Radiology: Quantitative 3D MR aids diagnosis of kidney transplant dysfunction

Semiautomated segmentation process using coregistered 3D MR datasets. Final segmentation shows the cortex (red), medulla (blue), and the collecting system (green). Source: Radiology.

Low-dose MR renography may offer a noninvasive means to differentiate between acute kidney rejection and acute tubular necrosis after kidney transplantation, according to a study published in the September issue of Radiology.

Although renal transplant represents the optimal treatment for patients with chronic renal failure, 30 to 40 percent of transplant patients face at least one episode of acute graft dysfunction. While dysfunction related to anatomical causes can be diagnosed by ultrasound or CT, intrinsic parenchymal conditions—such as acute rejection and acute tubular necrosis, that can cause dysfunction—are more difficult to diagnose. Many of the symptoms, laboratory results and imaging findings of these conditions overlap, but treatments are dramatically different.

“Early characterization of the underlying cause of graft dysfunction is important, because delayed treatment can lead to irreversible loss of nephrons and hasten graft loss over time,” wrote Akira Yamamoto, MD, PhD, department of radiology at New York University School of Medicine in New York City, and colleagues.

The current gold standard for diagnosis is the percutaneous renal transplant biopsy, but the invasive procedure may result in complications and mask certain etiologies.

Given the emergence of MR renography as a measure of renal function, Yamamoto and colleagues sought to apply a multicompartment analytic model and evaluate the ability of the technique to help identify the cause of graft dysfunction.

The researchers prospectively enrolled 60 patients (41 men, 19 women; mean age, 49 years) with transplanted kidneys between December 2001 and May 2009. Thirty-one patients had clinically normal functioning transplants; 29 patients had acute dysfunction.

After patients underwent MR renography, images were registered and segmented to produce iliac artery, renal cortical and renal medullary signal intensity vs. time curves and volumes. Renal function was estimated using a modified three-compartment model that examines vascular, tubular and collecting ducts.

The model measured the following parameters: renal plasma flow (RPF), glomerular filtration rate (GFR), cortical vascular volume fraction, mean transit time for the vascular compartment (MTT_A), mean transit time for the tubular compartment (MTT_B), mean transit time for the collecting system compartment (MTT_C) and mean transit time for the whole kidney (MTT_K).

“As expected, the GFR of the transplanted kidney was significantly lower in the acute dysfunction group than in the normal function group,” the researchers wrote. Meanwhile, MTT_K was significantly higher in the acute dysfunction group than in the normal function group. However, there was considerable overlap with both measures.

The researchers also found significant differences in fractional MTT measures between the normal functioning and group and the dysfunction group. Logistic regression indicated that MTT_A/K offered the best prediction of acute rejection while MTT_T/K provided the best predictor of acute tubular necrosis.

The findings, noted the researchers, are in concordance with previously reported qualitative observations.

Yamamoto and colleagues acknowledged several shortcomings to the study, including its small sample size and the small number of subjects in each pathology group.

They suggested several avenues for future research including an assessment of whether a combination of diffusion-weighted and blood oxygen level-dependent MRI may provide complementary data and improve diagnosis.

Finally, Yamamoto and colleagues pointed out that the method adds no more than 10 minutes to scanning time and may decrease the need for percutaneous biopsy to differentiate the causes of graft dysfunction.