Original Paper
Urologia
Received: September 29, 2014 Accepted after revision: November 28, 2014 Published online: January 29, 2015
Urol Int 2015;94:210–214 DOI: 10.1159/000370247
Internationalis
Accuracy of Unenhanced Computerized Tomography Interpretation by Urologists in Patients with Acute Flank Pain Eyup Burak Sancak a Mustafa Resorlu b Orcun Celik c Berkan Resorlu a Murat Tolga Gulpinar a Alpaslan Akbas a Tolga Karakan d Omer Bayrak e Mucahit Kabar d Muzaffer Eroglu d Huseyin Ozdemir b
Departments of a Urology and b Radiology, Canakkale Onsekiz Mart University, Faculty of Medicine, Canakkale, c Department of Urology, Tepecik Training and Research Hospital, Izmir, d Department of Urology, Ankara Training and Research Hospital, and e Department of Urology, Gazi University, Faculty of Medicine, Ankara, Turkey
Abstract Purpose: The aim was to compare the findings of non-contrast computerized tomography (NCCT) evaluated by urology specialists with the findings of experienced radiologists, who are accepted as a standard reference for patients who present with acute flank pain. Materials and Methods: Five hundred patients evaluated with NCCT were included in the study. The NCCT images of these patients were evaluated by both radiologists and urology specialists in terms of the presence of calculus, size of calculus, the location of calculus, the presence of hydronephrosis, and pathologies other than calculus, and the results were compared. Results: The evaluations of urology specialists and standard reference radiology specialists are consistent with each other in terms of the presence of calculus (kappa [κ]: 0.904), categorical stone size (κ: 0.81), the location of calculus (κ: 0.88), and hydronephrosis (κ: 0.94). However, the evaluations of urology specialists in detecting pathologies other than calculus, which may cause acute flank pain or accompany renal colic, were found to be inadequate (κ: 0.37). The false-negative rate of detect-
© 2015 S. Karger AG, Basel 0042–1138/15/0942–0210$39.50/0 E-Mail [email protected] www.karger.com/uin
ing pathologies outside of the urinary system by the urology specialists is calculated as 0.86. Conclusion: Although the urology specialists can evaluate the findings related to calculus sufficiently with NCCT, they may not discover pathologies outside of the urinary system. © 2015 S. Karger AG, Basel
Introduction
The incidence and prevalence of urinary tract calculus disease is increasing worldwide, and the number of patients applying to emergency services due to renal colic is also increasing accordingly [1–6]. However, many urological and non-urinary disorders (appendicitis, ovarian torsion, epididymo-orchitis, diverticulitis, cholecystitis) apart from calculus disease are associated with acute flank pain [7, 8]. It is very important that diagnosis be quick and accurate for these patients, since a significant number of these disorders may require emergency treatment. In the current guidelines, non-contrast computerized tomography (NCCT) is recommended for the diagnosis of patients presenting with acute flank pain as the primary imaging method, which has a high sensitivity and specificity, and also the capability of detecting patholoBerkan Resorlu, MD Canakkale Onsekiz Mart Universitesi Terzioglu Yerleskesi, Barbaros Mh TR–17100 Canakkale (Turkey) E-Mail drberkan79 @ gmail.com
Downloaded by: NYU Medical Center Library 198.143.38.97 - 6/18/2015 9:40:12 AM
Key Words Acute flank pain · Computed tomography · Radiologist · Urologist
Patients and Methods Study Population The data of 579 patients, presenting to either the Emergency Service or Urology Department between January 2013 and January 2014 and assessed by NCCT, were evaluated retrospectively. However, since more than one stone was detected in 67 patients, some defects were detected in the images of 10 patients and some clinical data were missing in 2 patients; a total of 79 patients were excluded from the study, and the study comprised 500 patients. The NCCT images were evaluated by 2 experienced radiology specialists, and the results were accepted as a reference. Afterward, the images were evaluated by 5 experienced urology specialists, with each specialist interpreting 100 images. The main purpose was to compare the findings of the radiology specialists, which are considered the standard reference, and the interpretations of the urology specialists in terms of presence of stone, size of stone, location of stone, presence of hydronephrosis, and detecting pathologies other than calculus. No additional clinical information was given to either urology or radiology specialists, other than the side of the pain. The radiology specialists evaluated the images in a quiet and darkened environment on a 27-inch screen, and they used the coronal images if needed. Urology specialists examined only the images with transverse cross-sections on a 23-inch screen under proper clinical conditions and in a bright environment. The images were evaluated by both radiology and urology specialists in terms of the presence of the stone, size of the stone, categorical stone size (≤4, 5–9, ≥10 mm), location of the calculus (lower ureter [lower cross iliac], upper ureter [above cross iliac], renal pelvis, lower pole, medium pole, upper pole), presence of hydronephrosis, and presence of pathologies other than calculus, and the results were compared with each other. If there was no consensus on the presence or absence of the stone, a clinical examination was conducted. A number was given to every pathology other than calculus, and this numeric system was used for further evaluations. In addition, the pathologies other than calculus were categorized into three groups as follows: pathologies associated with the urinary tract; pathologies not related to the urinary tract; and the pathologies outside of both the urinary tract and urinary system.
Accuracy of Unenhanced CT Interpretation in Patients with Acute Flank Pain
Table 1. Demographic data and stone characteristics
Number of patients Mean age ± standard deviation, years Gender ratio Male Female Number with stone Stone laterality Right Left Stone location Upper pole Middle pole Lower pole Renal pelvis Distal ureter Proximal ureter Stone size, mm Stone size (categorical) ≤4 mm 5–9 mm 10 mm Hydronephrosis Yes No
500 47.8±16.5 59.8 40.2 273 52.8 47.2 14 (5.1%) 36 (13.1%) 72 (26.3%) 25 (9.1%) 75 (27.4%) 51 (18.6%) 5.04±6.2 50 127 96 123 377
Results
The study consisted of 299 male patients (59.8%) and 201 female patients (40.2%), which was 500 patients in total. The average age was 47.8 ± 16.5 (between 3–82 years old), and 99% were older than the age of 18. Stones were detected in 273 (54.6%) patients by the radiologists, in which the average size of the stone detected was 5.04 ± 6.2 mm. Looking at the localization of the stones; 14 (5.1%) were in the upper pole, 36 (13.1%) were in the middle pole, 72 (26.3%) were in the lower pole, 25 (9.1%) were in the renal pelvis, 51 (18.6%) were in the proximal ureter, and, finally, 75 (27.4%) were located in the distal ureter. The demographic data of the patients and characteristics of the calculi are summarized in table 1. There was a consensus for 476 patients in terms of the presence or absence of stone as a result of the evaluations of both urology and radiology specialists. The radiologists detected no stone in 6 out of 261 patients (2.3%) in whom the urology specialists detected a stone, and the radiologists detected a stone in 18 out of 239 patients (7.5%) in whom the urology specialists did not detect a stone (sensitivity 93.4%, specificity: 97.4%; negative predictive value [NPV]: 92.4%, positive predictive value [PPV]: 97.7%). The kappa value is calculated as 0.904 for this evaluation, which is described as a near-perfect match. Urol Int 2015;94:210–214 DOI: 10.1159/000370247
211
Downloaded by: NYU Medical Center Library 198.143.38.97 - 6/18/2015 9:40:12 AM
gies other than calculus [9]. However, despite providing quick detection with this technique, the radiologist’s report cannot be obtained so quickly. In this waiting period, NCCT scans are often interpreted by urologists themselves and treatment schedules are determined accordingly. Therefore, NCCT interpretation skills of urology specialists are very important to make the correct diagnosis and to start treatment quickly for patients presenting with acute flank pain. However, there is no study in the literature investigating the accuracy of diagnosis and interpretations of NCCT results by urology specialists. In order to evaluate this case, we compared the radiologists’ reports with the results of NCCT images interpreted by urology specialists for patients who presented with acute flank pain.
Table 2. The statistical analyses of presence of stone, presence of hydronephrosis, the categorical stone size, the stone localization and pathologies other than calculus out of interpretations by both urologists and radiologists
Presence of stone Hydronephrosis Stone size Stone localization Other pathologies
Kappa ± SD
PPV, % (95% CI)
NPV, % (95% CI)
Sensitivity, % (95% CI)
Specificity, % (95% CI)
0.904±0.01 0.940±0.04 0.816±0.02 0.880±0.17 0.376±0.02
97.7 (95–99) 95.9 (91–99)
92.4 (88–93) 98.3 (96–99)
93.4 (90–96) 95.1 (90–98)
97.3 (94–99) 98.5 (97–99)
SD = Standard deviation; PPV = positive predictive value; NPV = negative predictive value; CI = confidence interval.
212
Urol Int 2015;94:210–214 DOI: 10.1159/000370247
pathologies in the urinary system as well as for those outside of the urinary system is calculated as 0, and the false-negative rate is calculated as 0.52. The most common pathology detected in the urinary system other than calculus is the renal cyst, which was observed in 80 patients. The most common pathologies detected outside of urinary systems other than calculus are lymphadenopathy in 34 patients and umbilical hernia in 26 patients. In our study, there was no pathologic finding that needed immediate treatment. Twenty-five (5%) patients needed deferred treatment, and 17.6% of the pathologies had clinical importance.
Discussion
A rapid and accurate diagnosis can be achieved through non-contrast computerized tomography imaging in patients who present to the emergency services or the urology department with acute flank pain. Nowadays, this imaging method is accepted as a gold standard diagnosis method in the detection of calculus diseases in urinary systems by using low doses; this method has taken the place of conventional diagnostic methods such as intravenous urography (IVU) [9, 10]. It offers advantages over IVU of avoiding contrast allergy or nephrotoxicity and has the ability to determine other causes of abdominal pain [11]. However, despite all these advantages, this diagnosis arrives later mostly due to the reports of the radiologists. The delays in radiological evaluations often cause urologists to make their own decisions on treatment in accordance with their own interpretations. The most important step for patients presenting with acute flank pain is to decide whether there is a stone or not. As a result of this detection, many unnecessary acSancak et al.
Downloaded by: NYU Medical Center Library 198.143.38.97 - 6/18/2015 9:40:12 AM
The average stone size was measured as 5.04 ± 6.26 mm according to the evaluations of the radiologists, whereas it was measured as 4.79 ± 6.03 mm by the urologists. When the stone sizes are categorized into 3 groups, the kappa value is calculated as 0.81, which is the scale of the urologists to calculate the size of the stone accurately. The sensitivity, specificity, NPV, PPV, and kappa values are calculated as follows: 95.1, 98.5, 98.3, 95.9, and 0.94%, respectively, for urologists detecting the presence of hydronephrosis. The kappa value is 0.88 for the accurate detection of stone location. The average Hounsfield unit of the stone was measured as 752 (350–1,335) according to the evaluations of the radiologists, whereas it was measured as 765 (322–1,350) by the urologists (p > 0.05). In addition to the presence of the stone, the statistical analyses of categorical stone size, presence of hydronephrosis, and the stone location diagnosed by both urologists and radiologists are given in table 2. Other pathologies apart from calculus were detected in 262 (52.4%) patients: 39.1% of these patients were also diagnosed with a stone, whereas 68.2% of these patients were not diagnosed with a stone. The kappa value is calculated as 0.37 to detect the pathologies other than a stone, and it is interpreted as moderately consistent. Although the urology specialists found no pathologies other than calculus in 352 patients as a result of their evaluations, 25 of these (7.1%) patients had pathologies in their urinary systems, whereas 91 (25.8%) had pathologies outside of their urinary systems and 10 (2.8%) had pathologies in their urinary tracts as well as outside of their urinary tracts. The false-positive rate for the pathologies in the urinary system is 0, and the false-negative rate is 0.39. The false-positive rate for the pathologies outside of the urinary system is 0.05, and the falsenegative rate is 0.86. The false-positive rate for
The only similar study that we can find in the literature was conducted by Rafi et al., which compared the CT evaluations of emergency service specialists and radiology specialists in patients with flank pain [12]. In this study, CT images of 20 patients were evaluated by 10 emergency specialists. Evaluating the same 20 patients is positive in terms of the standardization of the images, but the number of the pathologies other than calculus is limited to 15. Since the number of patients was low, some pathologies were not included in the study, such as complicated renal cysts, adrenal adenoma, appendicitis, or ovarian cysts, as well as many others. The discordant rate increases in accordance with the increasing pathology rate in the evaluated CT images [14]. Therefore, our study aimed at achieving adequate sampling with 500 patients. The major problem with NCCT is high radiation exposure. Radiation risk of NCCT can be reduced by a low-dose protocol [9]. Using the low-dose technique offers a 56% decrease in the patient’s radiation exposure associated with a calculated effective dose of 2–2.5 mSv [15]. The specificity and sensitivity of low-dose CT were approximately 95% and 97%, respectively, in a metaanalysis of prospective studies [16]. However, this protocol may be associated with difficulties in the diagnosis of pathologies apart from stone detection; therefore, its effectiveness needs to be assessed in a prospective study. Although urology specialists are quite successful in the evaluation of parameters related to renal stones, they are quite inadequate in detecting pathologies other than renal stones (kappa value: 0.37). Especially, the false-negative rate seems to be quite high at 0.86 in the evaluation of systems other than the urinary system. When it is considered that the ratio of the pathology detections other than renal stones is 68% in patients who present with acute flank pain, the necessary consultations and the opinion of a radiologist are required for this group in particular. It is also necessary to ensure there is no other underlying pathology in patients diagnosed with renal stones. In addition to our study, further clinical evaluations were made for 24 patients for whom there was no consensus on the presence or absence of stones. Sixteen of these patients were evaluated, and the presence of the stone was proved. Accurate diagnoses were made in 14 out of these 16 patients by radiology specialists, whereas 2 of them were diagnosed accurately by the urology specialists. Since we have taken the evaluations of radiology specialists as a standard reference, these results are noted as independent from our study and as additional information.
Accuracy of Unenhanced CT Interpretation in Patients with Acute Flank Pain
Urol Int 2015;94:210–214 DOI: 10.1159/000370247
213
Downloaded by: NYU Medical Center Library 198.143.38.97 - 6/18/2015 9:40:12 AM
tions can be avoided by a preliminary diagnosis of acute pain. In our study, there was a consensus on the presence of stones in 476 out of 500 patients who were evaluated by both radiology and urology specialists. The positive predictive value of the evaluations made by the urology specialists is measured as 97.7%, whereas the negative predictive value is measured as 92.4%. Although our study is the only one in the literature focusing on this subject for urologists, in a study conducted by Rafi et al., in which the CT evaluation results of emergency service specialists were evaluated, PPD: 99%, NPD: 89% was found for the stones larger than 5 mm, and PPD: 84%, NPD: 77% was found for the stones smaller than 5 mm [12]. These results are consistent with the results of urologists. If the cause of acute flank pain is urolithiasis, the two most important factors in making treatment decisions are the size of the stone and the location. Although 68% of the stones with a size of 4 mm or smaller can pass spontaneously, it is not recommended to just monitor the stones larger than 1 cm. Therefore, the dimensions of the stones are categorized as less than 4 mm, between 5 and 9 mm, and greater than 1 cm. While making a decision about treatment, the size and localization of the stone affects the decision process as well as the presence of hydronephrosis. In our study, the kappa value was calculated as 0.8, which is the indicator of accurate interpretation by urologists based on these 3 parameters. This value shows that in cases with a stone, the urology specialists can make the correct treatment decisions by evaluating the NCCT results accurately. It is not appropriate to only use NCCT results for stone detection in patients presenting with acute side pain. It is important not to ignore pathologies other than calculus that can act similar to stones, such as aneurysm rupture, ovarian torsion, acute appendicitis, and acute cholecystitis. Hoppe et al. evaluated the additional findings accompanying urolithiasis or producing similar results in patients with acute side pain [13]. They found additional or different pathologies in 71% of all patients. They found that 6% of the pathologies needed immediate treatment; 8% of the pathologies needed deferred treatment, 31% of the pathologies had little clinical importance; and 26% of the pathologies had no clinical importance. In our study, the rate of pathology apart from calculus was calculated as 53% for all patients. The tomography findings are not classified, since the groups with and without clinical significance would be subjective, and emergencies can vary from patient to patient.
This case constitutes the most important limitation of our study. Urology specialists are quite successful at evaluating parameters that are either related or not related to calculus in the NCCT interpretations of patients with
acute flank pain. However, if there is no stone detected, radiology specialist assistance should be obtained in the determination of pathologies that might not be detected by urologists but can cause similar clinical results as stones.
References
214
Urol Int 2015;94:210–214 DOI: 10.1159/000370247
6 Shakhssalim N, Basiri A, Houshmand M, et al: Genetic polymorphisms in calcitonin receptor gene and risk for recurrent kidney calcium stone disease. Urol Int 2014;92:356–362. 7 Heidenreich A, Desgrandschamps F, Terrier F: Modern approach of diagnosis and management of acute flank pain: review of all imaging modalities. Eur Urol 2002;41:351–362. 8 Portis AJ, Sundaram CP: Diagnosis and initial management of kidney stones. Am Fam Physician 2001;63:1329–1338. 9 Turk C, Knoll T, Petrik A, Sarica K, Skolarikos A, Straub M, Seitz C; members of the European Association of Urology (EAU) Guidelines Office: European Association of Urology Guidelineson Urolithiasis. 28. Milano, EAU Annual Congress, 2013. 10 Hale Z, Hanna E, Miyake M, et al: Imaging the urologic patient: the utility of intravenous pyelogram in the CT scan era. World J Urol 2014;32:137–142. 11 Resorlu M, Adam G, Uysal F, et al: Re: Imaging the urologic patient: the utility of intravenous pyelogram in the CT scan era. World J Urol 2014;32:837.
12 Rafi M, Shetty A, Gunja N: Accuracy of computed tomography of the kidneys, ureters and bladder interpretation by emergency physicians. Emerg Med Australas 2013;25:422–426. 13 Hoppe H, Studer R, Kessler TM, et al: Alternate or additional findings to stone disease on unenhanced computerized tomography for acute flank pain can impact management. J Urol 2006;175:1725–1730. 14 Arendts G, Manovel A, Chai A: Cranial CT interpretation by senior emergency department staff. Australas Radiol 2003;47:368–374. 15 Zilberman DE, Tsivian M, Lipkin ME, et al: Low dose computerized tomography for detection of urolithiasis – its effectiveness in the setting of the urology clinic. J Urol 2011;185: 910–914. 16 Niemann T, Kollmann T, Bongartz G: Diagnostic performance of low-dose CT for the detection of urolithiasis: a meta-analysis. AJR Am J Roentgenol 2008;191:396–401.
Sancak et al.
Downloaded by: NYU Medical Center Library 198.143.38.97 - 6/18/2015 9:40:12 AM
1 West B, Luke A, Durazo-Arvizu RA, et al: Metabolic syndrome and self-reported history of kidney stones: the National Health and Nutrition Examination Survey (NHANES III) 1988–1994. Am J Kidney Dis 2008; 51: 741–747. 2 Scales CD Jr, Smith AC, Hanley JM, et al; Urologic Diseases in America Project: Prevalence of kidney stones in the United States. Eur Urol 2012;62:160–165. 3 Romero V, Akpinar H, Assimos DG: Kidney stones: a global picture of prevalence, incidence, and associated risk factors. Rev Urol 2010;12:e86–e96. 4 Lieske JC, Peña de la Vega LS, Slezak JM, et al: Renal stone epidemiology in Rochester, Minnesota: an update. Kidney Int 2006; 69: 760– 764. 5 Wang S, Zhang Y, Mao Z, et al: A meta-analysis of coffee intake and risk of urolithiasis. Urol Int 2014;93:220–228.
Urologia
Received: September 29, 2014 Accepted after revision: November 28, 2014 Published online: January 29, 2015
Urol Int 2015;94:210–214 DOI: 10.1159/000370247
Internationalis
Accuracy of Unenhanced Computerized Tomography Interpretation by Urologists in Patients with Acute Flank Pain Eyup Burak Sancak a Mustafa Resorlu b Orcun Celik c Berkan Resorlu a Murat Tolga Gulpinar a Alpaslan Akbas a Tolga Karakan d Omer Bayrak e Mucahit Kabar d Muzaffer Eroglu d Huseyin Ozdemir b
Departments of a Urology and b Radiology, Canakkale Onsekiz Mart University, Faculty of Medicine, Canakkale, c Department of Urology, Tepecik Training and Research Hospital, Izmir, d Department of Urology, Ankara Training and Research Hospital, and e Department of Urology, Gazi University, Faculty of Medicine, Ankara, Turkey
Abstract Purpose: The aim was to compare the findings of non-contrast computerized tomography (NCCT) evaluated by urology specialists with the findings of experienced radiologists, who are accepted as a standard reference for patients who present with acute flank pain. Materials and Methods: Five hundred patients evaluated with NCCT were included in the study. The NCCT images of these patients were evaluated by both radiologists and urology specialists in terms of the presence of calculus, size of calculus, the location of calculus, the presence of hydronephrosis, and pathologies other than calculus, and the results were compared. Results: The evaluations of urology specialists and standard reference radiology specialists are consistent with each other in terms of the presence of calculus (kappa [κ]: 0.904), categorical stone size (κ: 0.81), the location of calculus (κ: 0.88), and hydronephrosis (κ: 0.94). However, the evaluations of urology specialists in detecting pathologies other than calculus, which may cause acute flank pain or accompany renal colic, were found to be inadequate (κ: 0.37). The false-negative rate of detect-
© 2015 S. Karger AG, Basel 0042–1138/15/0942–0210$39.50/0 E-Mail [email protected] www.karger.com/uin
ing pathologies outside of the urinary system by the urology specialists is calculated as 0.86. Conclusion: Although the urology specialists can evaluate the findings related to calculus sufficiently with NCCT, they may not discover pathologies outside of the urinary system. © 2015 S. Karger AG, Basel
Introduction
The incidence and prevalence of urinary tract calculus disease is increasing worldwide, and the number of patients applying to emergency services due to renal colic is also increasing accordingly [1–6]. However, many urological and non-urinary disorders (appendicitis, ovarian torsion, epididymo-orchitis, diverticulitis, cholecystitis) apart from calculus disease are associated with acute flank pain [7, 8]. It is very important that diagnosis be quick and accurate for these patients, since a significant number of these disorders may require emergency treatment. In the current guidelines, non-contrast computerized tomography (NCCT) is recommended for the diagnosis of patients presenting with acute flank pain as the primary imaging method, which has a high sensitivity and specificity, and also the capability of detecting patholoBerkan Resorlu, MD Canakkale Onsekiz Mart Universitesi Terzioglu Yerleskesi, Barbaros Mh TR–17100 Canakkale (Turkey) E-Mail drberkan79 @ gmail.com
Downloaded by: NYU Medical Center Library 198.143.38.97 - 6/18/2015 9:40:12 AM
Key Words Acute flank pain · Computed tomography · Radiologist · Urologist
Patients and Methods Study Population The data of 579 patients, presenting to either the Emergency Service or Urology Department between January 2013 and January 2014 and assessed by NCCT, were evaluated retrospectively. However, since more than one stone was detected in 67 patients, some defects were detected in the images of 10 patients and some clinical data were missing in 2 patients; a total of 79 patients were excluded from the study, and the study comprised 500 patients. The NCCT images were evaluated by 2 experienced radiology specialists, and the results were accepted as a reference. Afterward, the images were evaluated by 5 experienced urology specialists, with each specialist interpreting 100 images. The main purpose was to compare the findings of the radiology specialists, which are considered the standard reference, and the interpretations of the urology specialists in terms of presence of stone, size of stone, location of stone, presence of hydronephrosis, and detecting pathologies other than calculus. No additional clinical information was given to either urology or radiology specialists, other than the side of the pain. The radiology specialists evaluated the images in a quiet and darkened environment on a 27-inch screen, and they used the coronal images if needed. Urology specialists examined only the images with transverse cross-sections on a 23-inch screen under proper clinical conditions and in a bright environment. The images were evaluated by both radiology and urology specialists in terms of the presence of the stone, size of the stone, categorical stone size (≤4, 5–9, ≥10 mm), location of the calculus (lower ureter [lower cross iliac], upper ureter [above cross iliac], renal pelvis, lower pole, medium pole, upper pole), presence of hydronephrosis, and presence of pathologies other than calculus, and the results were compared with each other. If there was no consensus on the presence or absence of the stone, a clinical examination was conducted. A number was given to every pathology other than calculus, and this numeric system was used for further evaluations. In addition, the pathologies other than calculus were categorized into three groups as follows: pathologies associated with the urinary tract; pathologies not related to the urinary tract; and the pathologies outside of both the urinary tract and urinary system.
Accuracy of Unenhanced CT Interpretation in Patients with Acute Flank Pain
Table 1. Demographic data and stone characteristics
Number of patients Mean age ± standard deviation, years Gender ratio Male Female Number with stone Stone laterality Right Left Stone location Upper pole Middle pole Lower pole Renal pelvis Distal ureter Proximal ureter Stone size, mm Stone size (categorical) ≤4 mm 5–9 mm 10 mm Hydronephrosis Yes No
500 47.8±16.5 59.8 40.2 273 52.8 47.2 14 (5.1%) 36 (13.1%) 72 (26.3%) 25 (9.1%) 75 (27.4%) 51 (18.6%) 5.04±6.2 50 127 96 123 377
Results
The study consisted of 299 male patients (59.8%) and 201 female patients (40.2%), which was 500 patients in total. The average age was 47.8 ± 16.5 (between 3–82 years old), and 99% were older than the age of 18. Stones were detected in 273 (54.6%) patients by the radiologists, in which the average size of the stone detected was 5.04 ± 6.2 mm. Looking at the localization of the stones; 14 (5.1%) were in the upper pole, 36 (13.1%) were in the middle pole, 72 (26.3%) were in the lower pole, 25 (9.1%) were in the renal pelvis, 51 (18.6%) were in the proximal ureter, and, finally, 75 (27.4%) were located in the distal ureter. The demographic data of the patients and characteristics of the calculi are summarized in table 1. There was a consensus for 476 patients in terms of the presence or absence of stone as a result of the evaluations of both urology and radiology specialists. The radiologists detected no stone in 6 out of 261 patients (2.3%) in whom the urology specialists detected a stone, and the radiologists detected a stone in 18 out of 239 patients (7.5%) in whom the urology specialists did not detect a stone (sensitivity 93.4%, specificity: 97.4%; negative predictive value [NPV]: 92.4%, positive predictive value [PPV]: 97.7%). The kappa value is calculated as 0.904 for this evaluation, which is described as a near-perfect match. Urol Int 2015;94:210–214 DOI: 10.1159/000370247
211
Downloaded by: NYU Medical Center Library 198.143.38.97 - 6/18/2015 9:40:12 AM
gies other than calculus [9]. However, despite providing quick detection with this technique, the radiologist’s report cannot be obtained so quickly. In this waiting period, NCCT scans are often interpreted by urologists themselves and treatment schedules are determined accordingly. Therefore, NCCT interpretation skills of urology specialists are very important to make the correct diagnosis and to start treatment quickly for patients presenting with acute flank pain. However, there is no study in the literature investigating the accuracy of diagnosis and interpretations of NCCT results by urology specialists. In order to evaluate this case, we compared the radiologists’ reports with the results of NCCT images interpreted by urology specialists for patients who presented with acute flank pain.
Table 2. The statistical analyses of presence of stone, presence of hydronephrosis, the categorical stone size, the stone localization and pathologies other than calculus out of interpretations by both urologists and radiologists
Presence of stone Hydronephrosis Stone size Stone localization Other pathologies
Kappa ± SD
PPV, % (95% CI)
NPV, % (95% CI)
Sensitivity, % (95% CI)
Specificity, % (95% CI)
0.904±0.01 0.940±0.04 0.816±0.02 0.880±0.17 0.376±0.02
97.7 (95–99) 95.9 (91–99)
92.4 (88–93) 98.3 (96–99)
93.4 (90–96) 95.1 (90–98)
97.3 (94–99) 98.5 (97–99)
SD = Standard deviation; PPV = positive predictive value; NPV = negative predictive value; CI = confidence interval.
212
Urol Int 2015;94:210–214 DOI: 10.1159/000370247
pathologies in the urinary system as well as for those outside of the urinary system is calculated as 0, and the false-negative rate is calculated as 0.52. The most common pathology detected in the urinary system other than calculus is the renal cyst, which was observed in 80 patients. The most common pathologies detected outside of urinary systems other than calculus are lymphadenopathy in 34 patients and umbilical hernia in 26 patients. In our study, there was no pathologic finding that needed immediate treatment. Twenty-five (5%) patients needed deferred treatment, and 17.6% of the pathologies had clinical importance.
Discussion
A rapid and accurate diagnosis can be achieved through non-contrast computerized tomography imaging in patients who present to the emergency services or the urology department with acute flank pain. Nowadays, this imaging method is accepted as a gold standard diagnosis method in the detection of calculus diseases in urinary systems by using low doses; this method has taken the place of conventional diagnostic methods such as intravenous urography (IVU) [9, 10]. It offers advantages over IVU of avoiding contrast allergy or nephrotoxicity and has the ability to determine other causes of abdominal pain [11]. However, despite all these advantages, this diagnosis arrives later mostly due to the reports of the radiologists. The delays in radiological evaluations often cause urologists to make their own decisions on treatment in accordance with their own interpretations. The most important step for patients presenting with acute flank pain is to decide whether there is a stone or not. As a result of this detection, many unnecessary acSancak et al.
Downloaded by: NYU Medical Center Library 198.143.38.97 - 6/18/2015 9:40:12 AM
The average stone size was measured as 5.04 ± 6.26 mm according to the evaluations of the radiologists, whereas it was measured as 4.79 ± 6.03 mm by the urologists. When the stone sizes are categorized into 3 groups, the kappa value is calculated as 0.81, which is the scale of the urologists to calculate the size of the stone accurately. The sensitivity, specificity, NPV, PPV, and kappa values are calculated as follows: 95.1, 98.5, 98.3, 95.9, and 0.94%, respectively, for urologists detecting the presence of hydronephrosis. The kappa value is 0.88 for the accurate detection of stone location. The average Hounsfield unit of the stone was measured as 752 (350–1,335) according to the evaluations of the radiologists, whereas it was measured as 765 (322–1,350) by the urologists (p > 0.05). In addition to the presence of the stone, the statistical analyses of categorical stone size, presence of hydronephrosis, and the stone location diagnosed by both urologists and radiologists are given in table 2. Other pathologies apart from calculus were detected in 262 (52.4%) patients: 39.1% of these patients were also diagnosed with a stone, whereas 68.2% of these patients were not diagnosed with a stone. The kappa value is calculated as 0.37 to detect the pathologies other than a stone, and it is interpreted as moderately consistent. Although the urology specialists found no pathologies other than calculus in 352 patients as a result of their evaluations, 25 of these (7.1%) patients had pathologies in their urinary systems, whereas 91 (25.8%) had pathologies outside of their urinary systems and 10 (2.8%) had pathologies in their urinary tracts as well as outside of their urinary tracts. The false-positive rate for the pathologies in the urinary system is 0, and the false-negative rate is 0.39. The false-positive rate for the pathologies outside of the urinary system is 0.05, and the falsenegative rate is 0.86. The false-positive rate for
The only similar study that we can find in the literature was conducted by Rafi et al., which compared the CT evaluations of emergency service specialists and radiology specialists in patients with flank pain [12]. In this study, CT images of 20 patients were evaluated by 10 emergency specialists. Evaluating the same 20 patients is positive in terms of the standardization of the images, but the number of the pathologies other than calculus is limited to 15. Since the number of patients was low, some pathologies were not included in the study, such as complicated renal cysts, adrenal adenoma, appendicitis, or ovarian cysts, as well as many others. The discordant rate increases in accordance with the increasing pathology rate in the evaluated CT images [14]. Therefore, our study aimed at achieving adequate sampling with 500 patients. The major problem with NCCT is high radiation exposure. Radiation risk of NCCT can be reduced by a low-dose protocol [9]. Using the low-dose technique offers a 56% decrease in the patient’s radiation exposure associated with a calculated effective dose of 2–2.5 mSv [15]. The specificity and sensitivity of low-dose CT were approximately 95% and 97%, respectively, in a metaanalysis of prospective studies [16]. However, this protocol may be associated with difficulties in the diagnosis of pathologies apart from stone detection; therefore, its effectiveness needs to be assessed in a prospective study. Although urology specialists are quite successful in the evaluation of parameters related to renal stones, they are quite inadequate in detecting pathologies other than renal stones (kappa value: 0.37). Especially, the false-negative rate seems to be quite high at 0.86 in the evaluation of systems other than the urinary system. When it is considered that the ratio of the pathology detections other than renal stones is 68% in patients who present with acute flank pain, the necessary consultations and the opinion of a radiologist are required for this group in particular. It is also necessary to ensure there is no other underlying pathology in patients diagnosed with renal stones. In addition to our study, further clinical evaluations were made for 24 patients for whom there was no consensus on the presence or absence of stones. Sixteen of these patients were evaluated, and the presence of the stone was proved. Accurate diagnoses were made in 14 out of these 16 patients by radiology specialists, whereas 2 of them were diagnosed accurately by the urology specialists. Since we have taken the evaluations of radiology specialists as a standard reference, these results are noted as independent from our study and as additional information.
Accuracy of Unenhanced CT Interpretation in Patients with Acute Flank Pain
Urol Int 2015;94:210–214 DOI: 10.1159/000370247
213
Downloaded by: NYU Medical Center Library 198.143.38.97 - 6/18/2015 9:40:12 AM
tions can be avoided by a preliminary diagnosis of acute pain. In our study, there was a consensus on the presence of stones in 476 out of 500 patients who were evaluated by both radiology and urology specialists. The positive predictive value of the evaluations made by the urology specialists is measured as 97.7%, whereas the negative predictive value is measured as 92.4%. Although our study is the only one in the literature focusing on this subject for urologists, in a study conducted by Rafi et al., in which the CT evaluation results of emergency service specialists were evaluated, PPD: 99%, NPD: 89% was found for the stones larger than 5 mm, and PPD: 84%, NPD: 77% was found for the stones smaller than 5 mm [12]. These results are consistent with the results of urologists. If the cause of acute flank pain is urolithiasis, the two most important factors in making treatment decisions are the size of the stone and the location. Although 68% of the stones with a size of 4 mm or smaller can pass spontaneously, it is not recommended to just monitor the stones larger than 1 cm. Therefore, the dimensions of the stones are categorized as less than 4 mm, between 5 and 9 mm, and greater than 1 cm. While making a decision about treatment, the size and localization of the stone affects the decision process as well as the presence of hydronephrosis. In our study, the kappa value was calculated as 0.8, which is the indicator of accurate interpretation by urologists based on these 3 parameters. This value shows that in cases with a stone, the urology specialists can make the correct treatment decisions by evaluating the NCCT results accurately. It is not appropriate to only use NCCT results for stone detection in patients presenting with acute side pain. It is important not to ignore pathologies other than calculus that can act similar to stones, such as aneurysm rupture, ovarian torsion, acute appendicitis, and acute cholecystitis. Hoppe et al. evaluated the additional findings accompanying urolithiasis or producing similar results in patients with acute side pain [13]. They found additional or different pathologies in 71% of all patients. They found that 6% of the pathologies needed immediate treatment; 8% of the pathologies needed deferred treatment, 31% of the pathologies had little clinical importance; and 26% of the pathologies had no clinical importance. In our study, the rate of pathology apart from calculus was calculated as 53% for all patients. The tomography findings are not classified, since the groups with and without clinical significance would be subjective, and emergencies can vary from patient to patient.
This case constitutes the most important limitation of our study. Urology specialists are quite successful at evaluating parameters that are either related or not related to calculus in the NCCT interpretations of patients with
acute flank pain. However, if there is no stone detected, radiology specialist assistance should be obtained in the determination of pathologies that might not be detected by urologists but can cause similar clinical results as stones.
References
214
Urol Int 2015;94:210–214 DOI: 10.1159/000370247
6 Shakhssalim N, Basiri A, Houshmand M, et al: Genetic polymorphisms in calcitonin receptor gene and risk for recurrent kidney calcium stone disease. Urol Int 2014;92:356–362. 7 Heidenreich A, Desgrandschamps F, Terrier F: Modern approach of diagnosis and management of acute flank pain: review of all imaging modalities. Eur Urol 2002;41:351–362. 8 Portis AJ, Sundaram CP: Diagnosis and initial management of kidney stones. Am Fam Physician 2001;63:1329–1338. 9 Turk C, Knoll T, Petrik A, Sarica K, Skolarikos A, Straub M, Seitz C; members of the European Association of Urology (EAU) Guidelines Office: European Association of Urology Guidelineson Urolithiasis. 28. Milano, EAU Annual Congress, 2013. 10 Hale Z, Hanna E, Miyake M, et al: Imaging the urologic patient: the utility of intravenous pyelogram in the CT scan era. World J Urol 2014;32:137–142. 11 Resorlu M, Adam G, Uysal F, et al: Re: Imaging the urologic patient: the utility of intravenous pyelogram in the CT scan era. World J Urol 2014;32:837.
12 Rafi M, Shetty A, Gunja N: Accuracy of computed tomography of the kidneys, ureters and bladder interpretation by emergency physicians. Emerg Med Australas 2013;25:422–426. 13 Hoppe H, Studer R, Kessler TM, et al: Alternate or additional findings to stone disease on unenhanced computerized tomography for acute flank pain can impact management. J Urol 2006;175:1725–1730. 14 Arendts G, Manovel A, Chai A: Cranial CT interpretation by senior emergency department staff. Australas Radiol 2003;47:368–374. 15 Zilberman DE, Tsivian M, Lipkin ME, et al: Low dose computerized tomography for detection of urolithiasis – its effectiveness in the setting of the urology clinic. J Urol 2011;185: 910–914. 16 Niemann T, Kollmann T, Bongartz G: Diagnostic performance of low-dose CT for the detection of urolithiasis: a meta-analysis. AJR Am J Roentgenol 2008;191:396–401.
Sancak et al.
Downloaded by: NYU Medical Center Library 198.143.38.97 - 6/18/2015 9:40:12 AM
1 West B, Luke A, Durazo-Arvizu RA, et al: Metabolic syndrome and self-reported history of kidney stones: the National Health and Nutrition Examination Survey (NHANES III) 1988–1994. Am J Kidney Dis 2008; 51: 741–747. 2 Scales CD Jr, Smith AC, Hanley JM, et al; Urologic Diseases in America Project: Prevalence of kidney stones in the United States. Eur Urol 2012;62:160–165. 3 Romero V, Akpinar H, Assimos DG: Kidney stones: a global picture of prevalence, incidence, and associated risk factors. Rev Urol 2010;12:e86–e96. 4 Lieske JC, Peña de la Vega LS, Slezak JM, et al: Renal stone epidemiology in Rochester, Minnesota: an update. Kidney Int 2006; 69: 760– 764. 5 Wang S, Zhang Y, Mao Z, et al: A meta-analysis of coffee intake and risk of urolithiasis. Urol Int 2014;93:220–228.
![[Emergency checklist: acute flank pain].](/assets/img/hold.png)
