Study Casts Doubt on Renal Insufficiency Risk Predictors

Jim Kling

May 03, 2012

May 3, 2012 (Vancouver, British Columbia) — Advanced age and diabetes are generally considered to be predictors of renal insufficiency, but a new study has found that history of renal disease, hypertension, heart disease, and sex are better predictors.

The study, which focused on patients receiving a contrast agent in advance of contrast-enhanced computed tomography (CT) scans, was presented here at the American Roentgen Ray Society 2012 Annual Meeting.

Contrast agent can cause contrast-induced nephropathy. Instead of testing a patient's estimated glomerular filtration rate (eGFR) to determine kidney function, hospitals typically screen patients on the basis of age (70 years or older) and the presence of diabetes.

At the Northwestern University Medical Center in Chicago, Illinois, eGFR is routinely tested in patients before they undergo contrast-enhanced CT. This gave researchers the the opportunity to compare eGFR with typical questionnaire risk factors and other patient characteristics.

"We don't really need the questionnaire because we know what the truth is, but [it] is a good lead for other institutions to risk-stratify their patients," Vahid Yaghmai, MD, who presented the research, told Medscape Medical News.

The study involved 2404 consecutive outpatients who were scheduled for contrast-enhanced CT. The eGFR was determined for all patients immediately prior to the scan. Patients were asked about their history of renal disease, hypertension, heart disease, liver disease, diabetes mellitus, their age, their sex, and their use of metformin.

The researchers then created a binary logistic regression model using questions with significant univariate correlation for an eGFR below 60 mL/min per 1.73 m² and below 45 mL/min per 1.73 m².

They found that 648 of the 2404 patients (27.0%) had an eGFR below 60 mL/min per 1.73 m², and 174 (7.2%) had an eGFR below 45 mL/min per 1.73 m². There was no correlation between renal insufficiency and metformin use or history of liver disease (P > .05).

Patients with renal disease had an odds ratio (OR) for eGFR below 60 mL/min per 1.73 m² of 3.95 (P < .001). Other factors that affected risk for eGFR below 60 mL/min per 1.73 m² included hypertension (OR, 2.39; P < .001), heart disease (OR, 1.46; P = .004), and being male (OR, 0.70; P < .001). No statistically significant relationship was seen for diabetes mellitus (OR, 1.16; P = .307) or age over 70 years (OR, 1.22; P = .089).

Risk factors for eGFR below 45 mL/min per 1.73 m² included renal disease (OR, 6.86; P < .001), hypertension (OR, 2.52; P < .001), heart disease (OR, 1.78; P = .004), and being male (OR, 0.58; P = .002). No statistically significant effects were seen for diabetes mellitus (OR, 1.48; P = .072) or age over 70 years (OR, 0.89; P = .588).

For eGFR below 60 mL/min per 1.73 m² and eGFR below 45 mL/min per 1.73 m², the researchers calculated positive predictive values for renal disease (59.7% and 32.3%, respectively), hypertension (39.2% and 12.7%), heart disease (39.0% and 15.1%), and being male (24.1% and 5.9%).

"The results were surprising because the risk factors used routinely in clinical practice to screen for renal insufficiency — mainly age and diabetes — were not independent predictors. Many of the patients being excluded [from contrast-enhanced CT] based on age should be tested, because we know from [our study] that patients with abnormal eGFR are the ones less than 70 years of age, not the ones over 70. Our whole concept of risk stratification is not accurate," said Dr. Yaghmai.

The study is the latest example of evidence-based medicine altering the clinical landscape — something that has been relatively late on the scene for diagnostics and radiology, according to Stefan Puig, MD, associate professor of radiology at Bern University Hospital in Switzerland, who comoderated the session.

The other comoderator, Jeffrey Dunkle, MD, emergency radiology section chief at the Indiana University School of Medicine in Indianapolis, agreed. "There's so much in medicine that we do for historical reasons, or because it seems to makes sense. Now we're really, in the last 10 to 15 years, taking a magnifying glass to all the things we do," he told Medscape Medical News.

Dr. Yaghmai, Dr. Puig, and Dr. Dunkle have disclosed no relevant financial relationships.

American Roentgen Ray Society (ARRS) 2012 Annual Meeting: Abstract 054. Presented May 1, 2012.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: