McCulloch, Andrew C.
The stress radionuclide assessment of diastolic function.
MD thesis, University of Glasgow.
Full text available as:
Many patients are referred from primary care with suspected heart failure and are found to have preserved systolic function. These patients may be labelled as having normal ejection fraction or diastolic heart failure, the diagnosis of which is both controversial and difficult. Previous work has identified a large proportion of these patients to have an alternative, pre- existing diagnosis. This thesis prospectively assesses the prevalence of undiagnosed ischaemic heart disease and respiratory disease in this patient group and assess diastolic function using multiple methods. The central hypothesis being tested was that first third fractional filling, a radionuclide ventriculogram (RNVG) parameter previously used to assess diastolic function at rest, would identify diastolic dysfunction more accurately under stress conditions.
Patients were recruited from an open access echocardiography service. Echocardiography, including tissue Doppler assessment, was carried out independently by 2 experienced observers. Confounding diagnoses including coronary artery disease and respiratory disease were actively sought by myocardial perfusion imaging and spirometry. N-terminal proBNP was measured. List mode radionuclide ventriculography was performed at rest supine and during upright bicycle exercise with simultaneous measurement of VO2 max.
Validation of the reliability and reproducibility of first third fractional filling, peak filling rate, time to peak filling and other radionuclide parameters of systolic and diastolic function was undertaken. This demonstrated that it was possible to measure both first third fractional filling and peak filling rate with the short acquisition times necessary for assessment during stress. Time to peak filling was poorly reproducible under these conditions.
A normal range for first third fractional filling at rest and during exercise was established. Due to a strong inverse relationship between heart rate and first third fractional filling, a continuous reference range was constructed using an exponential model. This unique approach enables the calculation of the lower limit of normal at any heart rate. A more conventional mean ± 2 standard deviations was used for the other RNVG parameters.
Eighty three patients were recruited and completed an extensive multi-modality assessment of systolic and diastolic function. As with previous work in this field, the patients were predominantly female (82%) and elderly (mean age 66.7). Mild left ventricular systolic dysfunction as determined by RNVG was missed by echocardiography in one third of patients. Systolic dysfunction more significant than this was not observed. N-terminal proBNP was elevated in 21 of 82 patients where it was available with no significant difference in left ventricular ejection fraction between those with normal and elevated levels. Myocardial perfusion scanning was normal in 46 of 83 patients and showed significant ischaemia in 20 of 83. Spirometry was normal in 58 of 82 patients, with mild airflow obstruction in 20 patients and moderate obstruction in 4. In only one patient were no alternative diagnoses present.
There was poor correlation between indices of diastolic function at rest including first third fractional filling, echocardiographic parameters and NT-proBNP. The assessment of diastolic function using stress radionuclide ventriculography did not improve the correlation between measured indices. On stress, however, low first third fractional filling predicted exercise intolerance as an inability to reach anaerobic threshold.
Alternative diagnoses to diastolic dysfunction are present almost universally in patients with suspected normal ejection fraction heart failure. This is true even where these diagnoses are not previously established. This thesis underlines the need to fully assess this patient group to allow appropriate targeting of therapy. It is also clear that echocardiography alone is potentially misleading and it is suggested that it is better placed within a tiered assessment process.
The assessment of diastolic function using stress radionuclide ventriculography, although an appealing concept, does not improve diagnostic accuracy within this patient group. The marked heterogeneity of this patient group is likely to have played a role in this and it may be of interest to reassess stress radionuclide ventriculography in a more acute heart failure population.
Actions (login required)