Smart-CKD, an ultrasound-derived, computer-aided diagnosis (CAD) tool, demonstrated an area under the curve of 81 percent and an 83 percent sensitivity rate in a validation cohort for differentiating between mild and moderate to severe fibrosis in patients with chronic kidney disease.
Preliminary results for an emerging ultrasound-based diagnostic tool may offer promise for facilitating timely risk stratification and appropriate interventions in the management of chronic kidney disease (CKD).
For the prospective study, recently published in Academic Radiology, researchers developed the Smart-CKD (S-CKD) diagnostic tool, which incorporates age, renal length, and end-diastolic velocity (EDR), to aid in the differentiation of mild renal fibrosis and moderate-severe fibrosis with CKD. They subsequently assessed the model in a 114-patient training cohort and a 48-patient validation cohort comprised of patients with CKD undergoing renal biopsies.
In the training set, the S-CKD model had an area under the curve (AUC) of 84 percent, a sensitivity rate of 83 percent, an accuracy rate of 76 percent and a specificity rate of 69 percent, according to the study. The study authors found that the S-CKD model retained its diagnostic ability in the validation set with similar AUC (81 percent), sensitivity (86 percent), accuracy (79 percent) and specificity (70 percent).
The researchers also found that the risk of developing moderate-severe renal fibrosis for people with CKD increased by 6 percent for each additional year of age, 58 percent for each one-unit reduction in renal length and 20 percent for each one-unit reduction in EDV.
“It is imperative to underline that persistently and dynamically risk stratifying CKD patients based on their own condition changes so as to timely identify those at high risk of developing moderate-severe renal fibrosis is the critical point for optimal management,” wrote study co-author Zhongzhen Su, M.D., Ph.D., who is affiliated with the Department of Ultrasound at the Fifth Affiliated Hospital of Sun Yat-sen University in Zhuhai, China.
(Editor’s note: For related content, see “Can Ultrasound Renal Denervation be a Game-Changer for Reducing Blood Pressure?”)
The study authors emphasized that the three variables measured with the S-CKD tool are non-invasive, relatively inexpensive, and easy to obtain from routine imaging and a patient’s medical record. The ultrasound-derived measurements of renal length and EDV are key to the ongoing assessment of these high-risk patients, according to the researchers.
“Changes in renal microvascular perfusion are the principal processes underlying CKD progression as well as renal fibrosis deterioration,” noted Su and colleagues. “During the pathological process of renal fibrosis, renal microvascular changes such as intimal proliferation and medial thickening result in the reduction of luminal diameter, which in turn causes increased resistance to flow and a decline in renal perfusion.”
In regard to study limitations, the authors acknowledged the data coming from a small patient cohort at a single institution. Su and colleagues emphasized that future research should explore the utility of the S-CKD model at a multicenter level and assess the impact of ethnic differences upon the model’s performance.