic information to the 10-minute standard when normalised metrics are employed. Image quality was fully preserved. The 5-minute acquisition is therefore recommended for routine clinical practice, improving patient comfort and workflow efficiency without compromising quantitative and qualitative information.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Optimal acquisition timing for perfusion imaging in dual-phase amyloid-PET: a within-patient list-mode comparison of 5-minute versus 10-minute protocols

  • 期刊论文

  • 发表日期：

    2026年8月3日

查看全文
  • Anna Margherita Maffione,
  • Alice Ferretti,
  • Paola Bartoletti,
  • Lara Pavan,
  • Arianna Massaro,
  • Sotirios Chondrogiannis,
  • Andrea Osele,
  • Marcello Gava,
  • Maria Cristina Marzola

摘要

Purpose

Dual-phase amyloid positron emission tomography (PET) combines early-phase perfusion assessment with late-phase amyloid deposition evaluation, but the optimal timing for early-phase acquisition remains debated. This study aimed to compare semiquantitative and qualitative image metrics between 5-minute and 10-minute early-phase acquisitions using a within-patient, within-acquisition design.

Methods

Nine patients underwent dual-phase [¹⁸F]-flutemetamol PET/CT with list-mode acquisition, allowing retrospective reconstruction of both timeframes from the same dataset. Semiquantitative analysis included regional absolute standardised uptake value mean (SUVmean), cerebellar-normalised standardised uptake value ratio (SUVr), and Z-scores calculated from cerebellar-normalised SUVr values using a fluorodeoxyglucose normative database as reference, across 6 clinically relevant brain regions analysed bilaterally. Image quality assessment included signal-to-noise ratio (SNR) and coefficient of variation (CV). In borderline cases, two independent nuclear medicine readers performed a blinded qualitative concordance evaluation.

Results

SUVmean was higher at 10 min (+ 3–5%), reflecting continued tracer uptake, with excellent proportional consistency (ICC > 0.97). SUVr showed no significant differences in any region (all p > 0.05), confirming complete compensation by cerebellar normalisation. Z-scores were more negative at 5 min in 4 out of 6 regions, though ICC remained near-perfect (≥ 0.98), indicating full preservation of hypoperfusion patterns. Image quality metrics were statistically consistent between acquisitions (all False Discovery Rate (FDR)-corrected p > 0.05). The Linear Mixed-Effects Model (LMM) analysis yielded consistent results, confirming that the effect of acquisition duration was small in magnitude and uniform. Z-score interpretation was identical in 93.8% of regions; in the 10 borderline cases, inter-reader concordance was 80%.

Conclusion

The 5-minute early-phase protocol provided non-inferior diagnostic information to the 10-minute standard when normalised metrics are employed. Image quality was fully preserved. The 5-minute acquisition is therefore recommended for routine clinical practice, improving patient comfort and workflow efficiency without compromising quantitative and qualitative information.