Klara Sahlén: Optimizing diagnostics and follow-up of patients with ureterolithiasis

Abstract

Urolithiasis is increasing in incidence worldwide and subsequently so is the demand for customized imaging. Optimizing imaging strategies is needed to reduce radiation dose, time, and cost. This thesis investigated aspects of CT-diagnostics for patients with ureterolithiasis.

In paper I the rate of urinary obstruction, the predictive value of secondary signs of obstruction, stone size and location, and renal cortical enhancement was assessed in 49 patients with a remaining stone at follow-up CT after a renal colic attack. A dynamic protocol was used to grade urinary obstruction and associated radiation doses evaluated. In paper II, renal parenchymal volumes were measured in CT-scans before, during and after an acute ureteral obstruction using 3D-segmentation in 20 patients. In paper III the interreader variability in stone size measurements in four different window settings was investigated in 124 patients with a proximal ureteral stone. In paper IV, patient reported symptoms was correlated to the degree of obstruction (defined with the dynamic protocol) in 81 patients with a remaining ureteral stone at follow-up.

Obstruction was present in 28% (n=14). The absence of hydronephrosis and hydroureter had a negative predictive value (NPV) of 1.0. Stone characteristics were not associated with obstruction. Cortical enhancement was lower in obstructed kidneys. 1.8 mSv was saved using a dynamic scan compared to an excretory phase. Renal parenchymal volume decreased with 24% in the obstructed and 5% in the healthy kidney after obstruction. Interreader variability was ±1.6 mm (bone window), ±0.4 mm (soft tissue window), ±0.3 mm (half-value MEAN window) and ±0.2 mm (half-value MAX window). In total 47% (n=38) with a remaining stone were asymptomatic. Of these 11% (n=4) had a mild degree of obstruction. All patients with moderate or severe obstruction reported symptoms at follow-up.

In conclusion, urinary obstruction was scarce in patients with a remaining ureteral stone at follow-up. Absence of hydronephrosis and hydroureter was a strong negative predictor of obstruction. A dynamic protocol saved radiation dose. Obstructed kidneys increased in parenchymal volume during obstruction and so did the contralateral kidney to a lesser extent. Interreader variability in stone size measurements is unsatisfying in the bone window and smallest in the half-value MAX window. Symptoms is an uncertain tool to rule out the presence of a remaining ureteral stone and obstruction at follow-up.

The findings in this thesis contribute to the knowledge of CT-diagnostics in ureterolithiasis that may help sharpen the radiological work-up for patients with ureterolithiasis.