Left ventricular volume assessment by planar radionuclide ventriculography evaluated by MRI

Sergiy V. Lazarenko, Pieter A. van der Vleuten, Rene A. Tio, Antoon T. M. Willemsen, Anne M. J. Paans, Jose E. Douma, Clark J. Zeebregts, Rudi A. Dierckx, Felix Zijlstra, Riemer H. J. A. Slart*

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

1 Citation (Scopus)


Background Assessment of left ventricular (LV) ejection fraction (LVEF) and LV volume are essential for the evaluation of prognosis in cardiac disease. LVEF and LV volumes can be measured with several imaging modalities, such as magnetic resonance imaging (MRI) or computed tomography; however, these are relatively expensive and time consuming. In contrast, planar radionuclide ventriculography (PRV) for LVEF assessment is a cost-effective, fast, and reliable technique, but PRV for LV volumes calculation is less common.

Aim Evaluation of a new hybrid geometrical count-based method (HGCBM) in comparison with two count-based methods (CBMs) and a geometrical method (GM) for the calculation of LV volumes with PRV using MRI as reference.

Methods Thirty cardiac patients underwent routine PRV with a standard dose of 500 MBq of (99m)Tc-pertechnetate and additional cardiac MRI as reference method. LV volumes of PRV data were calculated by four different methods. The CBMs and GM are based on the assumption that the shape of the LV can be approximated by an ellipsoid or sphere, and the new HGCBM extracts the volume from the projected count rates themselves.

Results All methods underestimated the LV volumes as compared with the MRI-measured volumes. The difference (mean +/- SD) of end-diastolic volume (EDV) between PRV and MRI was 33 +/- 23 ml for GM, 12 +/- 26 ml for HGCBM, 50 +/- 38 ml for CBM1, and 13 +/- 40 ml for CBM2. The correlation coefficients for EDV between PRV methods and MRI were r = 0.90 for GM and r = 0.85 for HGCBM. The CBMs showed poor correlation r = 0.64 with the MRI data and a high SD. The difference of end-systolic volume (ESV) between PRV and MRI was 23 +/- 19 ml for GM, 9 +/- 22 ml for HGCBM, 29 +/- 29 ml for CBM1, and 9 +/- 28 ml for CBM2. The correlation coefficients for ESV between PRV methods and MRI were r = 0.955 for GM and r = 0.914 for HGCBM, r = 0.85 for CBM1 and CBM2. Although GM showed a slightly higher correlation than HGCBM, the difference of EDV and ESV between PRV and MRI was much higher for GM in comparison with HGCBM. Both CBMs showed poor agreement with MRI data.

Conclusion PRV using the new HGCMB method in comparison with other methods is an easy and accurate method to determine LV volumes. However, all methods underestimate ESV and EDV slightly as compared with MRI. Nucl Med Commun 30:727-735 (C) 2009 Wolters Kluwer Health | Lippincott Williams & Wilkins.

Original languageEnglish
Pages (from-to)727-735
Number of pages9
JournalNuclear Medicine Communications
Issue number9
Publication statusPublished - Sep-2009


  • count-based method
  • geometrical method
  • left ventricular volumes
  • magnetic resonance imaging
  • planar radionuclide ventriculography

Cite this