Fluid assessment in dialysis patients by point-of-care magnetic relaxometry

Abstract

Magnetic resonance imaging (MRI) is a powerful diagnostic tool, but its use is restricted to the scanner suite. Here, we demonstrate that a bedside nuclear magnetic resonance (NMR) sensor can assess fluid status changes in individuals at a fraction of the time and cost compared to MRI. Our study recruited patients with end-stage renal disease (ESRD) who were regularly receiving hemodialysis treatments with intradialytic fluid removal as a model of volume overload and healthy controls as a model of euvolemia. Quantitative T ₂ measurements of the lower leg of patients with ESRD immediately before and after dialysis were compared to those of euvolemic healthy controls using both a 0.28-T bedside single-voxel NMR sensor and a 1.5-T clinical MRI scanner. In the MRI data, we found that the first sign of fluid overload was an expanded muscle extracellular fluid (ECF) space, a finding undetectable at this stage using physical exam. A decrease in muscle ECF upon fluid removal was similarly detectable with both the bedside sensor and MRI. Bioimpedance measurements performed comparably to the bedside NMR sensor but were generally worse than MRI. These findings suggest that bedside NMR may be a useful method to identify fluid overload early in patients with ESRD and potentially other hypervolemic patient populations.

Figure 1.. MRI pixel-wise analysis of changes within ROIs.

(A) ROIs were drawn on each slice of each scan for all subjects. Subcutaneous Tissue: includes skin, fat and blood vessels in the fat. Bone and Marrow: include tibia and fibula. Muscular Tissue: includes muscle, fascia, nerves, and blood vessels. Whole Leg: includes all tissues. (B) ROIs of sub-muscles were drawn on the first slice of each scan. (C) A histogram of the pixel-wise short (T_S) and long (T_L) relaxation values found in the muscular and subcutaneous tissue of a representative subject. The pre-post change in (D) T_S, (E) T_L, (F) R_L for each ROI across all HC and HD subjects. Bars represent the mean ± SD. ns denotes p ≥ 0.05, * for p < 0.05, ** for p < 0.01.

Figure 2.. MRI pixel-wise relaxation and relative amplitudes.

(A) CDFs of the pixel-wise T_L values found within the entire leg at baseline. The mean and 95% confidence interval (CI) of all subjects is in grey. (B) Summary of the proBNP and clinical examination results for HD subjects. (C) Heatmaps of T_S and T_L for a sample healthy control, HD1, and HD2b. Perifascial fluid deposits and/or subcutaneous edema are indicated by arrows. Average CDF of the pixel-wise R_L in the muscle for HC and HD subjects (D) pre- and (E) post-time points. (F) The change in R_L for HC and HD subject groups. HC2 and HC6 not included in HC average for figure 4F. All cdf figures are plotted as mean ± 95% CI.

Figure 3.. MRI muscle ROI bi-exponential fit results.

(A) The R_L values of the muscle ROI for each subject. (B) Pre- and post- muscle R_L values of HC and HD groups. (C) The change in muscle R_L for HC and HD subjects. The central mark in each boxplot indicates the median, and the bottom and top edges of the box indicate the 25th and 75th percentiles, respectively. The whiskers extend to the most extreme values not considered outliers. * denotes p < 0.05, ** denotes p < 0.01.

Figure 4.. Portable single-voxel NMR sensor for bedside measurements.

(A) Photo of the complete NMR sensor, RF coil, and box containing the matching circuitry for the coil. A US Quarter is used for scale. (B) Photo of the NMR sensor with the top and side removed. Arrows denote magnetic orientation of each slab. Red ellipsoid above sensor denotes approximate sensor measurement region. (C) Schematic of linear Halbach design showing magnetization orientation of the individual magnets as well as the net B_o and B₁ orientations. (D) Photo of the NMR sensor in use at the hospital for bedside assessment. (E) Sagittal and transverse MRI scans showing the location of the NMR sensor’s approximately measurement voxel in red.

Figure 5.. Comparison of T2 relaxation times from MRI pixel-by-pixel and NMR sensor.

Both mono- and bi-exponential fit results of each of the six phantoms and ex-vivo tissue samples. There is a strong correlation between the MRI and NMR sensor values (r²=0.966) suggesting that the results can be translated between the two sensors. Vertical (NMR sensor) error bars represent the 95% confidence interval for the fit. Horizontal (MRI) error bars represent the standard deviation of the pixel-by-pixel MRI results.

Figure 6.. NMR sensor results and future design criteria.

(A) Boxplots displaying R₂ values at pre- and post- time points, and (B) the change in R₂ for HC and HD subjects. The central mark in each box plot indicates the median, and the bottom and top edges of the box indicate the 25th and 75th percentiles, respectively. The whiskers extend to the most extreme values not considered outliers. (C) R₃ plotted against subcutaneous tissue thickness (r²=0.672). Note that the tissue thickness is compressed by a few millimeter when the leg is pressed against the NMR sensor for measurements. (D) Change in R₂ plotted against calf bioimpedance’s change in ECF-associated resistivity, R_e (r²=0.477). (E) MRI scan showing the size and location of some of the smaller ROIs. (F) Summary of P values comparing HC and HD subjects for each small ROI (detailed statistics in tables S19-S22). P-values are calculated by a two-sample permutation test with Monte Carlo estimation using 10⁵-1 repetitions. * signifies p < 0.05.

Figure 7.. Bioimpedance Comparison of HC and HD Subjects.

Panels (A)-(H) represent raw resistivity measurements whereas (I)-(P) represent TBW and ECF bioimpedance measurements. Panels (A)-(D) come from whole body bioimpedance measurements whereas panels (E)-(H) come from segmental leg bioimpedance measurements. The top row of panels shows R_e data, which corresponds to ECF. The second row of panels shows R_inf data, which corresponds to TBW. Fluid has a low resistivity. Low resistivity indicates more fluid. Higher resistivity indicates less fluid. An increase in resistivity indicates decrease of fluid. It is only possible to distinguish HD from HC subjects at a single time point with a whole body R_e measurement at baseline (A). Panels (F,H) show that it is possible to distinguish HD from HC subjects based on the change in R_e and R_inf in the leg. Panels (I)-(L) come from whole body bioimpedance measurements whereas panels (M)-(P) come from segmental leg bioimpedance measurements. The third row of panels shows ECF data. The bottom row of panels shows TBW data. It is only possible to significantly distinguish HC from HD subjects based on the change in whole-body (p=0.027) or leg (p=0.014 with permutation test; p=0.054 with Welch test) ECF. n.s. indicates p>0.05, * indicates p<0.05, ** indicates p<0.01.

Figure 8.. MR relaxometry findings at different fluid states.

Graphical summary of the relaxometry findings – through both traditional MRI and portable NMR sensor – at different clinical fluid states. All findings were observed in the muscular tissue.