Dynamic molecular imaging of cardiac innervation using a dual headpinhole SPECT system

Jicun Hu, Rostyslav Boutchko, Arkadiusz Sitek, B.w. Reutter, R.h. Huesman, G.t. Gullberg2008 · DOI

TL;DR

This paper presents a novel dynamic dual-head pinhole SPECT imaging protocol for assessing cardiac sympathetic innervation in rats. By using a faster, dynamic acquisition and advanced analysis techniques, it aims to provide a more comprehensive understanding of heart failure progression in animal models, potentially reducing scan times compared to traditional methods.

What Problem Does This Paper Address?

Assessing the health of the sympathetic nervous system in the heart is crucial for understanding and diagnosing heart failure. Current methods using 123I-MIBG (a radiotracer that mimics norepinephrine) typically involve two separate scans: one for initial uptake and another 3-4 hours later to measure tracer washout. This two-step process is time-consuming and may not fully capture the dynamic changes occurring in the heart. Furthermore, the rapid initial uptake of the tracer can make it difficult to obtain accurate early data. This study addresses these limitations by developing a single, dynamic SPECT acquisition that captures both uptake and washout over a shorter period, aiming to provide more detailed kinetic information about cardiac innervation.

Methods

The core of this work lies in developing and validating a dynamic imaging protocol using a dual-head pinhole SPECT system. Key methodological aspects include:

  • System Calibration: A sophisticated calibration procedure was implemented to accurately model the dual-head pinhole SPECT system's geometry. This involved a nonlinear least squares fitting method to estimate seven geometric parameters, accounting for angle-dependent shifts caused by gravitational torque.
  • Dynamic Acquisition: A continuous 360° data acquisition was performed every 90 seconds for 90-120 minutes after injecting 123I-MIBG into rats. This dynamic approach allows for the capture of tracer kinetics over time.
  • Data Analysis:
    • Factor Analysis (FA): For the initial, inconsistent projections (due to the rapid tracer bolus), FA was applied directly to the projections to extract time-activity curves (TACs) for the blood pool, serving as an input function.
  • Region of Interest (ROI) Sampling: For later, more consistent projections, ROIs were manually drawn on reconstructed images to obtain TACs for the blood pool and myocardium.
  • Two-Compartment Modeling: The extracted TACs were then fitted to a two-compartment model to estimate kinetic parameters, specifically wash-in (K21) and washout (K12) rates, and to calculate distribution volumes.
  • Image Reconstruction: An iterative ML-EM algorithm was used for image reconstruction, incorporating a multi-ray model for the point spread function to account for the finite size of the pinhole and detector pixels.

Datasets

The study utilized data from two groups of rats: spontaneous hypertensive rats (SHR) and normotensive Wistar Kyoto (WKY) rats. Four studies were performed on each rat over an eight-month period. No new benchmark datasets were introduced; the focus was on developing and validating a new imaging protocol.

Key Insights and Ideas

  • Dynamic Imaging for Kinetic Insights: The shift from static to dynamic SPECT imaging for 123I-MIBG allows for a more detailed understanding of tracer kinetics, moving beyond simple uptake measurements to quantify wash-in and washout rates.
  • Factor Analysis for Early Data: Applying FA directly to early, inconsistent projections is a clever workaround to obtain a usable input function when traditional methods fail due to rapid tracer dynamics.
  • SHR as a Heart Failure Model: The study leverages the SHR model to investigate changes in cardiac sympathetic innervation, which is known to be altered in heart failure.
  • Potential for Reduced Scan Time: The dynamic protocol aims to replace the traditional delayed scan, potentially shortening the overall imaging time while providing richer information.

Key Results

  • Calibration Accuracy: The developed calibration procedure demonstrated good fits to measured projections, indicating accurate geometric parameter estimation for the dual-head pinhole SPECT system.
  • Kinetic Parameter Estimation: The two-compartment model successfully fitted the obtained TACs, allowing for the estimation of wash-in and washout parameters.
  • Washout Differences in Late Stages: Towards the end of the rats' lives, SHR exhibited a trend towards higher 123I-MIBG washout rates compared to WKY rats, which is consistent with a compromised heart. However, the authors note that these results were not statistically significant due to the small sample size.
  • Distribution Volume Differences: Hypertensive rats generally showed lower distribution volumes compared to WKY controls, with one SHR that died showing a somewhat better distribution volume than other SHRs.

Limitations and Future Work

  • Statistical Significance: The study acknowledges that the observed differences in washout rates were not statistically significant due to the small number of animals (4 rats total: 2 WKY, 2 SHR).
  • Non-Uniqueness in FA: The paper mentions the known problem of non-uniqueness in factor analysis and suggests using orthogonal constraints or more detectors to address this in future work.
  • Collimator Penetration: While the pinhole collimators were designed for energies up to 140 keV, 123I emits higher energy photons, leading to some penetration that could bias kinetic data. Future work could involve experimentally measuring and correcting for this.
  • Attenuation and Scatter: These effects were not compensated for in the current implementation but could be incorporated if X-ray CT images are available.
  • Future Studies: The authors propose repeating the study with more animals, investigating other tracers (e.g., positron emitters), and exploring the effects of exercise on heart failure. They also suggest exploring voxel-based projection schemes and incorporating experimentally measured blurring kernels.

Summary updated

Abstract

Typically 123I-MIBG is used for the study of innervation andfunction of the sympathetic nervous system in heart failure. The protocolinvolves two studies: first a planar or SPECT scan is performed tomeasure initial uptake of the tracer, followed some 3-4 hours later byanother study measuring the wash-out of the tracer from the heart. A fastwash-out is indicative of a compromised heart. In this work, a dual headpinhole SPECT system was used for imaging the distribution and kineticsof 123I-MIBG in the myocardium of spontaneous hypertensive rats (SHR) andnormotensive Wistar Kyoto (WKY) rats. The system geometry was calibratedbased on a nonlinear point projection fitting method using a three-pointsource phantom. The angle variation effect of the parameters was modeledwith a sinusoidal function. A dynamic acquisition was performed byinjecting 123I-MIBG into rats immediately after starting the dataacquisition. The detectors rotated continuously performing a 360o dataacquisition every 90 seconds. We applied the factor analysis (FA)methodand region of interest (ROI) sampling method to obtain time activitycurves (TACs)in the blood pool and myocardium and then appliedtwo-compartment modeling to estimate the kinetic parameters. Since theinitial injection bolus is too fast for obtaining a consistenttomographic data set in the first few minutes of the study, we appliedthe FA method directly to projections during the first rotation. Then thetime active curves for blood and myocardial tissue were obtained from ROIsampling. The method was applied to determine if there were differencesin the kinetics between SHR and WKY rats and requires less time byreplacing the delayed scan at 3-4 hours after injection with a dynamicacquisition over 90 to 120 minutes. The results of a faster washout and asmaller distribution volume of 123IMIBG near the end of life in the SHRmodel of hypertrophic cardiomyopthy may be indicative of a failing heartin late stages of heart failure.
BibTeX
@misc{hu2008dynamic,
  author = {Jicun Hu and Rostyslav Boutchko and Arkadiusz Sitek and B.W. Reutter and R.H. Huesman and G.T. Gullberg},
  title = {{Dynamic molecular imaging of cardiac innervation using a dual headpinhole SPECT system}},
  year = {2008},
  doi = {10.2172/928712},
}
Figure