When does acute pyelonephritis require imaging?
A previously healthy 44-year-old woman presents to the emergency department with 1 day of fever, flank pain, dysuria, and persistent nausea and vomiting. Her temperature is 38.7°C (101.7°F), heart rate 102 beats per minute, and blood pressure 120/70 mm Hg. She has costovertebral angle tenderness. Laboratory testing reveals mild leukocytosis and a normal serum creatinine level; urinalysis shows leukocytes, as well as leukocyte esterase and nitrites. She has no personal or family history of nephrolithiasis. Urine cultures are obtained, and she is started on intravenous antibiotics and intravenous hydration to treat pyelonephritis.
Is imaging indicated at this point? And if so, which study is recommended?
KEY FEATURES
Acute pyelonephritis, infection of the renal parenchyma and collecting system, most often results from an ascending infection of the lower urinary tract. It is estimated to account for 250,000 office visits and 200,000 hospital admissions each year in the United States.1
Lower urinary tract symptoms such as urinary frequency, urgency, and dysuria accompanied by fever, nausea, vomiting, and flank pain raise suspicion for acute pyelonephritis. Flank pain is a key, nearly universal feature of upper urinary tract infection in patients without diabetes, though it may be absent in up to 50% of patients with diabetes.2
Additional findings include costovertebral angle tenderness on physical examination and leukocytosis, pyuria, and bacteriuria on laboratory studies.
PREDICTING THE NEED FOR EARLY IMAGING
Though guidelines state that imaging is inappropriate in most patients with pyelonephritis,2–4 it is nevertheless often done for diagnosis or identification of complications, which have been reported in more than two-thirds of patients.2–4
Acute pyelonephritis is generally classified as complicated or uncomplicated, though different definitions exist with regard to these classifications. The American College of Radiology’s Appropriateness Criteria2 consider patients with diabetes, immune compromise, a history of urolithiasis, or anatomic abnormality to be at highest risk for complications, and therefore recommend early imaging to assess for hydronephrosis, pyonephrosis, emphysematous pyelonephritis, and intrinsic or perinephric abscess.2
A clinical rule for predicting the need for imaging in acute pyelonephritis was developed and validated in an emergency department population in the Netherlands.3 The study suggested that restricting early imaging to patients with a history of urolithiasis, a urine pH of 7.0 or higher, or renal insufficiency—defined as a glomerular filtration rate (GFR) of 40 mL/min/1.73m2 or lower as estimated by the Modification of Diet in Renal Disease formula—would provide a negative predictive value of 94% to 100% for detection of an urgent urologic disorder (pyonephrosis, renal abscess, or urolithiasis). This high negative predictive value highlights that an absence of these signs and symptoms can safely identify patients who do not need renal imaging.
The positive predictive value was less useful, as only 5% to 23% of patients who had at least 1 risk factor went on to have urgent urologic risk factors.3
Implementation of this prediction rule would have resulted in a relative reduction in imaging of 40% and an absolute reduction of 28%. Of note, use of reduced GFR in this prediction rule is not clearly validated for patients with chronic kidney disease, as the previous GFR for most patients in this study was unknown.3
Based on these data, initial imaging is recommended in patients with diabetes, immune compromise, a history of urolithiasis, anatomic abnormality, a urine pH 7.0 or higher, or a GFR 40 mL/min or lower in a patient with no history of significant renal dysfunction. Early imaging would also be reasonable in patients with a complex clinical presentation, early recurrence of symptoms after treatment, clinical decompensation, or critical illness.
