Rc. Smith et al., ACUTE URETERAL OBSTRUCTION - VALUE OF SECONDARY SIGNS ON HELICAL UNENHANCED CT, American journal of roentgenology, 167(5), 1996, pp. 1109-1113
OBJECTIVE. The purpose of our study was to determine the value of seco
ndary signs of ureteral obstruction on helical unenhanced CT. MATERIAL
S AND METHODS. Over a 19-month interval, 312 patients with acute flank
pain were imaged with helical unenhanced CT. Ureteral stone disease w
as confirmed to be present in 109 patients and confirmed to be absent
in 111 patients. Ninety-two remaining patients had no confirmatory ima
ging studies or surgery and were unable to be contacted for follow-up.
For each of the 220 patients with a confirmed diagnosis, we determine
d the presence or absence of ureteral or collecting system dilatation,
perinephric stranding, symmetry of renal size, and renal stones. In a
ll patients with a ureteral stone, we noted the presence or absence of
a circumferential rim of soft-tissue attenuation (''tissue-rim'' sign
) surrounding each stone and each phlebolith. RESULTS. The sensitivity
of each secondary sign was ureteral dilatation, 90%; perinephric stra
nding, 82%; collecting system dilatation, 83%; and renal enlargement,
71%. The specificity of each secondary sign was ureteral dilatation, 9
3%; perinephric stranding, 93%; collecting system dilatation, 94%; and
renal enlargement, 89%. Ureteral dilatation and perinephric stranding
were both present or both absent in 181 of the 220 patients with a co
nfirmed diagnosis. In this subgroup, this combination of signs had a p
ositive predictive value of 99% and a negative predictive value of 95%
. The odds ratio for the frequency of the tissue-rim sign with stones
versus tissue-rim with phleboliths was 31:1. CONCLUSION. When using un
enhanced CT to help diagnose acute flank pain, if one does not see a u
reteral stone or does see an indeterminate but suspicious calcificatio
n, then secondary signs of obstruction are important for diagnosis. Th
e results of this study form the basis of an imaging algorithm that ca
n be used when interpreting unenhanced CT images of patients with acut
e flank pain.