Metabolic imaging of in vivo myocardium

Background The equilibrium cellular water efflux rate constant [kio; mean water lifetime inverse] from contrast agent [CA]enhanced MRI measures on-going cellular Na,K-ATPase activity [turnover]. Good literature [4 different labs] agreement shows substantial kio decreases in myocardial ischemia, hypertension, or infarct regions (Table). The 3 methods used differ in extracellular ("outside”) CAo level manipulation to change the MR shutter-speed relative to kio and the MR exchange condition reached: A) CAo steady-state, slow-exchange-regime; B) CAo titration, fastexchange-regime [FXR]; and C) CAo wash-out, FXR. The independent intracellular volume fraction [ICV] cell density•volume product and ≈ 1 ECV [extracellular volume fraction] also decreases in pathology. We hypothesize that kio mapping shows metabolic compromise most effectively. We report initial experience with tissue near a repaired ventricular septal defect [VSD].


Background
The equilibrium cellular water efflux rate constant [k io ; mean water lifetime inverse] from contrast agent [CA]enhanced MRI measures on-going cellular Na + ,K + -ATPase activity [turnover]. Good literature [4 different labs] agreement shows substantial k io decreases in myocardial ischemia, hypertension, or infarct regions (Table). The 3 methods used differ in extracellular ("outside") CA o level manipulation to change the MR shutter-speed relative to k io and the MR exchange condition reached: A) CA o steady-state, slow-exchange-regime; B) CA o titration, fastexchange-regime [FXR]; and C) CA o wash-out, FXR. The independent intracellular volume fraction [ICV] -cell density• volume product and ≈ 1 -ECV [extracellular volume fraction] -also decreases in pathology. We hypothesize that k io mapping shows metabolic compromise most effectively. We report initial experience with tissue near a repaired ventricular septal defect [VSD].

Methods
We acquired serial 1.5T 1 H 2 O T 1 -weighted data from a 27 yo male before and 3 times after a bolus IV 0.15 mmol/kg CA [Omniscan] injection. Quantitative Look-Locker T 1 measurements [non-selective inversion, 21 recovery times] imaged an 8 mm slice with a mid-ventricular short axis location inferior to the VSD patch. Method C (CA o washout, FXR) determined k io and ICV values in six LV wall segments.

Results
The Figure shows a post-CA T 1 -w image: the endo-and epicardial LV wall edges as bright orange and green, respectively [light orange circle, an LV ROI]. Segmental ICV and k io values are given (yellow). Segments S5 and S6 comprise the septum. The ICV values for segments S1 -S4 are reasonable for normal myocardium (Table). Thus, we have indicated (*) a control myocardial k io value [5 s -1 , Table], since the CA wash-out data quantity [3 points] and quality from these normal myocardium segments yielded insufficient precision. Interestingly, the k io value is reduced [4.5 s -1 ] in segment S6, and dramatically so [1.7 s -1 ; 66%↓] in segment S5, immediately inferior to the VSD patch.

Conclusions
The k io biomarker is a sensitive measure of on-going myocardial metabolic activity. Our result suggests that tissue nearby a VSD patch can be, or become, metabolically compromised.
The ultimate goal is pixel-wise k io and ICV maps. [Here, nominal voxels are 2x2x8 mm 3 = 32 μL.] For this, one needs data with good S/N and more than 3 wash-out points. Also, method C has systematic error absent in methods A and B, which cannot be used for humans. It assumes the CA o concentration equals that of CA p [in plasma] during wash-out. This is invalid for finite CA intravasation kinetics, which may be particularly slow in myocardial lesions due to common reduced

• Research which is freely available for redistribution
Submit your manuscript at www.biomedcentral.com/submit Figure 2 Short axis T 1 -w image slice inferior to VSD patch. The k io and ICV values of six LV wall segments are given. k io and ICV are reduced (66% and 30%, respectively) in segment S5,immediately below the patch.