American Journal of (2010) 28, 813–816

www.elsevier.com/locate/ajem

Brief Report Can the degree of on predict kidney stone size?☆ Jacob K. Goertz MD a,⁎, Seth Lotterman MD a,b aDepartment of Emergency Medicine, Long Island Jewish Medical Center, New Hyde Park, NY 11040, USA bDepartment of Emergency Medicine, Wilford Hall Medical Center, Lackland Air Force Base, TX 78236, USA

Received 7 April 2009; revised 10 June 2009; accepted 10 June 2009

Abstract Objective: The aim of the study was to determine if the degree of hydronephrosis on focused emergency renal ultrasound correlates with kidney stone size on computed . Methods: A retrospective study was performed on all adult patients in the who had a focused emergency renal ultrasound and ureterolithiasis on noncontrast computed tomography. Severity of hydronephrosis was determined by the performing physician. Ureteral stone size was grouped into 5 mm or less and larger than 5 mm based on likelihood of spontaneous passage. Results: One hundred seventy-seven ultrasound scans were performed on patients with ureteral calculi. When dichotomized using test characteristic analysis, patients with none or mild hydronephrosis (72.9%) were less likely to have ureteral calculi larger than 5 mm than those with moderate or severe hydronephrosis (12.4% vs 35.4%; P b .001) with a negative predictive value of 0.876 (95% confidence interval, 0.803-0.925). Conclusion: Patients with less severe hydronephrosis were less likely to have larger ureteral calculi. © 2010 Elsevier Inc. All rights reserved.

1. Introduction angle tenderness, and hematuria [2]. Multiple imaging modalities, including intravenous pyelography, noncontrast Symptomatic renal colic is a common complaint present- computed tomography (CT), and ultrasound, have been used ing to the emergency department (ED). An estimated 5% to to diagnose renal colic [3-11]. In many EDs, noncontrast CT 15% of the population will have a kidney stone during their has become the dominant imaging modality because it allows lifetime with an ED use rate of 126 to 226 per 100 000 visits determination of stone size and location, degree of hydro- [1,2]. Diagnosis of renal colic is often made solely on the nephrosis, and evaluation of other pathophysiologic pro- basis of history, physical examination, and urinalysis, cesses that may masquerade as renal colic [2,10-12]. Despite especially in patients presenting with typical complaints of the advantages of CT, drawbacks include cost, accessibility, flank pain radiating to the ipsilateral groin, costovertebral availability, increased length of ED stay, and risks from [6,13-15]. Focused emergency ultrasound is advantageous because it is nonionizing, inexpensive, and can be performed at the ☆ Previously presented at the American College of Emergency Physicians Scientific Assembly, Seattle, Wash, October 2007. patient's bedside thereby becoming an integral part of the ⁎ Corresponding author. Tel.: +1 718 470 7501; fax: +1 718 470 9113. physical examination [16,17]. Multiple studies using ultra- E-mail address: [email protected] (J.K. Goertz). sound in the evaluation of renal colic have demonstrated that

0735-6757/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.ajem.2009.06.028 814 J.K. Goertz, S. Lotterman

curvilinear or 3S 1.5 to 3.5-MHz phased array ultrasound Table 1 Interobserver agreement between emergency physician and QA ultrasound findings probe (General Electric, Fairfield, Conn) or a SonoSite MicroMaxx ultrasound machine with a C60e 2 to 5-MHz Emergency QA hydronephrosis curvilinear or P17 1 to 5-MHz phased array ultrasound probe physician None Mild Moderate Severe Total (SonoSite, Bothell, Wash). All ultrasound examinations hydronephrosis were recorded onto S-VHS videotape or onto DVD-R for None 22 4 1 0 27 video quality assurance review. The impression of the Mild 7 90 3 0 100 performing clinician was concurrently entered into the Moderate 0 1 42 0 43 ultrasound database. Severe 0 0 0 5 5 Hydronephrosis was defined as none, mild, moderate, or Total 29 95 46 5 175 severe according to standard definitions. Mild hydronephro- κ = 0.847 (95% confidence interval, 0.777-0.918). sis was defined as enlargement of the calices with preservation of the renal papillae, moderate hydronephrosis was defined as rounding of the calices with obliteration of the ultrasound is accurate at detecting hydronephrosis and that renal papillae, and severe hydronephrosis was defined as emergency physicians can accurately determine the presence and degree of hydronephrosis but that ultrasound has limited caliceal ballooning with cortical thinning [17]. All ultra- sound examinations were subsequently reviewed for quality use in directly detecting calculi or determining stone location assurance (QA) by an emergency ultrasound fellowship- [4-9]. Furthermore, the presence of hydronephrosis on trained emergency physician and an independent nonblinded focused emergency ultrasound and hematuria on urinalysis quantification of the degree of hydronephrosis was made is accurate in detecting and sufficient for diagnosing renal using the above standardized definitions. Ureteral stone size colic [18,19]. Despite ultrasound not being as accurate as was extracted from the CT report in the medical record after other modalities in determining stone location, stone size is a QA review. more important determinant of likelihood of stone passage and clinical outcome. Ureteral calculi less than 5 mm are The performing physician's interpretation of the degree of hydronephrosis, the degree of hydronephrosis on QA review, likely to pass spontaneously regardless of location in the and the ureteral stone size on CT were entered into SPSS ureter, whereas those greater than 5 mm have a decreased (version 16.0; SPSS Inc, Chicago, Ill) and analyzed using χ2 likelihood of spontaneous passage or an increased likelihood testing. Ureteral stone size was stratified into 2 groups, those of need for eventual urologic intervention [20-22]. For larger 5 mm or smaller and those larger than 5 mm, based on the stones, current guidelines recommend a trial of passage with likelihood of successfully spontaneous stone passage, and medical expulsion therapy for calculi between 5 and 10 mm; test characteristics were calculated [20-22]. stones larger than 10 mm will require surgical removal in most cases [21]. The objective of our study is to determine if the degree of hydronephrosis on focused emergency renal ultrasound 3. Results correlates with ureteral calculi size on noncontrast CT and to ascertain if these results would allow for prediction of One hundred seventy-seven were performed stone size. in the ED on patients who had confirmed ureteral calculi on CT. Of these patients, 120 (67.8%) were male, and the mean age was 48.1 years old (range, 18-89 years old). One hundred 2. Methods forty-four patients (81.4%) had ureteral calculi 5 mm or smaller, and 33 patients (18.6%) had ureteral calculi larger We performed a retrospective chart review using an than 5 mm. For the ultrasounds performed, there was very emergency ultrasound database containing the findings and good interobserver agreement between the degree of interpretations of the focused emergency ultrasounds hydronephrosis as determined by the performing emergency performed in the ED of an academic medical center with an annual census of 70 000. All consecutive adult patients (≥18 years) in the database who had a complete focused Table 2 Hydronephrosis vs ureteral calculi size emergency renal ultrasound performed and a noncontrast CT Hydronephrosis Calculi ≤ 5 mm Calculi N 5 mm Total documenting ureterolithiasis for a 4-year period from March 2004 to March 2008 were enrolled. Approval for review with None 25 3 28 a waiver of informed consent was obtained from the Mild 88 13 101 Moderate 30 13 43 institutional review board. Severe 1 4 5 Ultrasounds were performed in the ED by emergency Total 144 33 177 medicine resident or attending physicians using either a GE b Logiq 5 ultrasound machine with a 3.5C 2 to 5-MHz P .001. Hydronephrosis and kidney stone size 815

likely to pass spontaneously, if a CT were not performed on Table 3 Hydronephrosis categories vs ureteral calculi size all patients with none or mild hydronephrosis, the need for Hydronephrosis Calculi ≤ 5 mm Calculi N 5 mm Total CT would be decreased by 73% at the risk of missing only None or mild 113 16 129 16 patients (9%) with calculi larger than 5 mm. Although this Moderate or severe 31 17 48 is not insignificant, all had calculi smaller than 10 mm Total 144 33 177 making them eligible for medical expulsion therapy; none P b .001. had calculi larger than 10 mm, the size at which stones will likely require urologic intervention [20-21]. The sensitivity for detection of stones larger than 5 mm is poor. Of the patients with more severe hydronephrosis, over physician and the QA review (Table 1), with κ = 0.847 (95% one third had calculi larger than 5 mm; of those, 35% had confidence interval, 0.777-0.918). calculi larger than 10 mm. Although the poor sensitivity and Increasing degree of hydronephrosis on ultrasound was the small number of patients with larger stones, especially associated with an increasing proportion of ureteral calculi those larger than 10 mm, make it difficult to draw larger than 5 mm as shown in Table 2 (P b .001). Analysis of conclusions, current guidelines suggest a trial of passage the test characteristics demonstrated best discrimination by dichotomizing the degree of hydronephrosis into none or with medical expulsion therapy for stones sized 5 to 10 mm [21]. This suggests that in the appropriate clinical setting, mild hydronephrosis (less severe hydronephrosis) and patients with less severe hydronephrosis on ED ultrasound moderate or severe hydronephrosis (more severe). When might be candidates for discharge with a trial of passage with the degree of hydronephrosis was collapsed into these medical expulsion therapy and outpatient urologic follow- 2 categories, 129 patients (72.9%) had less severe hydrone- up, whereas those with more severe hydronephrosis should phrosis and 48 patients (27.1%) had more severe hydrone- have further evaluation to determine stone size and the need phrosis. One hundred thirteen patients (87.6%) with less for possible emergent urologic consultation or intervention severe hydronephrosis had ureteral calculi 5 mm or smaller. None of the 16 patients with less severe hydronephrosis and vs discharge with medical expulsion therapy and urgent urologic follow-up. ureteral calculi larger than 5 mm had stones larger than This study suggests that bedside focused emergency 10 mm, whereas 6 (35.3%) of the 17 patients with more ultrasound may be a method to screen patients and use CT severe hydronephrosis and larger ureteral calculi had stones scans more prudently because most patients in our study had larger than 10 mm. Of the patients with calculi 5 mm or neither larger kidney stones nor more severe hydronephrosis. smaller, 113 (78.5%) had less severe hydronephrosis; Avoiding unnecessary CT scans could have a significant 17 patients (51.5%) with calculi larger than 5 mm had impact in the ED and on public heath. Emergency more severe hydronephrosis. As shown in Table 3, patients with none or mild department throughput could be significantly enhanced; one study showed an average delay of 147 minutes between hydronephrosis were less likely to have larger ureteral the performance of the focused emergency ultrasound and calculi than those with moderate or severe hydronephrosis the CT scan [9]. Concerns have been raised about the risk of (12.4% vs 35.4%; P b .001). This dichotomization had a cancer from exposure to ionizing radiation from CT scans, sensitivity of 0.515 (95% confidence interval, 0.339-0.688), and recommendations have been made to limit CT scanning a specificity of 0.785 (95% confidence interval, 0.707- to that which is medically justified [14]. Because the 0.847), a positive predictive value of 0.354 (95% confidence recurrence of renal calculi in patients may be as high as interval, 0.226-0.506), and a negative predictive value of 0.876 (95% confidence interval, 0.804-0.925). 50%, were these patients to have multiple CT scans, the cumulative dose of radiation over their lifetimes becomes critically important [1,14,15]. This study suggests that in an appropriate clinical setting, such as for a patient with 4. Discussion symptoms of recurrent renal colic and in whom alternative serious pathologic condition can be ruled out, focused Our study demonstrated a relationship between the degree emergency renal ultrasound may be able to obviate the need of hydronephrosis found on focused emergency ultrasound for CT scanning in the ED. Further study is needed for and ureteral stone size on CT. We observed that increasing evaluation and validation of such a diagnostic paradigm. degree of hydronephrosis on ultrasound was associated with The major limitations of this study are due to the an increasing proportion of ureteral calculi larger than 5 mm. retrospective nature of the study. Although all consecutive Most patients in this study had less severe (none or mild) patients identified in the ultrasound database whom had a hydronephrosis; these patients were less likely to have larger focused emergency renal ultrasound performed and a CT ureteral calculi. The specificity of an ED ultrasound scan demonstrating ureterolithiasis were included, there were demonstrating none or mild hydronephrosis is 78.5% with likely other patients presenting with renal colic who either a negative predictive value to predict a ureteral calculus sized had a CT but no ultrasound or who had no imaging 5 mm or less of 87.6%. Because calculi less than 5 mm are performed in the ED, thereby, raising potential selection bias. 816 J.K. Goertz, S. Lotterman

It is unknown if the ultrasounds were performed before the [3] Miller OF, Rineer SK, Reichard SR, et al. Prospective comparison of CT or if the performing physician had access to the CT unenhanced spiral computed tomography and intravenous urogram in the evaluation of acute flank pain. Urology 1998;52:982-7. results before performing the ultrasound. The QA review, [4] Yilmaz S, Sindel T, Arslan G, et al. Renal colic: comparison of spiral however, was performed before extracting the CT results CT, US and IVU in the detection of ureteral calculi. Eur Radiol 1998;8: from the medical record, and the agreement between the 212-7. performing physician's interpretation and the independent [5] Erwin BC, Carroll BA, Sommer FG. Renal colic: the role of ultrasound QA review was very good, therefore, making it unlikely that in initial evaluation. 1984;152:147-50. [6] Sheafor DH, Hertzberg BS, Freed KS, et al. Nonenhanced helical CT that possible nonblinding of the performing physician to the and US in the emergency evaluation of patients with renal colic: CT results contributed significant bias. We were unable to prospective comparison. Radiology 2000;217:792-7. control for degree of hydration, and we were also unable to [7] Sinclair D, Wilson S, Toi A, Greenspan L. The evaluation of suspected determine if the duration of symptoms (as a surrogate for renal colic: ultrasound scan versus excretory urography. Ann Emerg duration of ureteral obstruction) correlated with the degree of Med 1989;18:556-9. [8] Henderson SO, Hoffner RJ, Aragona JL, et al. Bedside emergency hydronephrosis. Furthermore, this study is limited in that we department ultrasonography plus of the kidneys, ureters, are studying a group of patients with a known diagnosis of and bladder vs intravenous pyelography in the evaluation of suspected ureterolithiasis on CT rather than all patients coming to the ureteral colic. Acad Emerg Med 1998;5:666-71. ED with a presentation suggestive of renal colic. Variation in [9] Rosen CL, Brown DFM, Sagarin MJ, et al. Ultrasonography by presenting complaints such as location or intensity of pain, emergency physicians in patients with suspected ureteral colic. J Emerg Med 1998;16:865-70. results of other ED testing such as urinalysis demonstrating [10] Vieweg J, Teh C, Freed K, et al. Unenhanced helical computerized hematuria, or in ED management, patient course, and tomography for the evaluation of patients with acute flank pain. J Urol disposition is unknown. Finally, we were unable to follow 1998;160:679-84. patient outcomes such as the rate of and time to successful [11] Brown J. Diagnostic and treatment patterns for renal colic in US passage of calculi to determine the effects of the degree of emergency departments. Int Urol Nephrol 2006;38:87-92. [12] Katz DS, Scheer M, Lumerman JM, et al. Alternative or additional hydronephrosis or of stone size. diagnoses on unenhanced helical computed tomography for suspected In conclusion, our results demonstrate a relationship renal colic: experience with 1000 consecutive examinations. Urology between the degree of hydronephrosis as determined by 2000;56:53-7. emergency physicians on focused emergency ultrasound and [13] Rekant E, Gilbert CL, Counselman FL. Emergency department time ureteral calculi size; patients with less severe hydronephrosis for evaluation of patients discharged with a diagnosis of renal colic: unenhanced helical computed tomography versus intravenous uro- were less likely to have larger ureteral calculi. This suggests graphy. J Emerg Med 2001;21:371-4. that ultrasound can help identify many, but not all, patients [14] Brenner DJ, Hall EJ. Computed tomography—an increasing source of who are at lower risk for having larger ureteral calculi. radiation exposure. N Engl J Med 2007;357:2277-84. [15] Katz SI, Saluja S, Brink JA, et al. Radiation dose associated with unenhanced CT for suspected renal colic: impact of repetitive studies. AJR 2006;186:1120-4. Acknowledgments [16] Swadron S, Mandavia D. Renal. In: Ma OJ, Mateer JR, Blaivas M, editors. Emergency ultrasound. 2nd ed. New York: McGraw-Hill; 2008. p. 229-55. Statistical analysis was assisted by Cristina Sison, PhD, [17] Noble VE, Brown DFM. Renal ultrasound. Emerg Med Clin N Am Senior Research Statistician, Feinstein Institute for Medical 2004;22:641-59. Research. Alexander Tsukerman, MD, Department of [18] Gaspari RJ, Horst K. Emergency ultrasound and urinalysis in the Emergency Medicine, Long Island Jewish Medical Center evaluation of flank pain. Acad Emerg Med 2005;12:1180-4. [19] Kartal M, Eray O, Erdogru T, et al. Prospective validation of a current performed data entry. algorithm including bedside US performed by emergency physicians for patients with acute flank pain suspected for renal colic. Emerg Med J 2006;23:341-4. [20] Miller OF, Kane CJ. Time to stone passage for observed ureteral References calculi: a guide for patient education. J Urol 1999;162:688-91. [21] Preminger GM, Tiselius HG, Assimos DG, et al. 2007 guideline for the [1] Pearle MSP, Calhoun EA, Curhan GC, et al. Urologic diseases in management of ureteral calculi. J Urol 2007;178:2418-34. America project: urolithiasis. J Urol 2005;173:848-57. [22] Papa L, Steill IG, Wells GA, et al. Predicting intervention in renal colic [2] Teichman JMH. Clinical practice. Acute renal colic from ureteral patients after emergency department evaluation. Can J Emerg Med calculus. N Engl J Med 2004;350:684-93. 2005;7:78-86.