Bone turnover prediction in patients with chronic kidney disease (CKD) undergoing hemodialysis using shortened dynamic 18F-NaF PET/CT Ki–Patlak

Abstract

Abstract This study investigated whether Ki–Patlak derived from a shortened scan time for dynamic 18F-NaF PET/CT in chronic kidney disease (CKD) patients undergoing hemodialysis can provide predictive accuracy comparable to that obtained from a longer scan. Twenty-seven patients on chronic hemodialysis, involving a total of 42 scans between December 2021 and August 2023 were recruited. Dynamic 18F-NaF PET/CT scans, lasting 60–90 min, were immediately acquired post-injection, covering the mid-twelfth thoracic vertebra to the pelvis region. Ki–Patlak analysis was performed on bone time–activity curves at 15, 30, 45, 60, and 90 min in the lumbar spine (L1–L4) and both anterior iliac crests. Spearman’s rank correlation (rs) and interclass correlation coefficient were used to assess the correlation and agreement of Ki–Patlak between shortened and standard scan times. Bone-specific alkaline phosphatase (BsAP) and tartrate-resistant acid phosphatase isoform 5b (TRAP5b) were tested for their correlation with individual Ki–Patlak. Strong correlations and good agreement were observed between Ki–Patlak values from shortened 30-min scans and longer 60–90-min scans in both lumbar spine (rs = 0.858, p < 0.001) and anterior iliac crest regions (rs = 0.850, p < 0.001). The correlation between BsAP and Ki–Patlak in the anterior iliac crests was weak and statistically insignificant. This finding suggests that a proposed shortened dynamic 18F-NaF PET/CT scan is effective in assessing bone metabolic flux in CKD patients undergoing hemodialysis, offering a non-invasive alternative approach for bone turnover prediction.

Keywords

• Chronic kidney disease
• Renal osteodystrophy
• Bone turnover prediction
• Metabolic bone disease
• 18F-NaF kinetic modeling
• Ki–Patlak