| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, University of Michigan Hospital, Rm. B1D502G,
University of Michigan Hospital, Ann Arbor, MI 48109-0030.
2
Department of Internal Medicine, University of Michigan Hospital, Ann Arbor,
MI 48109-0030.
Received December 18, 2000;
accepted after revision June 19, 2001.
Presented at the 100th annual meeting of the American Roentgen Ray Society,
Washington, DC, May 2000.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Two radiologists and a radiology resident, unaware of the final diagnosis, reviewed preselected helical CT images from renal stone examinations in patients with 65 indeterminate pelvic calcifications. Assessment of calcifications for rim or comet-tail signs was performed independently of an assessment for the following five secondary signs of urinary tract obstruction: caliectasis, pelviectasis, ureterectasis, perinephric stranding, and renal enlargement. Agreement in identifying rim and comet-tail signs was assessed by obtaining kappa statistics. The utility the of rim or comet-tail signs in determining whether ureterolithiasis was present in patients in whom perinephric stranding and ureterectasis were present or absent was determined. The frequency with which one or more of each of the five assessed secondary signs was identified ipsilateral to a calcification having rim or comet-tail signs was also tabulated.
RESULTS. Kappa values for interobserver agreement ranged from 0.49 to 0.73. In only one patient was a rim sign detected in the absence of ureterectasis and perinephric stranding. Reviewers identified at least three of the five assessed secondary signs ipsilateral to calcifications showing a rim sign in all but one patient (by each radiologist) and four patients (by the resident). When three or more secondary signs of obstruction were seen ipsilateral to a calcification having a comet-tail sign, in all but one instance, this was because the calcification was a ureteral calculus or because there was a separate ipsilateral ureteral calculus.
CONCLUSION. In many instances, observers did not agree about whether the rim and comet-tail signs were present. The rim sign was observed in the absence of any secondary signs of urinary tract obstruction in only one (1.5%) of the 65 patients in our series (95% confidence interval, 0-5.3%). The comet-tail sign, when accompanied by secondary signs of obstruction, should indicate that an ipsilateral ureteral stone is present and not the reverse.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The rim and comet-tail signs have been described as helpful aids in evaluating pelvic calcifications on CT when their relationship to the ureter is uncertain [12, 13, 15, 17]. The rim sign, a circumferential rim of soft-tissue attenuation surrounding an abdominal or pelvic calcification, indicates that the calcification is ureteral (positive predictive value, 84-100% [12, 13, 15]). The comet-tail sign, a linear or curvilinear soft-tissue structure extending from an abdominal or pelvic calcification, generally indicates that a calcification is a phlebolith (positive predictive value, 100% [15, 17]).
In our experience, these signs are rarely a deciding factor in identifying ureterolithiasis on CT because the correct diagnosis is usually made by identifying a calcification that is obviously ureteral or by detecting ipsilateral secondary signs of ureteral obstruction in a patient with an indeterminate calcification. To our knowledge, no prior reports have assessed the frequency with which the rim and comet-tail signs have been associated with ipsilateral secondary signs of ureteral obstruction. Our study was undertaken to determine how often a rim sign was detected when few or no other secondary signs of ureteral obstruction were present and how often a comet-tail sign adjacent to an indeterminate calcification was associated with one or several ipsilateral secondary signs.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Scan Selection and Determination of Final Diagnosis
The initial reviewer evaluated all indeterminate lower abdominal or pelvic
calcifications (thought to be located at or near the distal ureter) for the
presence or absence of the rim and comet-tail signs. All scans containing an
indeterminate calcification with a rim sign (n = 20) or a comet-tail
sign (n = 24) were selected for further analysis. Twenty-one
additional scans containing a pelvic calcification believed to be located in
or near a distal ureter that did not show either sign were then selected at
random from the remaining 206 CT scans. These patients were selected as
control subjects for our study so that the subsequent reviewers would not be
able to assume that all calcifications showed either one sign or the other.
Thus, in all, 65 calcifications were included in the study. The initial
reviewer then evaluated each calcification chosen for subsequent analysis
using all images on the CT scan to determine whether the calcification
represented a ureteral calculus or a phlebolith. This determination was
correlated with clinical and surgical follow-up.
Scan Analysis
Three independent reviewers (two experienced senior genitourinary
radiologists [henceforth referred to as "radiologist 1" and
"radiologist 2"] and a third-year radiology resident [henceforth
referred to as "resident"]), unaware of patient symptoms and the
final radiologic and clinical diagnosis, evaluated the selected CT scans
containing the 65 calcifications. Each reviewer analyzed the images during two
separate independent imaging sessions.
During the first session, only images containing the suspicious pelvic calcifications were evaluated. This allowed the reviewers to assess only the features of the calcification in question. The reviewers were asked whether the calcification exhibited a rim sign (defined as a circumferential rim or halo of soft-tissue attenuation surrounding the calcification) or a comet-tail sign (defined as an adjacent eccentric tapering soft-tissue mass or a linear or curvilinear area of soft-tissue attenuation adjacent to the calcification or when the calcification is in the crescentic soft-tissue attenuation rectovesical venous plexus). These definitions, used by the authors who first identified these signs [12, 13, 15, 17], were provided on each data sheet used by the reviewers. Reviewers were then asked whether they believed that the calcification was located in the ureter on the basis of only their assessment of the position of the calcification. Using a 5-point scale, reviewers graded responses to all questions as "definitely," "definitely not," "probably," "probably not," or "can't tell."
The second review session was separated from the first by at least 1 week. At this time, each of the reviewers evaluated corresponding images of only the kidneys and proximal ureters in the same patients. The reviewers assessed images for the presence or absence of the following five potential secondary signs of urinary tract obstruction: caliectasis, pelviectasis, ureterectasis, perinephric stranding, and renal enlargement.
Data Analysis
Interobserver agreement for the presence of the rim sign, the comet-tail
sign, and for the calcification location (as ureteral or extraureteral) was
calculated using weighted kappa statistics
[18]. The five-point scale
just described was condensed into a weighted three-point scale. Responses of
"definitely" and "probably" were considered
affirmative and responses of "probably not" and "definitely
not" were considered negative. A response of "can't tell"
was considered as a third indeterminate category. The weighted kappa values
were interpreted as follows: a score of less than 0.21 indicated poor
agreement; 0.21-0.40, fair agreement; 0.41-0.60, moderate agreement;
0.61-0.80, good agreement; and greater than 0.80, very good agreement
[19].
The number of ureteral calculi and phleboliths around which each reviewer identified rim and comet-tail signs was determined. When a rim or a comet-tail sign was identified as being probably or definitely present by a reviewer, we determined which secondary signs of obstruction were identified on the same CT scan by that reviewer. The 95% confidence interval for the frequency with which a rim sign would be present around a calcification in a patient with no secondary signs of urinary tract obstruction was calculated using the normal approximation to the binomial with continuity correction [18]. We also tabulated the total number of ipsilateral secondary signs identified by each reviewer on CT scans that contained a calcification showing rim or comet-tail signs.
Finally, because ureterectasis and perinephric standing have been the best predictors in determining the presence or absence of ureterolithiasis [20], patients were divided into groups on the basis of the presence or absence of these two signs. For each group, we determined whether the rim sign or comet-tail sign was present and whether these signs were associated with diagnosed calculi or phleboliths.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Identification of Rim and Comet-Tail Signs
Radiologists 1 and 2 and the resident identified a rim sign around 16, 19,
and 21 of the 37 ureteral calculi, respectively. Neither radiologist
identified a rim sign around any of the 28 phleboliths, whereas the resident
noted a rim sign around two. Although radiologists 1 and 2 and the resident
identified comet-tail signs adjacent to 14, 20, and 20, respectively, of the
phleboliths, they also identified comet-tail signs adjacent to two, three, and
four of the calculi, respectively.
Interobserver Agreement
When one reviewer's observations were compared with those of the other
reviewers, agreement was moderate to good for identification of both the rim
and comet-tail signs (with weighted kappa statistics ranging between 0.49 and
0.73). However, many instances occurred in which at least one of the three
reviewers disagreed with the interpretations of one or both of the others.
Although at least one reviewer believed that the rim and comet-tail signs were
probably or definitely associated with 27 and 29 of the analyzed
calcifications, all three reviewers agreed that the rim sign was probably or
definitely present around only 13 calcifications and the comet-tail sign was
probably or definitely adjacent to only 11 calcifications (Figs.
1,2,3,4).
|
|
|
|
The Rim Sign: Identification of Ipsilateral Secondary Signs of
Urinary Tract Obstruction
One or more of the five assessed secondary signs of urinary tract
obstruction were usually detected ipsilateral to a calcification that was
surrounded by a rim (Table 1).
Caliectasis and pelviectasis were most frequently identified, and renal
enlargement was least frequently identified (87-95% and 50-64% of patients,
respectively).
|
The Rim Sign: Association with Ureterectasis and Perinephric
Stranding
In most instances, when a reviewer identified a rim sign surrounding a
calcification, both ipsilateral ureterectasis and perinephric stranding were
also visualized by the same reviewer (Table
2). A rim sign was detected around a calcification diagnosed as a
ureteral calculus in the absence of both of these secondary signs in only one
instance (by both radiologists but not by the resident). Thus, when these two
secondary signs of urinary tract obstruction were considered, in only one
instance was the rim sign the sole indicator that a pelvic calcification was a
ureteral calculus. In all, this finding represents only one of 37 patients
with ureterolithiasis in the group of 65 patients whose scans were selected
from a larger group of 250 CT scans. Therefore, the rim sign was only rarely
the sole indicator of ureterolithiasis (0.4% of all cases). Furthermore,
according to the resident, two phleboliths in patients with no ipsilateral
secondary signs present also showed a rim sign. If the resident had used the
rim sign as an indicator for urolithiasis, these phleboliths might have been
erroneously diagnosed as ureteral calculi.
|
Table 2 reviews the potential ability of the rim sign to differentiate calculi from phleboliths in the following four subgroups of patients: those with both ureterectasis and perinephric stranding; those with ureterectasis but no perinephric stranding; those with perinephric stranding but no ureterectasis; and those without both ureterectasis and perinephric stranding. For the rim sign to be maximally helpful, the percentages of calculi showing this sign and the percentages of phleboliths not showing this sign should be high in each group. When ureterectasis (with or without perinephric stranding) was identified ipsilateral to the calcification being assessed, the majority of such calcifications represented ureteral calculi regardless of whether or not a rim sign was present. In comparison, when perinephric stranding alone (without ureterectasis) was present, most calcifications showing a rim sign were calculi, whereas most of those not showing a rim sign were phleboliths. When neither ureterectasis nor perinephric stranding was identified, the calcification being assessed was almost always a phlebolith. As previously mentioned, in the subgroup of patients without ureterectasis and perinephric stranding, in only one instance was a rim sign detected around a calcification by either radiologist 1 or 2 (and this calcification was determined to be a calculus). In contrast, the resident identified a rim sign surrounding two calcifications, both of which were phleboliths.
The Rim Sign: Total Number of Ipsilateral Secondary Signs
Three or more of all five assessed secondary signs of urinary tract
obstruction were identified by radiologist 1 and radiologist 2 ipsilateral to
18 of 19 and 15 of 16 calcifications, respectively, that were surrounded by a
rim (Table 3). Neither
radiologist 1 nor radiologist 2 identified secondary signs ipsilateral to the
remaining calcification around which a rim was detected. The resident
identified three or more secondary signs ipsilateral to 19 of 23
calcifications that were surrounded by a rim
(Table 3). The resident
identified only one or two secondary signs ipsilateral to the four other
calcifications in which a rim sign was identified (two of which were diagnosed
on initial review as phleboliths). The resident detected ipsilateral
ureterectasis in the one patient with a calcification showing a rim sign in
which both radiologists saw no secondary signs.
|
The Comet-Tail Sign: Ipsilateral Secondary Signs
Secondary signs were usually absent ipsilateral to calcifications showing a
comet-tail sign (Table 1).
The Comet-Tail Sign: Association with Ureterectasis and Perinephric
Stranding
In this series, there were many instances in which the reviewers identified
ureterectasis or perinephric stranding ipsilateral to calcifications adjacent
to a comet-tail sign that was visualized on CT
(Table 4). This finding was
usually a result of the presence of ipsilateral ureteral calculi. Also,
although the comet-tail sign was frequently identified adjacent to phleboliths
in patients who had neither ipsilateral ureterectasis nor perinephric
stranding and who did not have ipsilateral ureterolithiasis, radiologist 1 and
the resident each identified a comet-tail sign adjacent to a ureteral calculus
(the "pseudotail" sign
[17]) in a patient who had
neither ureterectasis nor perinephric stranding. Thus, the presence of a
comet-tail sign was not helpful in determining whether or not ipsilateral
ureterolithiasis was present in any of the groups.
|
The Comet-Tail Sign: Total Number of Secondary Signs
Radiologist 1, radiologist 2, and the resident identified three or more
secondary signs ipsilateral to calcifications showing the comet-tail sign in
four, six, and five instances, respectively
(Table 3). In all but one of
these patients, an ipsilateral ureteral calculus was diagnosed on initial
review, with the calcification showing the comet-tail sign representing a
calculus in two, four, and three instances (a pseudotail sign)
(Fig. 5). In two, one, and two
cases a separate ipsilateral ureteral calcification was located elsewhere in
the pelvis. In only one instance did a reviewer (radiologist 2) identify three
secondary signs ipsilateral to a calcification showing a comet-tail sign when
no ureteral calculus was diagnosed on that side.
|
Diagnosing the Nature of the Calcification by Analysis of
Position
When attempting to determine whether pelvic calcifications were ureteral or
extraureteral exclusively on the basis of an assessment of the position of the
calcification in relation to the expected location of the ureter, radiologist
1, radiologist 2, and the resident correctly identified 24 (65%), 27 (73%),
and 31 (84%) of the 37 calculi as being located in the ureter and 21 (75%), 22
(79%), and 23 (82%) of the 28 phleboliths as being extraureteral.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The positive predictive value of secondary signs in diagnosing ureterolithiasis has been previously shown to be extremely high. The combination of ureterectasis and perinephric stranding, for example, has been reported to correctly predict the presence of an ipsilateral ureteral stone with an accuracy of 99% [20]. Because in the majority of cases in which the rim sign is present it is possible to determine whether ipsilateral ureterolithiasis is present, reliance on the presence or absence of the rim sign is usually unnecessary. Assessing a pelvic calcification for a rim sign would be warranted only when some or all secondary signs are absent or when the urologist would like to know which of several indeterminate calcifications might represent an obstructing ureteral calculus.
Indeed, the rim sign may be more useful when only a few secondary signs are present. Our results suggest that the rim sign might be useful for predicting that a calcification is a ureteral calculus in those patients who have only perinephric stranding identified. In our study, in this group of patients, all calcifications surrounded by a rim sign were ureteral calculi sign, and all but one of the calcifications adjacent to a comet-tail were phleboliths. Interestingly, even in these instances, our reviewers did not usually rely on the rim sign to make a diagnosis but instead were able to correctly determine whether or not the calcification was a calculus on the basis of its position in the pelvis in relation to the expected course of the ureter.
As previously mentioned, the rim sign may be useful in determining which of several indeterminate pelvic calcifications might represent the obstructing ureteral calculus. Although secondary signs may alert the radiologist to the presence or absence of ipsilateral ureterolithiasis, they do not help in determining whether any individual calcification is a ureteral stone. In our experience, most problematic calcifications are small, and the majority of these pass with conservative treatment (of analgesics and hydration). In general, it is the larger calculi (which are less likely to exhibit the rim sign) or the calculi that remain symptomatic (often leading to additional imaging) that are treated with urologic intervention. Thus, in the acute setting, it is usually only important that a diagnosis of ureterolithiasis be made or excluded.
Additionally, in some cases, it may not be possible to determine which of several tiny indeterminate pelvic calcifications represents an obstructing calculus. In fact, the only way a suspected calculus seen on CT can be definitively determined to represent an obstructing calculus is if it is absent on a repeated renal stone CT scan obtained after a calculus has been passed. Such routine follow-up examinations are rarely performed on patients who have passed ureteral calculi. Lack of repeated imaging represents a limitation of this and most of the other previously published studies because a direct correlation between a passed calculus and a suspected calculus seen on CT is possible only on rare occasions.
Urinary tract obstruction must be present for at least a few hours for secondary signs of obstruction to develop [21,22,23]. The rim sign might be more beneficial in the subgroup of patients with acutely obstructed urinary tracts who are referred for a CT scan shortly after they become symptomatic and who may not have enough time to develop secondary signs. However, the rim sign usually does not appear for 4-24 hr after the patient becomes symptomatic [23]. Thus, this sign will also frequently be absent in patients who have an acutely obstructed collecting system that has not yet had time to develop the other secondary signs of obstruction.
Our evaluation also showed limitations in the clinical utility of the comet-tail sign, because there are occasional instances in which multiple secondary signs are noted ipsilateral to calcifications showing the comet-tail sign. In our study, this finding was usually a result of the presence of an ipsilateral obstructing ureteral calculus (with the calcification being analyzed representing the calculus or a separate calculus being located elsewhere). In these instances, the use of the comet-tail sign to exclude a diagnosis of ureterolithiasis might have led the radiologist astray from the correct diagnosis.
Our study differs in several ways from previous reports. First, previous studies have assumed that reviewers can consistently identify the rim and comet-tail signs [12, 13, 15, 17]. To our knowledge, none of these reports used more than one reviewer to evaluate the same CT images. We found that although agreement among the reviewers was moderate to good, there were many instances in which one reviewer believed one of the signs to be "probably" or "definitely" present, whereas one or both of the other reviewers did not. This disagreement occurred despite our inclusion of the definitions of the rim and comet-tail signs on each data sheet that our reviewers completed.
Second, in prior studies assessing calcifications for the rim and comet-tail signs, CT scans have been evaluated in their entirety [12, 13, 15, 17]. Reviewers may have been biased when evaluating a pelvic calcification if they had already determined whether secondary signs of urinary tract obstruction were present. By separating the evaluation of the abdominal or pelvic calcifications from that of the kidneys and upper ureters and by withholding patient history (including side of abdominal pain), our reviewers were able to make an assessment for the presence of the rim and comet-tail signs with less bias. Although they could occasionally determine whether ipsilateral ureterectasis was present, they were unaware of the appearance of the ipsilateral proximal collecting systems (for caliectasis, pelviectasis, perinephric stranding, and renal enlargement). Similarly, when evaluating the kidneys and proximal ureters for secondary signs of collecting system obstruction, our reviewers were unaware of the side of patient pain and of the appearance of the calcification in question. This type of evaluation better assesses the utility of the rim and comet-tail signs as independent indicators of the diagnosis of renal stones on CT.
Third, our study also purposefully included a resident reviewer, because many CT examinations for ureteral stones are performed during the evening and are initially interpreted by a resident. If the rim and comet-tail signs were even of some exclusive benefit to radiology residents who might be less skilled in identifying secondary signs of obstruction, then these signs would still be of some value. We did not find the signs to be any more effective for the resident than for the radiologists. In fact, the resident (but neither of the radiologists) identified a rim sign around a phlebolith in two patients with no secondary signs of obstruction present. This finding could have led to the false assumption that the phleboliths were calculi. Also, the resident (and one of the radiologists) identified a comet-tail sign adjacent to a ureteral calculus in a patient who had neither ipsilateral ureterectasis nor perinephric stranding. This could have led to the false assumption that this calculus was a phlebolith.
Our study had a number of limitations. When evaluating the accuracy of ureteral stones identified on CT, we found no true gold standard for determining the ability of CT to differentiate ureteral calculi from pelvic phleboliths unless a follow-up CT examination was performed and showed interval movement or disappearance of the calcification in question (thereby indicating that it represented a calculus). The follow-up examination was rarely performed in our patient population. Instead, in our study, the initial reviewer used the presence or absence of secondary signs of urinary tract obstruction and the location of the calcification to reach a final CT diagnosis. Furthermore, there was no guarantee that when a stone was recovered from a patient, the stone corresponded with a suspicious calcification seen on CT. Previous studies have indicated that there are inherent inaccuracies in determining the presence or absence of stone disease by clinical follow-up [10, 12, 21]. Thus, although flawed, the combination of clinical follow-up and radiologist assessment using all images and clinical information available was the only feasible means of determining a diagnosis in our study.
Additionally, despite our attempts to limit bias by separating the evaluation of the calcification from the evaluation for secondary signs of obstruction, all information about the nature of the calcification in question could not be hidden from our reviewers. For example, when evaluating pelvic calcifications, our reviewers may have been biased by the presence or absence of ureterectasis immediately adjacent to a calcification. Finally, our study did not contain any patients with secondary signs that were produced by abnormalities other than urolithiasis (including patients with acute pyelonephritis, obstruction not caused by calculus disease, renal infarction, or renal vein thrombosis).
In summary, our study indicates that the rim and comet-tail signs are seldom the sole indicators of the presence or absence of ipsilateral ureterolithiasis. Determination of the presence or absence of secondary signs of obstruction ipsilateral to a suspicious calcification and estimation of the position of a calcification in relation to the observed or expected location of the ureter is nearly always sufficient to allow a correct diagnosis. Identification of a rim sign around a ureteral calculus without any ipsilateral secondary signs of obstruction is rare (in our study, occurring in only one patient). Although the rim sign may occasionally contribute to the diagnosis in patients with only perinephric stranding present, it is usually not helpful. Additionally, whereas the comet-tail sign may be identified adjacent to phleboliths, it is occasionally seen adjacent to calculi or in patients with ipsilateral calculi located elsewhere in the pelvis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  P.-A. Poletti, A. Platon, O. T. Rutschmann, F. R. Schmidlin, C. E. Iselin, and C. D. Becker Low-Dose Versus Standard-Dose CT Protocol in Patients with Clinically Suspected Renal Colic Am. J. Roentgenol., April 1, 2007; 188(4): 927 - 933. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Rucker, C. O. Menias, and S. Bhalla Mimics of Renal Colic: Alternative Diagnoses at Unenhanced Helical CT RadioGraphics, October 1, 2004; 24(suppl_1): S11 - S28. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
