INTRODUCTION
Clinical trials of 31 P-magnetic resonance spectroscopy (MRS) have been conducted for various organs (1–4) 31 P-MRS has been widely used to study pretransplant renal viability (5–8) (9–12) (9,10) 31 P-MRS (13) 31 P-MRS to assess the long-term prognosis.
MATERIALS AND METHODS
We studied 20 patients with renal allografts. The age of the patients ranged from 16 to 42. Eleven kidneys were transplanted from living related donors and 9 from cadaveric donors. The mean duration from transplantation to MRS was 33.2 months (ranging from 1 to 121 months) (Table 1 ). Clinical data included serum creatinine level before MRS, at the time of MRS, and after MRS. Clinical course of renal function at MRS was evaluated from changes of serum creatinine levels. The immunosuppressive protocol in our patients consists of cyclosporine A, azathioprine, and a low-dose predonisolone.
Table 1: Case summaries
A Gyroscan ACS-II (Philips Medicals, Holland) operating at 1.5 Tesla and 15 cm surface coil was used for 31 P-MRS. Before 31 P-MRS, the magnetic field homogeneity was optimized by shimming on the proton signal. The shim-value was less than 5 Hz in each experiment. Localized 31 P-MRS was performed using the image selected in vivo spectroscopy (ISIS) method. Spectra with a repetition time of 2000 msec and 512 acquisitions were recorded (Fig. 1-a ). Figure 1-b shows a representative 31 P-MRS spectrum. Seven peaks were identified for fitting; the phosphomonoester peak (PME), the inorganic phosphate peak (Pi), the phosphodiester peak (PDE), the phosphocreatine peak (PCr), and the three peaks of adenosine triphosphate (γ-, α- and β-ATP). Individual peaks were fitted by Lorenzian line-shapes with a least square method (Fig. 2-a, 2-b ).
Figure 1: 31 P spectrum. (a) Magnetic resonance image of a transplanted kidney shows the position of the volume of interest (VOI) that was used to obtain the 31 P spectrum of Figure 1-b. (b) 31 P spectrum of a transplanted kidney. PME, phosphomonoesters; Pi, inorganic phosphate; PDE, phosphodiesters; PCr, phosphocreatine; ATP, adenosine triphosphate.
Figure 2: The line fitting of the 31 P spectrum of transplanted kidney. (a) The line-fitted spectrum (solid line) was obtained from the original (dotted line). (b) The line-fitted spectrum (solid line) was made assuming the presence of seven signals with a Lorenzian line shape (dotted line).
We reported on the relation of β-ATP/Pi with the clinical course of renal function when MRS was performed. One group with gradual increase in serum creatinine has lower β-ATP/Pi ratio than another group with stable course (13) 31 P-MRS peak ratios were also evaluated, including β-ATP/PME, β-ATP/PDE, PDE/PME, PDE/Pi, γ-ATP/Pi, γ-ATP/PME, γ-ATP/PDE, α-ATP/Pi, α-ATP/PME, and α-ATP/PDE.
Informed consent was obtained in all cases before the 31 P-MRS experiments, which the ethical committee of the National Institute of Radiological Sciences approved in advance.
RESULTS
The Survival of the Kidney after 31 P-MRS
Survival is shown in Figure 3 . It reflects patients with high β-ATP/Pi ratio (β-ATP/Pi≥1.2) and low β-ATP/Pi ratio (β-ATP/Pi<1.2). The high β-ATP/Pi group survived significantly longer than the low β-ATP/Pi group (Fig. 3 ). The horizontal axis indicates months since MRS study.
Figure 3: Survival rate of transplanted kidneys after MRS. High: β-ATP/Pi≥1.2 Low: β-ATP/Pi<1.2
Renal Survival Ratio at 3 Years and 5 Years
At 3 years after 31 P-MRS, the graft was still functional in all 13 subjects with a high β-ATP/Pi but in only one of 7 subjects with a low β-ATP/Pi. This difference was significant by chi-square test (P <o.oo1) (Table 2 ).
Table 2: Prognosis of high β-ATP/Pi group and low β-ATP/Pi group
At 5 years after 31 P-MRS, the graft was functional in 10 of 13 subjects with a high β-ATP/Pi and in one of 7 subjects with a low β-ATP/Pi. This difference was also significant by chi-square test (P <o.o1) (Table 2 ).
A β-ATP/Pi ratio >1.2 had sensitivity of 92.8%, specificity of 100%, and accuracy of 95% for predicting 3-year renal survival; a β-ATP/Pi ratio >1.2 had sensitivity of 90.9%, specificity of 66.7%, and accuracy of 76.9% for predicting 5-year renal survival.
Comparison of Kidneys Derived from Living Related Donors with Kidneys from Cadaveric Donors
Seven of 11 kidneys derived from living related donors and 6 of 9 from cadaveric donors survived 3 years. There were five cases in each kidney group in which the kidney survived for 5 years after 31 P-MRS. There was no significant difference between two groups (Table 3 ).
Table 3: Survival rate of living related donor group and cadaveric group after MRS
Cases in which 31 P-MRS Was Performed Within 3 Months after Transplantation
All three renal transplants with a β-ATP/Pi ratio >1.2 survived at 3 years. At 5 years one of them had failed when the ratio was >1.2.
Other Data
We compared 31 P-MRS spectra data between the survived group and the failed group (Fig. 4-a and 4-b ). Data are expressed as the mean±standard deviation. Open symbols represent the survived group and closed symbols represent the failed group. The survived group had significantly higher β-ATP/Pi, α-ATP/Pi, and PDE/Pi ratios than the failed group.
Figure 4: Peak ratios of 31 P spectrum of survived group and failed group (a) 3 years after MRS (b) 5 years after MRS SD, standard deviation. Data were analyzed by Wilcoxon’s test.
DISCUSSION
In the case of cadaveric kidney it takes a long time to select the recipient and then to transport the kidney. Consequently cold storage in a perfusion solution was used in the case of cadaveric kidney to protect against renal damage. It is important to evaluate renal viability before transplantation. Bretan et al. and Kurkova et al. reported the effectiveness of MRS for evaluating hypothermically stored cadaveric kidneys (5,6)
Chan et al. reported that the β-ATP value of hypothermically stored cadaveric kidneys is a good indicator of postplantation renal viability (7) (9,10) (13)
However, there had been no study of the efficacy of MRS for predicting the long-term survival of transplants. We followed the recipients for at least 5 years from when imaging was performed. When the cut-off level for β-ATP/Pi in the good renal function group was 1.2, the 3-year positive predictive value with respect to the renal survival was 100% and the 5-year positive predictive value was 76.9%. There were clear significant differences in both survival rates.
There have been several reports that a low β-ATP/Pi value was indicative of ischemia (14,16,17) (15) (14)
We discussed the β-ATP/Pi value as a parameter for predicting long-term survival of transplanted kidneys from when MRS was performed. A value >1.2 suggests a high probability of 3-year renal survival, whereas a value >2.5 indicates that the transplanted kidney could survive over 5 years. 31 P-MRS may be useful for predicting long-term survival of transplanted kidneys, but additional studies are needed.
REFERENCES
1. Naruse S, Tanaka C, Higuchi T, et al. Non-invasive quantitation of phosphorus metabolites in human brain and brain tumour by magnetic resonance spectroscopy. Jpn J Magn Reson Med 1989; 8: 243.
2. Bell JD, Bakhoo KK. Metabolic changes underlying
31 P MR spectral alterations in human hepatic tumors. NMR Biomed 1998; 11: 354.
3. Sijens PE. Phosphorus MR spectroscopy in the treatment of human extremity sarcomas. NMR Biomed 1998; 11: 341.
4. Illing B, Horn M, Urban B, et al. Changes of myocardial high-energy phosphates with the cardiac cycle during acute or chronic myocardial stress. Magn Reson Med 1998; 40: 727.
5. Bretan PN Jr, Baldwin N, Novick AC, et al. Pretransplant assessment of renal viability by phosphorus-31 magnetic resonance spectroscopy clinical experience in 40 recipient patients. Transplantation 1989; 48: 48.
6. Kurkova D, Herynek V, Gintelova J, Taborsky P, Hajek M. Potential of
31 P magnetic resonance spectroscopy in monitoring the viability of human renal grafts stored in Euro-Collins perfusion solution. Physiol Res 1995; 44: 327.
7. Chan L, French ME, Gadian DG, et al. Studies of human kidneys prior to transplantation by phosphorus nuclear magnetic resonance. In: Pegy, Halasz, Jacobsen, eds. Organ preservation, Vol 3. Lancaster: MTP Press, 1981.
8. Hene RJ, van der Grond J, Boer WH, Mali WP, Koomans HA. Pre-transplantation assessment of renal viability with
31 P magnetic resonance spectroscopy. Kidney Int 1994; 46: 1694.
9. Chan L, Shapiro JI. Magnetic resonance study of renal transplantation. Ren Physiol Biochem 1989; 12: 181.
10. Grist TM, Charles HC, Sostman HD. 1990 ARRS Executive Council Award. Renal transplant rejection: diagnosis with
31 P MR spectroscopy. Am J Roentgenol 1991; 156: 105.
11. Boska MD, Meyerhoff DJ, Twieg DB, Karczmar GS, Matson GB, Weiner MW. Image-guided
31 P magnetic resonance spectroscopy of normal and transplanted human kidneys. Kidney Int 1990; 38: 294.
12. Klemm A, Rzanny R, Funfstuck R, et al.
31 P-magnetic resonance spectroscopy (
31 P-MRS) of human allografts after renal transplantation. Nephrol Dial Transplant 1998; 13: 3147.
13. Seto K. Phosphorus-31 magnetic resonance spectroscopy for evaluation of renal transplants. Jpn J Magn Reson Med 1991; 11: 54.
14. Marshall V. Organ and tissue preservation for transplantation: monitoring by P-31 nuclear magnetic resonance. Transplant Proc 1985; 17: 1693.
15. Zollinger HU. Morphology and pathogenesis of glomerulopathy in cadaver kidney allografts treated with anti-lymphocyte globulin. Curr Top Pathol 1973; 57: 1.
16. Weiner MW, Adam WR. Magnetic resonance spectroscopy for evaluation of renal function. Semin Urol 1985; 3: 34.
17. Bretan PN Jr, Vigneron DB, Hricak H, et al. Assessment of in situ renal transplant viability by
31 P-MRS: experimental study in canines. Am Surg 1993; 59: 182.
