a b s t r a c t

Purpose: The aim of this study was to determine if contrast-enhanced CT can safely exclude obstructive urolith- iasis in patients with flank plain. We performed a retrospective cohort analysis to compare the negative predic- tive values of contrast-enhanced and non-contrast CTs for the detection of obstructing urolithiasis.

Methods: Through report analysis, we identified all non-contrast and contrast-enhanced CT examinations of the abdomen and pelvis performed on adult patients in the emergency department at a single, multi-site academic medical institution in 2017 with an indication of flank pain. The prevalence of obstructive urolithiasis in each group was calculated. We subsequently analyzed 200 consecutive studies from each of these groups (reported negative for obstructive urolithiasis) for negative predictive value calculation. Follow up Abdominal imaging within 7 days from original presentation was used as a reference standard for analysis.

Results: In the noncontrast group, 1 study out of 200 was false negative (negative predictive value = 99.5%). In the contrast-enhanced group, there were no false negatives (negative predictive value = 100%). The prevalence of obstructive urolithiasis was 44.0% (351/797) in the noncontrast group and 18.7% (86/459) in the contrast- enhanced group.

Conclusion: Our results suggest that contrast-enhanced CT can safely exclude obstructing ureteral calculi in the setting of acute flank pain. This finding is of clinical relevance given the inherent benefit of IV contrast in diagnos- ing abdominopelvic pathology.

(C) 2021

1. Introduction

Flank pain is a common complaint among patients seen in the emer- gency department (ED), with an estimated 2.1 million ED visits in the US annually [1]. When obstructive urolithiasis is the suspected underly- ing cause, noncontrast computed tomography (NCCT) is the gold stan- dard diagnostic test and reliably identifies stones while also determining their size and location [1-7]. Utilization of NCCT increased significantly from a rate of 19.6% of encounters for acute flank pain in 2000 to 45% in 2008 [1]. Clinical suspicion for urolithiasis accounts for over 20% of all CT scans performed on patients presenting with non- traumatic Abdominal complaints [8,9].

Prospective studies on the performance of NCCT in detecting ure- teral calculi have demonstrated sensitivities ranging from 95 to 100% and specificities ranging from 92 to 100% [7,10]. However, 54-67% of patients presenting with acute flank pain do not have urolithiasis [7,11]. Among such cases, Nachmann et al. reported in their study that

* Corresponding author.

E-mail address: [email protected] (N. Harfouch).

¹ Permament Address: Staten Island University Hospital, Northwell Health, 475 Seaview Ave, Staten Island, NY 10305 USA.

32% were positive for other significant findings such as appendicitis, di- verticulitis, and renal/adnexal/colonic mass [11]. These pathologies are among a broad differential of urologic, gynecologic, gastrointestinal, hepatobiliary, and pancreatic processes which are recognized mimickers of renal colic [5,10,12] and also may be incompletely charac- terized or even undetected on NCCT [5,11,12]. It has therefore been sug- gested that indiscriminate use of NCCT can lead to operational inefficiency due to need for repeat and/or additional testing [12].

According to the 2018 American College of Radiology Manual on

Contrast Media, contrast-enhanced computed tomography (CECT) “has superior diagnostic performance compared to [NCCT] for a wide array of indications.” Thus, in equivocal cases of flank pain where under- lying causes other than Urinary stones require significant consideration, CECT should generally be the preferred test [5,12]. CECT also possesses advantages over NCCT when obstructing urolithiasis is present. Specifi- cally, CECT enables detection of certain secondary signs of Obstructive uropathy such as a delayed nephrogram [6] and increased ureteral en- hancement which are not seen on NCCT. Additionally, physiologic en- hancement of the ureter may aid in anatomic localization [5] and therefore differentiation between minimally obstructive Distal ureteral calculi and pelvic phleboliths, which has been described as a notable limitation of NCCT [13]. Obscuration of stones by intravenous contrast

https://doi.org/10.1016/j.ajem.2021.03.059

0735-6757/(C) 2021

media that has been excreted into the urinary collecting system could potentially be viewed as a drawback of CECT. However, in the era of multidetector CT, progression of contrast into the ureters where

Table 1

Study demographics and exclusions.

NCCT CECT

obstructing calculi often occur at the time of image acquisition is not likely to occur when standard imaging protocol is followed [5].

While NCCT has been established as a very effective test when uro- lithiasis is present, CECT can further enhance diagnostic performance in such cases and is also comprehensively superior for other underlying etiologies for acute flank pain. Given the prevalence of flank pain as a presenting complaint in the ED as well as the large proportion of these cases that do not involve obstructive urolithiasis, consideration of CECT over NCCT as a principal diagnostic tool may impact operational efficiency on a broad scale. However, direct, evidence-driven compari- sons between the two tests are needed. The ability of CECT to rule out obstructive urolithiasis compared to that of NCCT is of particular signif- icance and no scientific data on this specific matter has been previously submitted in the literature. We hypothesize that the negative predictive value (NPV) of CECT for obstructive urolithiasis would be similar to the NPV of NCCT.

1. Materials and methods

This study was approved by an appropriate institutional review board with a waiver for informed consent.

• 1. Study design

We conducted a retrospective cohort analysis to determine and compare the NPV of CECT (specifically, standard post-contrast examina- tions with image acquisition in the portal venous pHase) vs NCCT for the detection of obstructing urolithiasis. Obstructing urolithiasis was de- fined as one or more calculi within the ureteropelvic junction, ureter, or ureterovesicular junction in symptomatic patients. Follow up abdom- inal imaging (including ultrasound) within 7 days from original presen- tation was used as a reference standard for analysis. If the patient did not obtain repeat imaging within 7 days, the original negative CT was designated as true negative. However, if a patient was found to have an obstructing Ureteral stone or new hydronephrosis on follow up ab- dominal imaging within 7 days of the original study, the initial CT was designated as a false negative.

• 1. Subjects

Our institutional database, Montage, was queried to identify all NCCT (817) and CECT (469) studies of the abdomen and pelvis per- formed on adult patients in the ED in 2017 with an indication of “flank pain.” Any study with a history suggestive of trauma in the indi- cation (i.e words such as “trauma,” “motor vehicle collision,” or “fall”) was removed from the data set. The study was done at a two campus tertiary care medical institution, which is also the largest hospital on the residing island, comprising of approximately 800 beds. Each imag- ing report was individually reviewed further to apply exclusion criteria (Table 1). These included miscoded or otherwise incorrectly categorized examinations, patients with staghorn calculi, non-diagnostic examina- tions due to significant motion or beam-hardening artifact, and contrast within the urinary collecting system due to suboptimal technique or ad- ministration during a preceding test. The prevalence of obstructive uro- lithiasis i.e. the percentage of cases positive for obstructing ureteral calculi in each group was calculated. To determine the NPV of each exam type (CECT and NCCT), 400 consecutive examinations that were reported negative for obstructive urolithiasis were tabulated. Their breakdown is the following: 200 negative CECT and 200 negative NCCT exams (Fig. 1).

Total exams (after exclusions) 817 469

Male (%) 370 (45.3%) 171 (36.5%)

Female (%) 447 (54.7%) 298 (63.5%)

Mean Age (Standard Deviation) 49.8 (16.3) 50.4 (15.5)

Positive for obstructive urolithiasis 351 (44.0%) 86 (18.7%)

Exclusions^a 20 10

Technique 3 1

Artifact 4 1

Otherwise nondiagnostic 2 2

Staghorn calculi 6 1

Mislabeled/miscoded 5 5

*NCCT - noncontrast computed tomography.

**CECT - contrast-enhanced computed tomography.

^a All cases of traumatic flank pain were excluded at the time of the original search query for flank plain.

1286 Total Subjects

797 NCCT

459 CECT

200 NCCT

selected for NPV analysis

200 CECT

selected for NPV analysis

*NCCT - noncontrast computed tomography

**CECT - contrast-enhanced computed tomography

Fig. 1. Total subjects for prevalence analysis and subjects selected for NPV analysis. *NCCT - noncontrast computed tomography. **CECT - contrast-enhanced computed tomography.

• 1. CT technique

Images were obtained using either the Toshiba Aquilion One 320 CT (Toshiba, Japan) or the GE Revolution EVO 64 slice CT (GE Healthcare, Milwaukee). Slice thickness ranged from 1 to 5 mm and images were reformatted to include coronal and sagittal reconstructions. The scanner parameters for the Toshiba scanner were a rotation time of 0.5 s, a ref- erence tube current of 100-500mAs and a tube voltage of 120kVp. The GE scanner parameters were a rotation time of 0.6 s, a tube current of 100-500mAs and a tube voltage of 120kVp. When intravenous contrast was administered, iodinated contrast at a dose of 320 mg/ml was used (Optiray, Guerbet North Carolina). Contrast was injected at a rate of 2 ml/s via a power injector. Scan times began 75 s after administration of intravenous contrast for adequate portal phase protocol.

• 1. Report analysis

A total of 1286 CT exam reports were reviewed by an attending body radiologist (with 7 years experience) and a third year radiology resi- dent. Reports were analyzed for the presence of an obstructing ureteral

calculus in symptomatic patients. In cases where an initial negative exam was reported but was later found to be positive (within a 7 day period) i.e. a false negative exam, both radiologists reviewed the images

Table 2

Negative predictive value of NCCT and CECT.

NCCT^a CECT^b

with consensus agreement.

• 1. Statistical methods

Study participants in the two groups were described with respect to all characteristics recorded at baseline. Categorical variables were sum- marized by number and percentage in each category and continuous variables were summarized by mean and standard deviation or median and interquartile range as appropriate. Statistical significance was eval- uated at a one-sided alpha level of 0.05. All Statistical analyses were per- formed in SAS version 9.3 (Statistical Analysis Systems Inc., Cary, NC, USA).

1. Results

Subjects included in the study after exclusion criteria were applied include 797 subjects in the NCCT cohort and 459 in CECT cohort. Subject demographics are included in Table 1.

Among the 200 consecutive NCCT studies that were initially read as negative for obstructive urolithiasis, 1 study was a false negative (NPV

= 99.5%) (Fig. 2). Among the 200 consecutive CECT studies that were initially read as negative for obstructive urolithiasis, there were no false negatives (NPV = 100%) (Table 2).

The prevalence of obstructive urolithiasis was also calculated for both groups to ensure that the lack of false negatives was not due to scarcity of pathology (i.e. a prevalence near 0%). The prevalence of ob- structive urolithiasis was 44.0% (351/797) in the NCCT group and 18.7% (86/459) in the CECT group, ensuring that neither prevalence was extremely low.

1. Discussion

Our study provides a data-driven comparison between the NPV of CECT versus NCCT in ruling out obstructing ureteral stones in the setting of flank pain. The two Radiologic tests were highly and equally reliable

Fig. 2. Axial NCCT of a 36-year-old male with Right flank pain is significant for a punctate calculus (arrow) at the right ureterovesical junction (UVJ). This case represents the false negative exam in the non-contrast cohort.

Consecutive exams selected for analysis 200 200

False negatives 1 0

Negative predictive value (CI 0.99,1) 99.5% 100%

^a NCCT - noncontrast computed tomography.

^b CECT - contrast-enhanced computed tomography.

for excluding obstructing stones with NPVs that are very similar to the previously reported value of 98% for NCCT [11]. Our observations from this study evoke a number of different narratives pertaining to the work up of acute flank pain and imaging utilization.

We conclude that CECT can be employed for diagnostic evaluation of

a broad scope of abdominopelvic pathology without compromising per- formance in identifying obstructing urinary calculi relative to NCCT. In fact, as stated earlier, CECT even possesses certain advantages over NCCT in cases of ureteral obstruction, namely a delayed nephrogram (Fig. 3) and increased ureteral enhancement (Fig. 4), both of which are absent on NCCT. Moreover, our understanding of the risks of con- trast utilization has recently changed. As of January 2020, a consensus statement from the American College of Radiology and National Kidney Foundation has concluded that the risk of acute renal injury in patients with reduced kidney function after exposure to iodinated contrast is sig- nificantly less than previously believed. Whereas iodinated contrast media was previously contraindicated for individuals with estimated glomerular filtration rate of 30 or less who were not undergoing dialy- sis, Davenport et al. now recommend that patients in this subgroup may receive contrast following prophylaxis with intravenous normal saline. For individuals with estimated glomerular filtration rate of 45 or greater, studies have shown that the risk of contrast-induced acute kidney injury is near 0% [14]. When renal function screening would be indicated prior to CECT, the additional costs in time and resources for both the patient and healthcare system should be acknowledged. How- ever, such costs must be weighed against the disadvantages of NCCT im- aging and the possibility of incompletely characterized and undetected findings which may result in the need for additional testing - including further imaging - and/or delayed treatment.

The advent of Dual-energy CT (DECT) technology provides a power- ful diagnostic tool for identifying urolithiasis and characterizing stone composition in vivo with an unprecedented degree of specificity [15]. In general, this newer modality is capable of yielding the benefits of both NCCT and CECT from a single image acquisition with a breadth of potential useful applications. DECT is therefore a potent study for eval- uating a host of abdominopelvic pathologies although, with respect to

Fig. 3. Coronal CECT of a 32-year-old female with flank pain. Left nephrogram (arrows) and perinephric fluid (arrowheads). Delayed nephrogram is a conspicuous indirect sign of obstructive uropathy and is only seen after administration of intravenous contrast.

Fig. 4. Coronal CECT of a 51-year-old male with flank pain. The kidneys are symmetrically enhancing but there is urothelial enhancement (arrow heads) of the left renal pelvis. This finding can denote underlying pathology which may go undetected in the absence of intravenous contrast.

excluding obstructing stones specifically, the NPVs of NCCT and CECT we have observed leave minimal room for improvement. The capability of DECT to enhance contrast-to-noise ratio may allow for reduction of intravenous contrast media doses although it cannot eliminate the need for contrast administration altogether [16]. DECT is also of limited availability at this time and we therefore recommend continued atten- tion to its utility in the work up of suspected obstructive urolithiasis going forward.

We note incidentally that a disparity in prevalence of obstructive urolithiasis between the CECT (18.7%) and NCCT (44.0%) groups was demonstrated by our study. While our ED does not utilize a formalized algorithm to determine whether patients presenting with flank pain should undergo CECT versus NCCT, the decision is based on the degree of clinical suspicion for obstructive urolithiasis relative to alternative etiologies. However, this difference does not influence the inherent abil- ity of CECT or NCCT in ruling out obstructive urolithiasis by any means we are aware of.

Limitations of our study include its retrospective design and the use of patient re-presentation within 7 days with persistent flank pain as the criterion for determining if a CT scan was true versus false negative. However, few other practical reference standards for determining neg- ative predictive value in this particular setting are known to us. We be- lieve this test possesses adequate construct validity as symptomatic patients with missed obstructing stones are indeed likely to require short interval follow up. However, it is possible obstructing calculi that were truly present passed spontaneously after initial CT, leading to alle- viation of symptoms and forgoing need for follow up. It is also possible that patients with missed stones did re-present but to a different insti- tution. It is also feasible that a true negative could be incorrectly catego- rized as false negative if a non-obstructing renal stone advanced and became an obstructing ureteral stone in between the time of the initial CT and subsequent evaluation. For this reason, staghorn calculi were ex- cluded from the data set.

1. Conclusion

Our study demonstrates that contrast administration does not ad- versely affect the capability to exclude obstructing ureteral calculi and that the NPV of both CECT and NCCT are very high. Our findings cohere with other suggestions in the recent literature that CECT may be the test of choice for a broader subgroup of patients presenting with flank pain and that CECT is acceptable even when stone disease is suspected.

However, robust prospective randomized clinical trials should be per- formed to substantiate our initial observations.

Credit authors statement

Brandon Lei: Data curation, Investigation, Writing- original draft.

Nassier Harfouch: Formal analysis, Writing.

Jonathan Scheiner: Methodology, Project administration.

Seleshi Demissie: Formal analysis, Methodology, Validation.

Morris Hayim: Conceptualization, Data Curation, Formal analysis, Investigation, Project administration, Writing.

Declaration of Competing Interest

The authors have no relevant financial or non-financial interests to disclose.

References

1. Hyams ES, Korley FK, Pham JC, Matlaga BR. Trends in imaging use during the emer- gency department evaluation of flank pain. J Urol. 2011;186(6):2270-4. https://doi. org/10.1016/j.juro.2011.07.079.
2. ACR manual on contrast media version 10.3. Retrieved from https://www.acr.org/-/ media/ACR/Files/Clinical-Resources/Contrast_Media.pdf; 2018, June.
3. Coursey CA, Casalino DD, Remer EM, Arellano RS, Bishoff JT, Dighe M, et al. ACR ap- propriateness criteria(R) acute onset flank pain - Suspicion of stone disease. Ultra- sound Q. 2012;28(3):227-33. https://doi.org/10.1097/ruq.0b013e3182625974.
4. Kambadakone A, Andrabi Y, Patino M, Das C, Eisner B, Sahani D. Advances in CT im- aging for urolithiasis. Ind J Urol. 2015;31(3):185-93. https://doi.org/10.4103/0970- 1591.156924.
5. Corwin MT, Lee JS, Fananapazir G, Wilson M, Lamba R. Detection of renal stones on portal venous phase CT: comparison of thin axial and coronal maximum-intensity- projection images. AJR Am J Roentgenol. 2016;207(6):1200-4. https://doi.org/10. 2214/ajr.16.16099.
6. Cheng PM, Moin P, Dunn MD, Boswell WD, Duddalwar VA. What the radiologist needs to know about Urolithiasis: part 2 - CT findings, reporting, and treatment. AJR Am J Roentgenol. 2012;198(6):548-54. https://doi.org/10.2214/ajr.11.8462.
7. Dalrymple NC, Verga M, Anderson KR, Bove P, Covey AM, Rosenfield AT, et al. The value of unenhanced helical computerized tomography in the management of acute flank pain. J Urol. 1998;159(3):735-40. https://doi.org/10.1016/s0022-5347 (01)63714-5.
8. Rosen MP, Siewert B, Sands DZ, Bromberg R, Edlow J, Raptopoulos V. Value of ab- dominal CT in the emergency department for patients with abdominal pain. Eur Radiol. 2003;13(2):418-24. https://doi.org/10.1007/s00330-002-1715-5.
9. Abujudeh HH, Kaewlai R, Mcmahon PM, Binder W, Novelline RA, Gazelle GS, et al. Abdominopelvic CT increases Diagnostic certainty and guides management deci- sions: a prospective investigation of 584 patients in a large Academic Medical Cen- ter. AJR Am J Roentgenol. 2011;196(2):238-43. https://doi.org/10.2214/ajr.10.4467.
10. Colistro R, Torreggiani WC, Lyburn ID, Harris AC, Al-Nakshabandi NA, Nicolaou S, et al. Unenhanced helical CT in the investigation of acute flank pain. Clin Radiol. 2002;57(6):435-41. https://doi.org/10.1053/crad.2001.0871.
11. Nachmann MM, Harkaway RC, Summerton SL, Horrow MM, Kirby CL, Fields RG, et al. Helical CT scanning: the primary imaging modality for acute flank pain. Am J Emerg Med. 2000;18(6):649-52. https://doi.org/10.1053/ajem.2000.16292.
12. Desai V, Cox M, Deshmukh S, Roth CG. Contrast-enhanced or noncontrast CT for renal colic: utilizing urinalysis and patient history of urolithiasis to decide. Emerg Radiol. 2018;25(5):455-60. https://doi.org/10.1007/s10140-018-1604-0.
13. Freed KS, Paulson EK, Frederick MG, Preminger GM, Shusterman DJ, Keogan MT, et al. Interobserver variability in the interpretation of unenhanced helical CT for the diagnosis of ureteral stone disease. J Comput Assist Tomogr. 1998;22(5): 732-7. https://doi.org/10.1097/00004728-199809000-00013.
14. Davenport MS, Perazella MA, Yee J, Dillman JR, Fine D, RJ McDonald, et al. Use of in- travenous iodinated contrast media in patients with kidney disease: consensus statements from the American College of Radiology and the National Kidney Foun- dation. Radiology. 2020;294(3):660-8. https://doi.org/10.1148/radiol.2019192094.
15. Hidas G, Eliahou R, Duvdevani M, Coulon P, Lemaitre L, Gofrit ON, et al. Determina- tion of renal stone composition with dual-energy CT: in vivo analysis and compari- son with X-ray diffraction. Radiology. 2010;257(2):394-401. https://doi.org/10. 1148/radiol.10100249.
16. Goo HW, Goo JM, Dual-Energy CT. New horizon in medical imaging. Korean J Radiol. 2017;18(4):555-69. https://doi.org/10.3348/kjr.2017.18.4.555.