Ultrasound vs CT for the detection of ureteric stones in

patients with renal colic

1M PATLAS, MD, 2A FARKAS, MD, 1D FISHER, MD, 1I ZAGHAL, MD and1I HADAS-HALPERN, MD

Departments of 1Radiology and 2Urology, Shaare Zedek Medical Center, Jerusalem 91031, Israel

Abstract. The aim of our study was to compare the accuracy of non-contrast spiral CT with

ultrasound (US) for the diagnosis of ureteral calculi in the evaluation of patients with acute flank

pain. 62 consecutive patients with flank pain were examined with both CT and US over a period

of 9 months. All patients were prospectively defined as either positive or negative for

ureterolithiasis, based on follow-up evaluation. 43 of the 62 patients were confirmed as having

ureteral calculi based on stone recovery or urological interventions. US showed 93% sensitivity

and 95% specificity in the diagnosis of ureterolithiasis; CT showed 91% and 95%, respectively.

Pathology unrelated to urinary stone disease was demonstrated in six patients. Although both

modalities were excellent for detecting ureteral stones, consideration of cost and radiation lead

us to suggest that US be employed first and CT be reserved for when US is unavailable or

non-diagnostic.

Imaging evaluation of patients with acute flankpain is traditionally based on intravenous urog-raphy (IVU) as the standard screening tool fordetecting urinary calculi. IVU requires iv contrastmedium, with its associated potential risks [1]. Inaddition, the length of this examination maypreclude rapid evaluation of patients in an emer-

gency setting. These considerations have led to theuse of other techniques, such as the combinationof plain abdominal radiography and ultrasound(US) [2], and more recently unenhanced helicalCT [3]. Plain radiographs are not sensitive to non-radio-opaque calculi or to non-calculous obstruc-tion. Plain radiography also lacks specificity, asphleboliths, which are common pelvic calcifica-tions, are not always readily differentiated fromurinary tract calculi [4]. The advantages of CTover IVU are well documented and includeshorter examination time, avoidance of iv contrastmedium, greater sensitivity for stone detectionand increased detection of abnormalities unre-lated to ureteral stones. However, radiation doseis high [5]. Transabdominal US has the advantageof being universally available, does not exposethe patient to radiation, requires no iv contrastmedium and is independent of kidney function;US is therefore attractive as the modality ofchoice for the initial evaluation of urinarysymptoms. This prospective study compared the

accuracy of spiral CT with US in the evaluationof patients with acute flank pain. Plain radiograph

studies were excluded since plain radiography andUS have already been compared with CT [6].

Patients and methods

62 consecutive patients, seen for suspected renalcolic in our emergency department over a 9month period, were enrolled in a standardizeddouble-blinded protocol that consisted of USexamination followed by CT. There were 42 menand 20 women. The age range was 26–89 years.All imaging studies were conducted within 4 h ofadmission to the emergency department.

US examination was performed transabdomin-ally, after ingestion of 400 ml of water, with anATL Ultramark 9 system (Advanced TechnologyLaboratories, Bothell, WA) using 3.5 MHz, 5 MHzand 7.5 MHz probes. Examinations were con-ducted by one of three experienced senior radiolo-gists (DF, IZ, IHH). US diagnosis of ureteralcalculi required the demonstration of an intra-luminal hyperechoic structure causing acousticshadowing. The presence of hydronephrosis andperinephric fluid were also noted.

Non-enhanced helical CT examinations wereperformed with an Elscint Helicat II scanner(Marconi Medical Systems, Cleveland, OH). CTimages were obtained from the upper renal polesto the bladder base. Helical data acquisitionconsisted of 6.5 mm thick sections and a pitch of1.5:1. No oral or iv contrast medium was admini-stered. The CT examinations were reviewed by anexperienced radiologist (MP) and were evaluatedfor the presence of ureteral calculi, perinephric orperiureteric stranding, and hydronephrosis. CT

Received 12 January 2001 and in revised form 29 May2001, accepted 22 June 2001.

Address correspondence to Irith Hadas-Halpern, MD.

The British Journal of Radiology, 74 (2001), 901–904 E 2001 The British Institute of Radiology

901The British Journal of Radiology, October 2001

diagnosis of ureteral calculi was established byvisualization of a high attenuation structure(greater than 100 Hounsfield units) within theureteral lumen.

The two sets of studies were reviewed byindependent radiologists who were blinded tothe patient’s identity and who noted all findingsincluding stone demonstration, stone size, andlocation and signs of obstruction. Findings notrelated to calculi were also noted.

Results

43 of the 62 patients were confirmed to haveureteral calculi based on stone recovery orurological interventions. The US and CT findingsare summarized in Table 1.

US demonstrated ureterolithiasis in 40 of the 43patients confirmed to have ureteral calculi (sensi-tivity 93%, specificity 95%, positive predictivevalue 98%, negative predictive value 86%). Fourcalculi were located in the upper third of theureter, four in the middle third (Figure 1) and 32in the distal ureter (Figure 2).

Hydronephrosis was noted in 44 cases. Thedegree of hydronephrosis demonstrated by USexamination was graded as minimal in 22patients, mild in 11 patients and moderate in 11patients. Perinephric fluid was demonstrated inthree patients.

Of the 43 patients with calculi, CT detected 39(sensitivity 91%, specificity 95%, positive predic-tive value 98%, negative predictive value 82%). 5calculi were demonstrated in the proximal ureter,4 in the midureter (Figure 3) and 30 in the distalureter (Figure 4).

Perinephric stranding was seen in 26 cases, andperiureteric stranding in 5 cases.

Pathology unrelated to urinary stones wasdemonstrated in six patients and included appen-dicitis, cholelithiasis, cholecystitis and adnexalmass in one patient each, and torted ovarian cystin two patients. All of these conditions weredetected by US and CT except the appendicitis,which was diagnosed by CT alone.

Discussion

Recent studies have shown that non-contrast

spiral CT is an excellent method for demonstrat-

ing ureteral calculi in patients with suspected

renal colic [2]. Smith et al [3] showed non-contrast

CT to be more effective than IVU in identifying

ureteral stones. In another comparative study,

Sommer et al [6] noted that reformatted, non-

contrast spiral CT images were superior to a

combination of US and plain abdominal radiog-

raphy for imaging ureteral calculi. In the current

study, a comparison was made between spiral CT

and US in 62 patients, with comparable results

for the two modalities in the demonstration of

ureteral calculi. In some cases it was difficult to

ascertain on CT whether calcification was within

the urinary tract or elsewhere, e.g. calcified phle-

boliths or a calcified seminal vesicle (Figure 5).In one case, CT interpretation was false positive

for a ureteral calculus, and retrospectively the

calcification was shown to be a pelvic phlebolith.

Four patients passed stones (2–5 mm in size),

none of which had been seen on CT. Non-

visualization of stones may be explained by

volume averaging, small stone size and/or low

attenuation value of the stones.US, which is universally available, non-inva-

sive, inexpensive and radiation free, is preferred

by some radiologists as the initial method for

evaluation of the kidneys and bladder. However,

US is considered to be of limited value in

demonstrating pathological conditions of the

ureter [8].All patients with ureterolithiasis described

herein had some degree of ureterohydronephrosis,

hence US was able to follow the ureter to the level

Table 1. Results of imaging with ultrasound and CTfor detection of ureteral calculi

Ureteralcalculipresent

Ureteralcalculiabsent

TotalNo.cases

UltrasoundPositive for ureteral calculi 40 1 41Negative for ureteral calculi 3 18 21Total 43 19 62CTPositive for ureteral calculi 39 1 40Negative for ureteral calculi 4 18 22Total 43 19 62

Figure 1. Ultrasound shows an intraluminal echogenicfocus with an acoustic shadow in the right midureter.

M Patlas, A Farkas, D Fisher et al

902 The British Journal of Radiology, October 2001

of the stone and demonstrate the exact nature of

the obstructing lesion. An intraluminal echogenic

focus with acoustic shadowing was clearly

depicted in all cases. Technical problems might

occur in assessing the ureter when the stone is in

the middle third, an area often obscured by bowel

gas; we overcame this problem by compressing

the area to be examined and changing the

patient’s position.Dalla Palma [9] evaluated 120 patients with

renal colic using US and plain radiographs, and

achieved 95% sensitivity but only 67% specificity.

US was classified as positive for ureteric colic in

the study when calculi or hydronephrosis were

present. In the current study, only cases with a

definite demonstration of ureteral calculi were

classified as positive and our results show a high

specificity of 95%.We did not evaluate resistive index (RI) values

or ureteric jets in our study. Others have recently

examined the role of RI with disappointing

results. Cronan [10] showed that the addition of

RI to renal US did not improve the 77%

(a) (b)

Figure 2. Ultrasound demonstration of distal ureteral calculi. (a) Minimal hydronephrosis. (b) Mildly dilateddistal ureter with an echogenic focus inside (arrow).

Figure 3. CT shows a calcified stone in the right mid-ureter.

Figure 4. CT shows a calcified stone in the rightdistal ureter.

Ultrasound vs CT for detection of ureteric stones

903The British Journal of Radiology, October 2001

sensitivity of US in that series. The use of colourDoppler for ureteric jets was studied by Burge et al[11]. Most cases of high-grade urinary tractobstruction had abnormal jets, whereas jetswere often normal in low-grade obstruction ornon-obstucting stones. In our study, CT and USwere equally sensitive in detecting ureteral calculi;91% and 93%, respectively. In the study bySommer et al, there were false negative USexaminations owing to a lack of significant hydro-nephrosis detectable on the examination [6]. Inour patients, US was also accurate in depictingstones in cases of minimal hydronephrosis.

Extraurinary causes mimicking renal colic weredemonstrated by both modalities except in onecase of appendicitis that was diagnosed by CTonly. However, the small number of cases withextraurinary causes precluded statistical analysis.

In summary, both spiral CT and US werefound to be excellent modalities for depictingureteral stones, but because of high cost, radiationdose and high workload of CT, we suggest thatUS should be performed first in all cases and

CT should be reserved for cases where USis unavailable or fails to provide diagnosticinformation.

References

1. Gavant ML. Low-osmolar contrast media in the1990s. Guidelines for urography in a cost-sensitiveenvironment. Invest Radiol 1993;28(Suppl. 5):S13–19.

2. Erwin BC, Carroll BA, Sommer FG. Renal colic: therole of ultrasound in initial evaluation. Radiology1984;152:147–50.

3. Smith RC, Rosenfield AT, Chol KA. Acute flankpain: comparison of non-contrast-enhanced CT andintravenous urography. Radiology 1995;194:789–94.

4. Koelliker SL, Cronan JJ. Acute urinary tractobstruction. Imaging update. Urol Clin North Am1997;24:571–83.

5. Denton ER, Mackenzie A, Greenwell T, Popert R,Rankin SC. Unenhanced helical CT for renal colic—is the radiation dose justifiable? Clin Radiol 1999;54:444–7.

6. Sommer FG, Jeffrey RB, Rubin GD, Napel S,Rimmer SA, Benford J, et al. Detection of ureteralcalculi in patients with suspected renal colic: value ofreformatted noncontrast helical CT. AJR 1995;165:509–14.

7. Vieweg J, Teh C, Freed K, Leder RA, Smith RHA,Nelson RH, et al. Unenhanced helical computerizedtomography for the evaluation of patients with acuteflank pain. J Urol 1998;160:679–84.

8. Wong-You-Cheong JJ, Wagner BJ, Davis CJ.Transitional cell carcinoma of the urinary tract.Radiologic–pathologic correlation. Radiographics1998;18:123–42.

9. Dalla Palma L, Stacul F, Bazzocchi M, Pagnan L,Festini G, Marega D. Ultrasonogrphy and plain filmversus intravenous urography in ureteric colic. ClinRadiol 1993;47:333–6.

10. Cronan JJ, Tublin ME. Role of the resistive indexin the evaluation of acute renal obstruction. AJR1995;164:377–8.

11. Burge HJ, Middleton WD, McClennan BL.Ureteral jets in healthy subjects and in patientswith ureteral calculi: comparison with color DopplerUS. Radiology 1991;180:437–42.

Figure 5. CT shows multiple calcifications of theseminal vesicles.

M Patlas, A Farkas, D Fisher et al

904 The British Journal of Radiology, October 2001