Imaging in Stone Diagnosis and Surgical Planning

Emily C. Serrell; Sara L. Best

Disclosures

Curr Opin Urol. 2022;32(4):397-404. 

In This Article

CT Assessment of Stones for Management Planning

Size

Abdominal CT remains the most accurate imaging modality for evaluation of stone-specific characteristics, though transverse measurements may underestimate maximal stone diameter. If the radiologist only measures stone diameter from transverse series, this may misrepresent actual stone size.[72] Coronal measurement of longitudinal stone diameter may be more helpful in stone measurement and has been shown to be a significant predictor of successful stone passage.[72–75] There are no appreciable measurement differences between low dose and conventional CT,[74] and improved accuracy can be obtained by measuring in bone window and small slice thickness.[76–79]

Stones are three-dimensional and thus irregularly shaped, and so diameter does not necessarily correlate closely with volume. Stone volume may be more helpful in patient counseling, treatment selection, and prediction of clinical risk. Stones <9 mm trend toward prolate (rugby-ball) ellipsoid, stones 9–15 mm toward oblate (disc) ellipsoid, and stones >15 mm toward scalene ellipsoid shapes.[80] There is thus far no optimal manual calculation for stone volume.[79–83] Automated assessment of stone volume is more accurate and reproducible than manual measurements.[79,81,83–92]

Stone volume has been used to predict stone behavior with some success according to recent investigations. Three studies in the past 2 years have evaluated machine learning models to predict spontaneous ureteral stone passage based on volume, with areas under the curve of 0.83–0.85.[91,93,94] This may also be applied to predict surgical efficacy. Larger volume stones are associated with a lower stone-free rate after SWL[84,95–97] and ureteroscopy,[85,98,99] as well as higher complication rates[100] and longer operative time for ureteroscopy.[85,98,101] For PCNL, larger volume stones are associated with lower stone free rate and need for second procedure,[83] increased operating time,[83,88,102] increased lasering time,[101] higher complication rates,[88] decreased hematocrit,[88,102] and need for secondary procedure. The distribution of stone burden volume in the pelvicalyceal system (ratio of stone volume to renal collecting system volume) has been associated with higher fluoroscopic timing and complication rates with an AUC 0.869.[88,103]

Composition

Another stone-specific characteristic being reported in recent literature is stone composition, particularly as uric acid stones may be managed with chemolysis. Historically the relative Hounsfield units density (HUD) on CT scan was used to predict stone composition.[104] The HUD is best measured in bone window while magnifying the stone using zoom; measured HUD may be higher on lower dose CT than standard CT.[105,106] Calcium oxalate stones have high HUD (650–1500).[107] Struvite have significant variability (500–1000) and cannot be accurately predicted by HUD.[108] There is significant overlap between cystine and uric acid stones (250–750).[107,109] Pure uric acid stones are identifiable by low HUD as well as being more homogenous than other stones (one study reports reproducibility using a stone heterogeneity index <140 HUD;[109] this index is also associated with higher stone free rates after SWL).[110] Automated pixel mapping software has been applied to CT imaging to detect uric acid stones with sensitivity 89% and specificity of 91%.[111]

A newer imaging modality used for the prediction of stone composition is dual-energy CT. Initially described in the mid-2000s, dual energy CT transmits both a low- and high-energy X-ray beam to one or two detectors.[112,113] In comparison, a conventional CT transmits a single beam to a single detector. With dual energy, the difference in attenuation of materials at the two different energy levels may be pictorially applied to CT imaging to predict stone composition or differentiate between ureteral stents and adjacent stones.[114]

In the ex vivo setting, dual energy CT has a reported 97–100% accuracy in differentiating between uric acid and nonuric acid stones.[115] A brief literature review identified seven papers evaluating the use of dual energy CT's accuracy in prediction of in vivo uric acid versus nonuric acid stones and reported: Sensitivity (85–100%), specificity (72–100%), positive predictive value (64–100%), and negative predictive value (96–98.5%) in differentiating uric acid vs nonuric acid stones.[116–122] This can be accomplished reproducibly without increased radiation exposure and at doses equivalent to low-dose CT (1.2–4.2 mSv).[118,121,123–125]

Dual-energy CT is superior to conventional CT in[124,126] differentiating stone composition, though its utility may be limited by very small stones or large patients.[120] However, quantitative models incorporating max HUD and peak point Laplacian-maximum attenuation algorithm may be applied to single energy CT to be comparable in accuracy to dual-energy CT.[127]

The benefit of identifying uric acid stones early is that these stones can be managed noninvasively with oral chemolysis through the alkalinization of urine. A 2020 randomized trial was carried out on medium (1–2.5 cm) radiolucent renal stones (≤500 HUD) in nonobese patients who self-regulated potassium citrate dose for a target urine pH 7–7.2. Oral dissolution therapy was associated with a stone-free rate of 16% in 1 month and 50% in 3 months.[128] Likewise, three nonrandomized recent studies have reported 3 months stone free rates of 61–64%[129,130] as well as a 6 months stone free rate of 83%.[131] Patient selection factors associated with response included smaller stones, lower density, and lower HUD.[129–131] Patient selection is imperative, and this may be a particularly appealing option for patients who want to avoid surgery.

processing....