From January 2007 to December 2013, we enrolled 556 APN patients who underwent a CE-CT scan within 24 hours after hospital admission before starting antibiotic treatment. Patients were excluded if they were younger than 18 years or had an hydronephrosis or renal abscess on CT scan. We also excluded the patients with post-contrast induced-acute kidney injury (CI-AKI) or low-quality image due to various causes. A clinical diagnosis of APN was made in patients with more than three of the following five diagnostic criteria: (1) clinical symptoms such as fever, chills, vomiting, or flank pain (2) costovertebral angle tenderness (3) fever of higher than 37.5 °C (4) leukocytosis in the complete blood count (5) abnormal urine test results (pyuria: white blood cell of ≥ 10/high-power field or positive urine culture of 10⁵ colony-forming unit/mL). A total of 537 patients were included in this study. Two experienced radiologist divided these patients into two groups by approximately quantifying renal parenchymal low density extent on CE-CT. APN extent was approximately quantified on CE-CT by calculating number of slices that contained less enhancing lesion divided by total number of slices that contained normal renal parenchyma (Fig. 1). We also checked the status of renal parenchymal involvement. In cases with bilateral APN of different grades in each kidney, the higher grade was recorded. Patients were divded into two groups: greater group (greater than 50% involvement, n = 143) and lesser group (lesser than 50% involvement, n = 394). This study was approved by the Institutional Review Board of the Presbyterian Medical Center, Jeonju, South Korea (IRB No. 2019-05-014).

All patients had a detailed clinical history and examination, a standard set of investigations including complete blood counts, liver function tests, serum creatinine, urea, electrolytes, chest radiograph, three peripheral blood smears for malaria, urinalysis, and two blood cultures. AKI was defined based on the RIFLE (Risk, Injury, Failure, Loss of kidney function, and End-stage kidney disease) criteria,9 and patients were categorized into the R, I or F categories. We defined CI-AKI with the RIFLE criteria as a relative increment in serum creatinine of ≥ 50%, or a decrease in estimated glomerular filtration rate of ≤ 25% from baseline, or an episode of oliguria lasting ≥ 6 hours within 48 to 72 hours following contrast administration.10 The estimated glomerular filtration rate (eGFR) was calculated using the abbreviated Modification of Diet in Renal Disease (MDRD) equation.11 When the baseline serum creatinine was not available, it was calculated using the standard four-variable MDRD formula assuming eGFR of 75 mL/min/1.73 m². The RIFLE class was determined based on the worst among serum creatinine levels, eGFR, and urine output criteria. Renal replacement therapy was initiated using the standard indications. Sepsis is defined as the presence (probable or documented) of infection together with systemic manifestations of renal infection. The CT images were reviewed by two radiologists who unaware of clinical information; the final decision of grade was reached by consensus.

CT scan was conducted on a 16-channel multi-detector row scanner (Somatom Sensation 16, Siemens Medical Solutions, Erlangen, Germany) with setting of 3 mm slice thickness. The CT findings were evaluated in the neprographic phases. Intravenous contrast material was administrated in an antecubital vein with injector at a dose of 1.5 mL/Kg body weight at a rate of 3 mL/s to a maximum of 120 mL. Nephrographic-phase scans were started 90 to 100 seconds after contrast injection.

All data are presented as means ± standard deviations unless otherwise specified. The baseline characteristics of patients in the greater and lesser groups were compared using t tests for continuous variables and chi-square tests for categorical variables. Clinically relevant parameters or the variables that were significantly associated with the presence of AKI in the univariate analysis were included in the multivariate analysis. A P-value of < 0.05 was considered to be statistically significant. All statistical analyses were carried out using SPSS version 22.0.

The baseline characteristics of the 537 study subjects are presented in Table 1. The patients included 504 (93.9%) men and 33 (6.1%) women, with a mean age of 55.2 ± 17.9 years. The mean estimated glomerular filtration rate was 70.6 ± 25.5 mL/min/1.73m². The initial white blood cell count (WBC) and serum albumin levels were 12.7 × 10³/mL and 3.8 g/dL, respectively. The mean hospital stay was 9.2 ± 4.6 days. Bilateral renal involvement in CE-CT and bacteremia were noted in 18.2% and 29.4% of the patients, respectively. Of 537 participants, 78 (14.5%) experienced AKI during hospitalization period.

When we compared clinical characteristics between greater (n = 143) and lesser (n = 394) groups, the patients in greater group had lower serum albumin levels (3.5 ± 0.5 vs 3.8 ± 0.6, P < 0.01) and longer hospital stay (10.1 ± 4.7 vs 8.8 ± 4.5, P < 0.05). Furthermore, bilateral renal involvement (40.6% vs 10.2%, P < 0.05) and bacteremia (36.4% vs 26.9%, P < 0.05) were frequently developed in greater group, respectively (Table 2). AKI (23.1% vs 11.4%, P < 0.05) occurred more frequently in greater group than lesser group. According to the RIFLE criteria, 52 (66.6%) and 26 (33.4%) in our study fell into the R and I categories, respectively. By univariate analysis, age, presence of diabetes mellitus or hypertension, serum albumin, total leukocyte count, bilateral renal involvement and 50% or greater renal parenchymal involvement in CT scan were significant predictors of AKI. Adjusting for these factors in a multivariate logistic regression analysis, older age and CT finding of 50% or greater renal parenchymal involvement were the only significant predictors of AKI (Table 3).