validation of emergency physician ultrasound in diagnosing hydronephrosis in ureteric colic
TRANSCRIPT
Emergency Medicine Australasia (2007) 19 188ndash195 doi 101111j1742-6723200700925x
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
Blackwell Publishing AsiaMelbourne AustraliaEMMEmergency Medicine Australasia1742-6731copy 2006 The Authors Journal compilation copy 2006 Australasian College for Emergency Medicine and Australasian Society forEmergency Medicine2007193188195Xxxx XxxxHydronephrosis studyS Watkins
et al
Correspondence Dr Stuart Watkins Department of Emergency Medicine Liverpool Hospital Locked Bag 7103 Liverpool BC NSW 1871 Australia Email stuartwatkinsswsahsnswgovau
Stuart Watkins MBChB Registrar Justin Bowra MB BS FACEM Staff Specialist Director of Emergency Medicine Training Praneal SharmaMB BS FRANZCR Director Anna Holdgate MB BS FACEM MMed Director of Emergency Medicine Research Unit Alan Giles MB BSFACEM Senior Staff Specialist Lewis Campbell MBChB Registrar
ORIGINAL RESEARCH
Validation of emergency physician ultrasound in diagnosing hydronephrosis in ureteric colicStuart Watkins1 Justin Bowra12 Praneal Sharma3 Anna Holdgate12 Alan Giles12 and Lewis Campbell1
Departments of 1Emergency Medicine and 3Radiology Liverpool Hospital and 2Conjoint University of New South Wales Sydney New South Wales Australia
Abstract
Objective Patients presenting to the ED with obstructive nephropathies benefit from early detectionof hydronephrosis Out of hours radiological imaging is expensive and disruptive toarrange Emergency physician ultrasound (EPU) could allow rapid diagnosis and dispo-sition If accurate it might avert the need for formal radiological imaging exclude anobstruction and improve patient flow through the ED
Methods This was a prospective study of a convenience sample of all adult non-pregnant patientswith presumed ureteric colic attending the ED with prior ethics committee approval Anemergency physician or registrar performed a focused ultrasound scan and were blindedto the patientrsquos other management A computerized tomography scan was also performedfor all patients while in the ED or within 24 h of the EPU The accuracy of EPU detectionof hydronephrosis was determined using computerized tomography scans reported by asenior radiologist as the lsquogold-standardrsquo
Results Sixty-three patients with suspected ureteric colic were enrolled of whom 57 completed bothEPU and computerized tomography imaging Forty-nine had confirmed nephrolithiasis bycomputerized tomography with 39 having evidence of hydronephrosis Overall prevalenceof hydronephrosis was 68 (95 confidence interval [CI] 56ndash79) compared with com-puterized tomography EPU had a sensitivity of 80 (95 CI 65ndash89) specificity of 83(95 CI 61ndash94) positive predictive value of 91 (95 CI 75ndash98) and negativepredictive value of 65 (95 CI 43ndash83) The overall accuracy was 81 (95 CI 69ndash89)
Conclusion Although the accuracy of detection of hydronephrosis after focused training in EPU isencouraging further experience and training might improve the accuracy of EPU andallow its use as a screening tool
Key words computerized tomography emergency physician hydronephrosis renal colic ultrasound
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
189
Introduction
Patients presenting to the ED with an obstructive neph-ropathy benefit from early detection and treatment par-ticularly in the context of associated renal tract infectionor renal failure Complete ureteric obstruction mightlead to loss of renal function with an increased occur-rence of irreversible damage after 1ndash2 weeks1 Conven-tional modalities available for detecting renal tractobstruction include formal ultrasound (US) and comput-erized tomography (CT) However both CT and formalUS require patient transfer and monitoring in areasremote to the ED and generally have limited availabil-ity out of hours
The detection of hydronephrosis by bedside US in theED could allow a more rapid diagnosis and dispositionof patients and be used in situations where iodizingradiation or intravenous contrast material are contrain-dicated andor formal US is not available Rapid EDscreening for hydronephrosis might exclude obstruc-tion and focus assessment on other potential diagnosesand might allow the selection of patients for furtherstudies to be refined
An estimated 2ndash5 of the population will form asymptomatic renal calculus at some point in their lives1
Clinical history and microscopic haematuria suggestthe diagnosis with a sensitivity of 69ndash892ndash4 CT hasproved a more accurate test than US with greater sen-sitivity for hydronephrosis and calculus detection5ndash10
With intravenous contrast it can give the same infor-mation on renal function as IVU12568911 but also carriesthe same risks of radiation exposure allergy and neph-rotoxicity7 The advantages of CT have made US asecond choice investigation because of its relatively lowsensitivity of 19 (specificity of 97) for detectingcalculi compared with sensitivity of 94ndash97 (specificityof 96ndash97) for CT68 However patients with uretericstones have a relatively high prevalence of partial andor temporary renal tract obstruction but seldom causecomplete obstruction1 Because of this high prevalencethis population provides a convenient group in whomto explore options for diagnosing obstruction and thesepatients will require definitive renal tract imaging aspart of their usual management
The use of emergency physician US (EPU) for trau-matic intraperitoneal and pericardial fluid and abdom-inal aortic aneurysm detection has been increasing inAustralasian ED with accreditation processes thatfollow the Australasian College of Emergency Medi-cine (ACEM) guidelines12ndash14 Internationally EPU hasexpanded into areas of ED practice such as intravenous
access lower limb deep venous thrombosis and fluidlocalization for either diagnosis or drainage12ndash14 Therole of EPU in assessing renal tract obstruction remainssomewhat controversial with various studies showinga wide range of accuracy215ndash17
We postulated that EPU might be safe and accurateand could be used to screen patients for hydronephrosisdue to ureteric calculi
Aim
The aim of the present study was to determine theaccuracy of EPU in detection of hydronephrosis com-pared with radiologist-reported non-contrast CT
Methods
Study design and setting
This was a prospective study of a convenience sampleof patients with presumed ureteric colic undertaken inthe ED of a tertiary teaching hospital with an annualcensus of 46 000 The study had Area Ethics committeeapproval
Study population and protocol
All non-pregnant patients over 18 years old whoattended the ED with a clinical diagnosis of uretericcolic as determined by the treating ED medical officerwere eligible for enrolment Written informed consentwas obtained from all patients Foreign language inter-preters were utilized when required
Emergency registrars and emergency physicians (EP)who had attended an ACEM-accredited US workshopand undertaken a further 1 h of training in basic renalUS by a senior radiologist were able to enrol patientsin the study Recruited patients underwent EPU by theenrolling doctor during their ED stay The doctor per-forming the EPU was not involved in the treatment ofthe patient and was blinded to the CT scan result Anon-contrast CT scan of the renal tract also was per-formed on each patient either while they were in theED or organized externally at a single designated privateradiology suite within 24 h of EPU One senior radiolo-gist blinded to the EPU result reported the CT scans
Measurements
Using a Toshiba US machine Model SSA-550A(Tochigi-Ken Japan) and a 35ndash5 MHz curved array
S Watkins et al
190 copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
probe on abdominal preset investigators obtained andrecorded images and measurements of both kidneys(diameters of each renal pelvis and of each kidneyrsquoslongitudinal and transverse sections) Images weresaved digitally and printed Investigators completed areporting form that included demographic data thelocation of symptoms the presence or absence ofhydronephrosis for each kidney and the severity ofhydronephrosis (mild moderate or severe) if presentAn US diagnosis of hydronephrosis was made based onthe features listed in Table 1 Bladder size was esti-mated and documented (as empty half or full)
CT images were performed on a four-slice ToshibaAquilion (TSX-101A) scanner or a 16-slice ToshibaAquilion scanner Information recorded by the radiolo-gist included the presence or absence of hydronephrosisand its severity the presence of calculus and incidentalfindings
Data analysis
The accuracy of the EPU was determined by calculatingthe sensitivity specificity positive (PPV) and negativepredictive values (NPV) with 95 confidence intervals(CI) using the CT scan report as the reference standardAssuming a rate of hydronephrosis of 75 in patientswith renal colic19 we estimated approximately 50patients would be required to detect a sensitivity of90 with 95 CI of plusmn10
Microsoft Excel 2004 for Mac version 112 andVassarstats Statistical Computation website (httpfacultyvassaredulowryVassarStatshtml) were usedfor analysis
Results
Sixty-three patients were enrolled in the study Sixpatients did not complete the study because CT scanswere not performed or were not available for analysisOf the remaining 57 there were 48 men (84) and 9women (16) In the present study 34 (54) patientspresented with left-sided symptoms and 29 (46)
presented with right-sided symptoms The mean agewas 437 years (range 18ndash67) Two patients presentedmore than once during the study period and each pre-sentation was recorded separately
Of the 57 patients 48 had CT-confirmed diagnosis ofnephrolithiasis and 39 had CT-confirmed hydro-nephrosis Nine patients had negative CT scans fornephrolithiasis and hydronephrosis The prevalence ofhydronephrosis in the study population was 3957 or68 (CI 56ndash79)
Thirty-one of 39 patients with CT-proved hydroneph-rosis had positive EPU scans The main results aresummarized in Table 2 EPU demonstrated sensitivityof 80 (CI 65ndash89) and specificity of 83 (CI 61ndash94) a PPV of 91 (CI 75ndash98) and an NPV of 65(CI 43ndash83) Overall accuracy of EPU for the detectionof hydronephrosis was 81 (CI 69ndash89)
The study authors performed 60 of the scans12 other EP or trainees (registrars) performed theremainder
Discussion
The present study found that with minimal training EPand trainees were able to achieve a sensitivity of 80and a specificity of 83 for the diagnosis of hydroneph-rosis using bedside US in the setting of suspected acuterenal colic
Emergency physician US has potential advantages inthe diagnosis of hydronephrosis It can be performed atthe bedside using a portable machine is immediatelyavailable and repeatable 24 h a day 7 days a weekPatients do not leave the department to go to potentiallyless monitored areas which obviates the need for por-table monitoring andor nurse escort EP have been
Table 1 Grades of hydronephrosis
Grade I Grade II ndash mild Grade III ndash moderate Grade IV ndash severe
Slight blunting of calyceal fornices
Obvious blunting of calyceal fornices andenlargement of calices but intruding shadows of papillae are easily seen
Rounding of calices withobliteration of papillae
Extreme calyceal ballooning
Adapted from the study by Grainger and Allison18
Table 2 Comparison of emergency physician US (EPU) andCT in detection of hydronephrosis
CT ndash positive CT ndash negative
EPU ndash positive 31 3EPU ndash negative 8 15
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
191
using bedside US in the ED over recent years and inAustralia have training and accreditation proceduresfor its use in other conditions12ndash14
In the present study we investigated the ability of EPwith 1 h of focused training in renal US in addition toan ACEM-accredited US workshop to detect the pres-ence of hydronephrosis Previous studies have com-pared EPU with IVP215ndash17 or compared radiologist-performed US with CT59 This is the first study thatdirectly compares EPU detection of hydronephrosiswith the current lsquogold-standardrsquo of radiologist-reportedCT scan
Previous studies have demonstrated US sensitivity of85ndash94 and specificity of 100 in detection of hydro-nephrosis when performed by radiologists or sonogra-phers19102021 CT has proved a more accurate test withgreater sensitivity for hydronephrosis and calculusdetection5ndash10
The accuracy reported in our study is comparable tothat previously reported in most other EPU trials Rosenet al with 5 h training compared EPU diagnosis ofhydronephrosis with IVP and CT in 126 patients find-ing a sensitivity of 72 specificity of 73 PPV of85 NPV of 54 and accuracy of 7216 By contrastHenderson et al found a sensitivity of 97 for lsquopathol-ogy consistent with nephro-ureterolithiasis when com-pared to IVPrsquo in 108 patients but did not specificallyreport the detection of hydronephrosis as an outcome2
Lanoix et al reported an accuracy of 94 and sensitiv-ity of 96 after 4 h tuition based on 45 subjects and 39EPtrainees15 However the reference standard used inthat study is unclear
In an Australian study with 3 days of US trainingRowland et al reported 68 accuracy for EPU usingthree grades of hydronephrosis nil subtle or obvious17
They reported a sensitivity of 93 but only a specific-ity of 47 (PPV 59 NPV 89) and used IVP formalUS within 24 h or radiologist review of the EPU as theirlsquogold-standardrsquo Four investigators obtained images
from 31 subjects They reported more false-positivesthan false-negatives whereas our study reports theopposite In the above studies the difference betweenthe US and CT grading were in subjects with low-gradehydronephrosis A comparison of these studies isshown in Table 3
It is worth noting that the sonographic grading ofhydronephrosis into mild moderate or severe correlatespoorly with the clinical severity of disease2223 Hencefor the purposes of data analysis in our study hydro-nephrosis was reported simply as either present orabsent
To be effective as a screening test EPU wouldrequire a high sensitivity (ie few false-negatives) Ourfinding of a sensitivity of 79 and NPV of 65 sug-gests that EPU is currently not an acceptable screeningtest to rule out hydronephrosis However althougheight cases of CT-confirmed hydronephrosis were notdetected by EPU seven of these false-negative scanswere reported as mild hydronephrosis on CT Theeighth patient with false-negative EPU had moderatehydronephrosis demonstrated on CT but this scan wasperformed more than 24 h after the EPU hence it isuncertain whether this truly reflects the presence ofhydronephrosis at the time of the EPU but has beenincorporated into our results for completeness
Radiological diagnosis of hydronephrosis on CT issubjective with several studies reporting inter-observervariability between radiologists and between radiolo-gists trainees and urologists23ndash27 The amount of hydro-nephrosis shown by US varies dynamically with partialobstruction and with hydration status of the patient1125
as hydronephrosis can be induced in healthy volunteerswith forced fluid intake In serial US following hydra-tion mildndashmoderate hydronephrosis was induced in80 of subjects28 Repeating the US in dehydratedpatients following hydration might alter previouslyfalse-negative results11 Studies have demonstrated sim-ilar dynamic changes with CT Perinephric stranding
Table 3 Comparison of the previous studies of emergency physician US and detection of hydronephrosis
Sensitivity()
NPV () Comment
The present study (2005) (n = 57) 80 65 CT only 1 h + courseRosen et al16 (1998) (n = 126) 72 54 IVU + CT 5 h trainingHenderson et al2 (1998) (n = 108) 97 92 IVU unclear diagnostic criteriaLanoix et al15 (2000) (n = 45) 94 94 Multiple reference standards 4 h training 39 investigatorsRowland et al17 (2001) (n = 31) 93 89 Used IVUUSradiologist 3 days training 68 accuracy
NPV negative predictive value
S Watkins et al
192 copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
ureteral dilatation perinephric fluid and collecting sys-tem dilatation showed statistically significant changeover 8 h of study29 Therefore correlation discrepancy inour study might be explained by any of these factorsIt has also been reported that false-negatives on US areusually followed by uncomplicated spontaneous stoneemission3031
What level of minimum training is required to makeEPU an effective screening tool for hydronephrosisLanoix et al15 and Rosen et al16 trained EP for 4 h and5 h respectively with markedly different results asnoted earlier perhaps because of the very different ref-erence standards used in their studies Rowland et aldemonstrated an overall accuracy of 68 for EPU after3 days training whereas our study demonstrates anaccuracy of 81 after 1 h of focused training in renalscans following completion of an ACEM-accreditedworkshop17
From our results it would seem prudent to state thatalthough additional training and experience mightimprove the accuracy of EPU it will not supplant theuse of CT in the foreseeable future However despitesimilar results to ours for EPU accuracy previousauthors have suggested a place for EPU in the detectionof hydronephrosis21516
Australasian and American Colleges for EmergencyMedicine have published policies on the training andaccreditation of EP in focused assessment with sonog-raphy for trauma and abdominal aortic aneurysm12ndash14
However currently there are no guidelines for minimumtraining and accreditation in renal sonography for EP
Limitations
As patients were enrolled on a convenience basis dueto the presence or absence of an investigator to per-form the scans this might have introduced one ormore unknown biases A trend towards improvedinvestigator performance was noted as scan qualityimproved with experience this might have affectedthe detection rate in the earlier stages of the presentstudy Some EP performed less than three studiesothers more than 10 however the sample size was toosmall to afford meaningful subgroup analysis for indi-vidual EP Some of these limitations would be over-come by larger studies
It is worth noting that several of the EPU sono-graphers were relatively inexperienced in the use ofbedside US and were not yet accredited in otherEPU applications such as focused assessment by
sonography in trauma (FAST)abdominal aortic aneu-rysm (AAA) It is possible that accredited EPU sonog-raphers would be more accurate
Because of the dynamic nature of renal colic andurinary obstruction ideally all participants would havehad their EPU and CT scan performed within minimaltime delay to ensure an accurate assessment of EPU inone patient CT scanning was performed more than 24 hlater Finally our study did not include routine evalua-tion of renal resistive indexes that might improve detec-tion of early obstruction28
Conclusion
Using non-contrast CT as the gold standard we havefound EPU detection of hydronephrosis to have anaccuracy of 81 which is comparable to previousstudies However on the basis of the present studyEPU is probably not accurate enough to rule out hydro-nephrosis Further experience and training mightimprove the accuracy of EPU and allow its use as ascreening tool
Acknowledgements
The authors express their thanks to the staff specialistsand registrars from the Departments of EmergencyMedicine and Radiology for their assistance
Author contributions
SW contributed to study design ethics submission con-sent and patient information sheets reporting sheetsdata collation and analysis literature research andmanuscript preparation (85) JB contributed to origi-nal concept study design ethics submission investiga-tion manuscript preparation and supervision (40) PScontributed to investigator training study design andCT reporting (25) AH contributed to research meth-ods data analysis and manuscript preparation (20)AG contributed to study design investigation andmanuscript preparation (10) LC contributed to studydesign investigation and data collation (10)
Competing interests
Justin Bowra is a member of the Ultrasound Committeeof the Australasian College for Emergency MedicineAnna Holdgate holds the position of Section Editor of
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
193
Original Research for Emergency Medicine AustralasiaAll other authors declare no competing interests
Accepted 11 October 2006
References
1 Rosen P Barkin R Ling LJ Emergency Medicine Concepts andClinical Practice 4th edn St Louis Mosby 1998
2 Henderson SO Hoffner RJ Aragona JL Groth DE Esekogwu VIChan D Bedside emergency department ultrasonography plusradiography of the kidneys ureters and bladder versus intrave-nous pyelography in the evaluation of suspected ureteral colicAcad Emerg Med 1998 5 666ndash71
3 Eray O Cubuk MS Oktay CEM Yilmaz S Cete Y Ersoy FFThe efficacy of urinalysis plain films and spiral CT in EDpatients with suspected renal colic Am J Emerg Med 2003 21152ndash4
4 Brown DFM Rosen CL Sagarin M McCabe C Wolfe RE Impactof bedside ultrasonography by emergency physicians on theclinical likelihood of nephrolithiasis Ann Emerg Med 1996 27818
5 Fowler KAB Locken JA Duchesne JH Williamson MR Ultra-sound for detecting renal calculi with non-enhanced CT as areference standard Radiology 2002 222 109ndash13
6 Yilmaz S Sindel T Arslau G et al Renal colic comparison ofspiral CT US and IVU in the detection of ureteral calculi EurRadiol 1998 8 212ndash17
7 Spencer BA Wood BJ Dretler SP Helical CT and ureteric colicUrol Clin North Am 2000 27 231ndash41
8 Smith RC Verga M McCarthy S Rosenfield AT Diagnosis ofacute flank pain value of un-enhanced helical CT AJR Am JRoentgenol 1996 166 97ndash101
9 Sheafor DH Hertzberg BS Freed KS et al Non-enhanced helicalCT and US in the emergency evaluation of patients with renalcolic prospective comparison Radiology 2000 217 792ndash7
10 Patlas M Farkas A Fisher D Zaghal I Hadas-Halpern I Ultra-sound vs CT for the detection of ureteric stones in patients withrenal colic BJR 2001 74 901ndash4
11 Noble VE Brown DFM Renal ultrasound Emerg Med ClinNorth Am 2004 22 641ndash59
12 American College of Emergency Physicians (ACEP) Board ofDirectors Use of Ultrasound Imaging by Emergency PhysiciansPolicy Statement June 2001 Available from URL httpwwwaceporg16840html [Accessed August 2006]
13 Australasian College of Emergency Medicine (ACEM) CouncilCredentialling for ED Ultrasonography Policy Document P22July 2000 Available from URL httpwwwacemorgauinfo-centreaspxdocId=59 [Accessed August 2006]
14 Australasian College of Emergency Medicine (ACEM) CouncilUse of Bedside Ultrasound by Emergency Physicians Policy Doc-ument P21 July 1999 Available from URL httpwwwacemorgauinfocentreaspxdocId=59 [Accessed August 2006]
15 Lanoix R Leak LV Gaeta T Gernsheimer JR A preliminaryevaluation of emergency ultrasound in the setting of an emer-gency medicine training program Am J Emerg Med 2000 1841ndash5
16 Rosen CL Brown DFM Sagarin MJ Chang Y McCabe CJ WolfeRE Ultrasonography by emergency physicians in patients withsuspected renal colic J Emerg Med 1998 16 865ndash70
17 Rowland JL Kuhn M Bonnin RLL Davey MJ Langlois SLAccuracy of emergency department bedside ultrasonographyEmerg Med 2001 13 305ndash13
18 Grainger RG Allison DJ (eds) Diagnostic Radiology A Textbookof Medical Imaging 4th edn London Churchill Livingstone2001 p 1594
19 Kiely EA Hartnell GG Gibson RN Measurement of bladderVolume by real-time ultrasound Br J Urol 1987 60 33ndash5
20 Sinclair D Wilson S Toi A Greenspan L The evaluation ofsuspected renal colic ultrasound scan vs excretory urographyAnn Emerg Med 1989 18 556ndash9
21 Dalla Palma L Stacul F Bazzocchi M et al Ultrasonography andplain film versus intravenous urography in ureteric colic ClinRadiol 1993 47 333ndash6
22 Oumlzden E Karamuumlrsel T Gouml uuml Ccedil Yaman Ouml Inal T Gouml uuml ODetection rate of ureter stones with US relationship with gradeof hydronephrosis J Ankara Med Sch 2002 24 183ndash6
23 King L Hydronephrosis when is obstruction not obstructionCommon problems in paediatric urology Urol Clin North Am1995 22 31ndash42
24 Holdgate A Chan T How accurate are emergency clinicians atinterpreting non-contrast CT for suspected renal colic AcadEmerg Med 2003 10 315ndash19
25 Jeffrey RB Federle MP CT and ultrasonography of acute renalabnormalities Radiol Clin North Am 1983 21 515ndash25
26 Morse JW Saracino BS Melanson SW Arcona S Heller MBUltrasound interpretation of hydronephrosis is improved by abrief educational intervention Ann Emerg Med 1998 32(Suppl Pt 2) S27
27 Freed KS Paulson EK Frederick MG et al Interobserver vari-ability in the interpretation of unenhanced helical CT for thediagnosis of ureteral stones J Comput Assist Tomogr 1998 22732ndash7
28 Morse JW Hill R Greissinger WP Patterson JW Melanson SWHeller MB Rapid oral hydration results in hydronephrosis asdetermined by bedside ultrasound Ann Emerg Med 1999 34134ndash40
29 Varanelli MJ Coll DM Levine JA Rosenfield AT Smith RCRelationship between duration of pain and secondary signs ofobstruction of the urinary tract on unenhanced helical CT AJRAm J Roentgenol 2001 177 325ndash30
30 Haddad MC Sharif HS Abomelha MS Colour Doppler sonogra-phy and plain abdominal radiography in the management ofpatients with renal colic Eur Radiol 1994 4 529ndash32
31 Catalano O Nunziata A Altei F Siani A Suspected ureteralcolic primary helical CT versus selective helical CT after unen-hanced radiography and sonography AJR Am J Roentgenol2002 178 379ndash87
(g s
(g s
S Watkins et al
194 copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
Appendix I Validation of ED physician US diagnosing hydronephrosis in ureteric colicCompleted forms to be placed in the marked box in the Resuscitation Room Date
Time
ED Physician name
Right Place patient details sticker here Location of
symptoms (Please circle or
comment)
Left
ULTRASOUND FINDINGS
Hydronephrosis Present
Estimate severity
NO MILD MODERATE SEVERE UNSURE
RIGHT
LEFT
Additional comments
Empty Half Full
Bladder size
Incidental Findings (eg Free fluidascites AAA effusion etc)
Study Group Use only CT scan
Performed Liverpool Hospital South West
RadiologyElsewhere
Date amp Time of scan
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
195
Appendix II Validation of ED physician US diagnosing hydronephrosis in ureteric colicReporting sheet for dr praneal sharma radiologist
Date Time
Place patient details sticker here
Diagnosis of Renal ureteric colic correct YES NO
Calculus Seen YES NO
Left Right
Position of Calculus
Size of Calculus
CT KUB FINDINGS
Hydronephrosis Present
Estimate severity
NO MILD MODERATE SEVERE UNSURE
RIGHT
LEFT
Additional Findings
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
189
Introduction
Patients presenting to the ED with an obstructive neph-ropathy benefit from early detection and treatment par-ticularly in the context of associated renal tract infectionor renal failure Complete ureteric obstruction mightlead to loss of renal function with an increased occur-rence of irreversible damage after 1ndash2 weeks1 Conven-tional modalities available for detecting renal tractobstruction include formal ultrasound (US) and comput-erized tomography (CT) However both CT and formalUS require patient transfer and monitoring in areasremote to the ED and generally have limited availabil-ity out of hours
The detection of hydronephrosis by bedside US in theED could allow a more rapid diagnosis and dispositionof patients and be used in situations where iodizingradiation or intravenous contrast material are contrain-dicated andor formal US is not available Rapid EDscreening for hydronephrosis might exclude obstruc-tion and focus assessment on other potential diagnosesand might allow the selection of patients for furtherstudies to be refined
An estimated 2ndash5 of the population will form asymptomatic renal calculus at some point in their lives1
Clinical history and microscopic haematuria suggestthe diagnosis with a sensitivity of 69ndash892ndash4 CT hasproved a more accurate test than US with greater sen-sitivity for hydronephrosis and calculus detection5ndash10
With intravenous contrast it can give the same infor-mation on renal function as IVU12568911 but also carriesthe same risks of radiation exposure allergy and neph-rotoxicity7 The advantages of CT have made US asecond choice investigation because of its relatively lowsensitivity of 19 (specificity of 97) for detectingcalculi compared with sensitivity of 94ndash97 (specificityof 96ndash97) for CT68 However patients with uretericstones have a relatively high prevalence of partial andor temporary renal tract obstruction but seldom causecomplete obstruction1 Because of this high prevalencethis population provides a convenient group in whomto explore options for diagnosing obstruction and thesepatients will require definitive renal tract imaging aspart of their usual management
The use of emergency physician US (EPU) for trau-matic intraperitoneal and pericardial fluid and abdom-inal aortic aneurysm detection has been increasing inAustralasian ED with accreditation processes thatfollow the Australasian College of Emergency Medi-cine (ACEM) guidelines12ndash14 Internationally EPU hasexpanded into areas of ED practice such as intravenous
access lower limb deep venous thrombosis and fluidlocalization for either diagnosis or drainage12ndash14 Therole of EPU in assessing renal tract obstruction remainssomewhat controversial with various studies showinga wide range of accuracy215ndash17
We postulated that EPU might be safe and accurateand could be used to screen patients for hydronephrosisdue to ureteric calculi
Aim
The aim of the present study was to determine theaccuracy of EPU in detection of hydronephrosis com-pared with radiologist-reported non-contrast CT
Methods
Study design and setting
This was a prospective study of a convenience sampleof patients with presumed ureteric colic undertaken inthe ED of a tertiary teaching hospital with an annualcensus of 46 000 The study had Area Ethics committeeapproval
Study population and protocol
All non-pregnant patients over 18 years old whoattended the ED with a clinical diagnosis of uretericcolic as determined by the treating ED medical officerwere eligible for enrolment Written informed consentwas obtained from all patients Foreign language inter-preters were utilized when required
Emergency registrars and emergency physicians (EP)who had attended an ACEM-accredited US workshopand undertaken a further 1 h of training in basic renalUS by a senior radiologist were able to enrol patientsin the study Recruited patients underwent EPU by theenrolling doctor during their ED stay The doctor per-forming the EPU was not involved in the treatment ofthe patient and was blinded to the CT scan result Anon-contrast CT scan of the renal tract also was per-formed on each patient either while they were in theED or organized externally at a single designated privateradiology suite within 24 h of EPU One senior radiolo-gist blinded to the EPU result reported the CT scans
Measurements
Using a Toshiba US machine Model SSA-550A(Tochigi-Ken Japan) and a 35ndash5 MHz curved array
S Watkins et al
190 copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
probe on abdominal preset investigators obtained andrecorded images and measurements of both kidneys(diameters of each renal pelvis and of each kidneyrsquoslongitudinal and transverse sections) Images weresaved digitally and printed Investigators completed areporting form that included demographic data thelocation of symptoms the presence or absence ofhydronephrosis for each kidney and the severity ofhydronephrosis (mild moderate or severe) if presentAn US diagnosis of hydronephrosis was made based onthe features listed in Table 1 Bladder size was esti-mated and documented (as empty half or full)
CT images were performed on a four-slice ToshibaAquilion (TSX-101A) scanner or a 16-slice ToshibaAquilion scanner Information recorded by the radiolo-gist included the presence or absence of hydronephrosisand its severity the presence of calculus and incidentalfindings
Data analysis
The accuracy of the EPU was determined by calculatingthe sensitivity specificity positive (PPV) and negativepredictive values (NPV) with 95 confidence intervals(CI) using the CT scan report as the reference standardAssuming a rate of hydronephrosis of 75 in patientswith renal colic19 we estimated approximately 50patients would be required to detect a sensitivity of90 with 95 CI of plusmn10
Microsoft Excel 2004 for Mac version 112 andVassarstats Statistical Computation website (httpfacultyvassaredulowryVassarStatshtml) were usedfor analysis
Results
Sixty-three patients were enrolled in the study Sixpatients did not complete the study because CT scanswere not performed or were not available for analysisOf the remaining 57 there were 48 men (84) and 9women (16) In the present study 34 (54) patientspresented with left-sided symptoms and 29 (46)
presented with right-sided symptoms The mean agewas 437 years (range 18ndash67) Two patients presentedmore than once during the study period and each pre-sentation was recorded separately
Of the 57 patients 48 had CT-confirmed diagnosis ofnephrolithiasis and 39 had CT-confirmed hydro-nephrosis Nine patients had negative CT scans fornephrolithiasis and hydronephrosis The prevalence ofhydronephrosis in the study population was 3957 or68 (CI 56ndash79)
Thirty-one of 39 patients with CT-proved hydroneph-rosis had positive EPU scans The main results aresummarized in Table 2 EPU demonstrated sensitivityof 80 (CI 65ndash89) and specificity of 83 (CI 61ndash94) a PPV of 91 (CI 75ndash98) and an NPV of 65(CI 43ndash83) Overall accuracy of EPU for the detectionof hydronephrosis was 81 (CI 69ndash89)
The study authors performed 60 of the scans12 other EP or trainees (registrars) performed theremainder
Discussion
The present study found that with minimal training EPand trainees were able to achieve a sensitivity of 80and a specificity of 83 for the diagnosis of hydroneph-rosis using bedside US in the setting of suspected acuterenal colic
Emergency physician US has potential advantages inthe diagnosis of hydronephrosis It can be performed atthe bedside using a portable machine is immediatelyavailable and repeatable 24 h a day 7 days a weekPatients do not leave the department to go to potentiallyless monitored areas which obviates the need for por-table monitoring andor nurse escort EP have been
Table 1 Grades of hydronephrosis
Grade I Grade II ndash mild Grade III ndash moderate Grade IV ndash severe
Slight blunting of calyceal fornices
Obvious blunting of calyceal fornices andenlargement of calices but intruding shadows of papillae are easily seen
Rounding of calices withobliteration of papillae
Extreme calyceal ballooning
Adapted from the study by Grainger and Allison18
Table 2 Comparison of emergency physician US (EPU) andCT in detection of hydronephrosis
CT ndash positive CT ndash negative
EPU ndash positive 31 3EPU ndash negative 8 15
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
191
using bedside US in the ED over recent years and inAustralia have training and accreditation proceduresfor its use in other conditions12ndash14
In the present study we investigated the ability of EPwith 1 h of focused training in renal US in addition toan ACEM-accredited US workshop to detect the pres-ence of hydronephrosis Previous studies have com-pared EPU with IVP215ndash17 or compared radiologist-performed US with CT59 This is the first study thatdirectly compares EPU detection of hydronephrosiswith the current lsquogold-standardrsquo of radiologist-reportedCT scan
Previous studies have demonstrated US sensitivity of85ndash94 and specificity of 100 in detection of hydro-nephrosis when performed by radiologists or sonogra-phers19102021 CT has proved a more accurate test withgreater sensitivity for hydronephrosis and calculusdetection5ndash10
The accuracy reported in our study is comparable tothat previously reported in most other EPU trials Rosenet al with 5 h training compared EPU diagnosis ofhydronephrosis with IVP and CT in 126 patients find-ing a sensitivity of 72 specificity of 73 PPV of85 NPV of 54 and accuracy of 7216 By contrastHenderson et al found a sensitivity of 97 for lsquopathol-ogy consistent with nephro-ureterolithiasis when com-pared to IVPrsquo in 108 patients but did not specificallyreport the detection of hydronephrosis as an outcome2
Lanoix et al reported an accuracy of 94 and sensitiv-ity of 96 after 4 h tuition based on 45 subjects and 39EPtrainees15 However the reference standard used inthat study is unclear
In an Australian study with 3 days of US trainingRowland et al reported 68 accuracy for EPU usingthree grades of hydronephrosis nil subtle or obvious17
They reported a sensitivity of 93 but only a specific-ity of 47 (PPV 59 NPV 89) and used IVP formalUS within 24 h or radiologist review of the EPU as theirlsquogold-standardrsquo Four investigators obtained images
from 31 subjects They reported more false-positivesthan false-negatives whereas our study reports theopposite In the above studies the difference betweenthe US and CT grading were in subjects with low-gradehydronephrosis A comparison of these studies isshown in Table 3
It is worth noting that the sonographic grading ofhydronephrosis into mild moderate or severe correlatespoorly with the clinical severity of disease2223 Hencefor the purposes of data analysis in our study hydro-nephrosis was reported simply as either present orabsent
To be effective as a screening test EPU wouldrequire a high sensitivity (ie few false-negatives) Ourfinding of a sensitivity of 79 and NPV of 65 sug-gests that EPU is currently not an acceptable screeningtest to rule out hydronephrosis However althougheight cases of CT-confirmed hydronephrosis were notdetected by EPU seven of these false-negative scanswere reported as mild hydronephrosis on CT Theeighth patient with false-negative EPU had moderatehydronephrosis demonstrated on CT but this scan wasperformed more than 24 h after the EPU hence it isuncertain whether this truly reflects the presence ofhydronephrosis at the time of the EPU but has beenincorporated into our results for completeness
Radiological diagnosis of hydronephrosis on CT issubjective with several studies reporting inter-observervariability between radiologists and between radiolo-gists trainees and urologists23ndash27 The amount of hydro-nephrosis shown by US varies dynamically with partialobstruction and with hydration status of the patient1125
as hydronephrosis can be induced in healthy volunteerswith forced fluid intake In serial US following hydra-tion mildndashmoderate hydronephrosis was induced in80 of subjects28 Repeating the US in dehydratedpatients following hydration might alter previouslyfalse-negative results11 Studies have demonstrated sim-ilar dynamic changes with CT Perinephric stranding
Table 3 Comparison of the previous studies of emergency physician US and detection of hydronephrosis
Sensitivity()
NPV () Comment
The present study (2005) (n = 57) 80 65 CT only 1 h + courseRosen et al16 (1998) (n = 126) 72 54 IVU + CT 5 h trainingHenderson et al2 (1998) (n = 108) 97 92 IVU unclear diagnostic criteriaLanoix et al15 (2000) (n = 45) 94 94 Multiple reference standards 4 h training 39 investigatorsRowland et al17 (2001) (n = 31) 93 89 Used IVUUSradiologist 3 days training 68 accuracy
NPV negative predictive value
S Watkins et al
192 copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
ureteral dilatation perinephric fluid and collecting sys-tem dilatation showed statistically significant changeover 8 h of study29 Therefore correlation discrepancy inour study might be explained by any of these factorsIt has also been reported that false-negatives on US areusually followed by uncomplicated spontaneous stoneemission3031
What level of minimum training is required to makeEPU an effective screening tool for hydronephrosisLanoix et al15 and Rosen et al16 trained EP for 4 h and5 h respectively with markedly different results asnoted earlier perhaps because of the very different ref-erence standards used in their studies Rowland et aldemonstrated an overall accuracy of 68 for EPU after3 days training whereas our study demonstrates anaccuracy of 81 after 1 h of focused training in renalscans following completion of an ACEM-accreditedworkshop17
From our results it would seem prudent to state thatalthough additional training and experience mightimprove the accuracy of EPU it will not supplant theuse of CT in the foreseeable future However despitesimilar results to ours for EPU accuracy previousauthors have suggested a place for EPU in the detectionof hydronephrosis21516
Australasian and American Colleges for EmergencyMedicine have published policies on the training andaccreditation of EP in focused assessment with sonog-raphy for trauma and abdominal aortic aneurysm12ndash14
However currently there are no guidelines for minimumtraining and accreditation in renal sonography for EP
Limitations
As patients were enrolled on a convenience basis dueto the presence or absence of an investigator to per-form the scans this might have introduced one ormore unknown biases A trend towards improvedinvestigator performance was noted as scan qualityimproved with experience this might have affectedthe detection rate in the earlier stages of the presentstudy Some EP performed less than three studiesothers more than 10 however the sample size was toosmall to afford meaningful subgroup analysis for indi-vidual EP Some of these limitations would be over-come by larger studies
It is worth noting that several of the EPU sono-graphers were relatively inexperienced in the use ofbedside US and were not yet accredited in otherEPU applications such as focused assessment by
sonography in trauma (FAST)abdominal aortic aneu-rysm (AAA) It is possible that accredited EPU sonog-raphers would be more accurate
Because of the dynamic nature of renal colic andurinary obstruction ideally all participants would havehad their EPU and CT scan performed within minimaltime delay to ensure an accurate assessment of EPU inone patient CT scanning was performed more than 24 hlater Finally our study did not include routine evalua-tion of renal resistive indexes that might improve detec-tion of early obstruction28
Conclusion
Using non-contrast CT as the gold standard we havefound EPU detection of hydronephrosis to have anaccuracy of 81 which is comparable to previousstudies However on the basis of the present studyEPU is probably not accurate enough to rule out hydro-nephrosis Further experience and training mightimprove the accuracy of EPU and allow its use as ascreening tool
Acknowledgements
The authors express their thanks to the staff specialistsand registrars from the Departments of EmergencyMedicine and Radiology for their assistance
Author contributions
SW contributed to study design ethics submission con-sent and patient information sheets reporting sheetsdata collation and analysis literature research andmanuscript preparation (85) JB contributed to origi-nal concept study design ethics submission investiga-tion manuscript preparation and supervision (40) PScontributed to investigator training study design andCT reporting (25) AH contributed to research meth-ods data analysis and manuscript preparation (20)AG contributed to study design investigation andmanuscript preparation (10) LC contributed to studydesign investigation and data collation (10)
Competing interests
Justin Bowra is a member of the Ultrasound Committeeof the Australasian College for Emergency MedicineAnna Holdgate holds the position of Section Editor of
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
193
Original Research for Emergency Medicine AustralasiaAll other authors declare no competing interests
Accepted 11 October 2006
References
1 Rosen P Barkin R Ling LJ Emergency Medicine Concepts andClinical Practice 4th edn St Louis Mosby 1998
2 Henderson SO Hoffner RJ Aragona JL Groth DE Esekogwu VIChan D Bedside emergency department ultrasonography plusradiography of the kidneys ureters and bladder versus intrave-nous pyelography in the evaluation of suspected ureteral colicAcad Emerg Med 1998 5 666ndash71
3 Eray O Cubuk MS Oktay CEM Yilmaz S Cete Y Ersoy FFThe efficacy of urinalysis plain films and spiral CT in EDpatients with suspected renal colic Am J Emerg Med 2003 21152ndash4
4 Brown DFM Rosen CL Sagarin M McCabe C Wolfe RE Impactof bedside ultrasonography by emergency physicians on theclinical likelihood of nephrolithiasis Ann Emerg Med 1996 27818
5 Fowler KAB Locken JA Duchesne JH Williamson MR Ultra-sound for detecting renal calculi with non-enhanced CT as areference standard Radiology 2002 222 109ndash13
6 Yilmaz S Sindel T Arslau G et al Renal colic comparison ofspiral CT US and IVU in the detection of ureteral calculi EurRadiol 1998 8 212ndash17
7 Spencer BA Wood BJ Dretler SP Helical CT and ureteric colicUrol Clin North Am 2000 27 231ndash41
8 Smith RC Verga M McCarthy S Rosenfield AT Diagnosis ofacute flank pain value of un-enhanced helical CT AJR Am JRoentgenol 1996 166 97ndash101
9 Sheafor DH Hertzberg BS Freed KS et al Non-enhanced helicalCT and US in the emergency evaluation of patients with renalcolic prospective comparison Radiology 2000 217 792ndash7
10 Patlas M Farkas A Fisher D Zaghal I Hadas-Halpern I Ultra-sound vs CT for the detection of ureteric stones in patients withrenal colic BJR 2001 74 901ndash4
11 Noble VE Brown DFM Renal ultrasound Emerg Med ClinNorth Am 2004 22 641ndash59
12 American College of Emergency Physicians (ACEP) Board ofDirectors Use of Ultrasound Imaging by Emergency PhysiciansPolicy Statement June 2001 Available from URL httpwwwaceporg16840html [Accessed August 2006]
13 Australasian College of Emergency Medicine (ACEM) CouncilCredentialling for ED Ultrasonography Policy Document P22July 2000 Available from URL httpwwwacemorgauinfo-centreaspxdocId=59 [Accessed August 2006]
14 Australasian College of Emergency Medicine (ACEM) CouncilUse of Bedside Ultrasound by Emergency Physicians Policy Doc-ument P21 July 1999 Available from URL httpwwwacemorgauinfocentreaspxdocId=59 [Accessed August 2006]
15 Lanoix R Leak LV Gaeta T Gernsheimer JR A preliminaryevaluation of emergency ultrasound in the setting of an emer-gency medicine training program Am J Emerg Med 2000 1841ndash5
16 Rosen CL Brown DFM Sagarin MJ Chang Y McCabe CJ WolfeRE Ultrasonography by emergency physicians in patients withsuspected renal colic J Emerg Med 1998 16 865ndash70
17 Rowland JL Kuhn M Bonnin RLL Davey MJ Langlois SLAccuracy of emergency department bedside ultrasonographyEmerg Med 2001 13 305ndash13
18 Grainger RG Allison DJ (eds) Diagnostic Radiology A Textbookof Medical Imaging 4th edn London Churchill Livingstone2001 p 1594
19 Kiely EA Hartnell GG Gibson RN Measurement of bladderVolume by real-time ultrasound Br J Urol 1987 60 33ndash5
20 Sinclair D Wilson S Toi A Greenspan L The evaluation ofsuspected renal colic ultrasound scan vs excretory urographyAnn Emerg Med 1989 18 556ndash9
21 Dalla Palma L Stacul F Bazzocchi M et al Ultrasonography andplain film versus intravenous urography in ureteric colic ClinRadiol 1993 47 333ndash6
22 Oumlzden E Karamuumlrsel T Gouml uuml Ccedil Yaman Ouml Inal T Gouml uuml ODetection rate of ureter stones with US relationship with gradeof hydronephrosis J Ankara Med Sch 2002 24 183ndash6
23 King L Hydronephrosis when is obstruction not obstructionCommon problems in paediatric urology Urol Clin North Am1995 22 31ndash42
24 Holdgate A Chan T How accurate are emergency clinicians atinterpreting non-contrast CT for suspected renal colic AcadEmerg Med 2003 10 315ndash19
25 Jeffrey RB Federle MP CT and ultrasonography of acute renalabnormalities Radiol Clin North Am 1983 21 515ndash25
26 Morse JW Saracino BS Melanson SW Arcona S Heller MBUltrasound interpretation of hydronephrosis is improved by abrief educational intervention Ann Emerg Med 1998 32(Suppl Pt 2) S27
27 Freed KS Paulson EK Frederick MG et al Interobserver vari-ability in the interpretation of unenhanced helical CT for thediagnosis of ureteral stones J Comput Assist Tomogr 1998 22732ndash7
28 Morse JW Hill R Greissinger WP Patterson JW Melanson SWHeller MB Rapid oral hydration results in hydronephrosis asdetermined by bedside ultrasound Ann Emerg Med 1999 34134ndash40
29 Varanelli MJ Coll DM Levine JA Rosenfield AT Smith RCRelationship between duration of pain and secondary signs ofobstruction of the urinary tract on unenhanced helical CT AJRAm J Roentgenol 2001 177 325ndash30
30 Haddad MC Sharif HS Abomelha MS Colour Doppler sonogra-phy and plain abdominal radiography in the management ofpatients with renal colic Eur Radiol 1994 4 529ndash32
31 Catalano O Nunziata A Altei F Siani A Suspected ureteralcolic primary helical CT versus selective helical CT after unen-hanced radiography and sonography AJR Am J Roentgenol2002 178 379ndash87
(g s
(g s
S Watkins et al
194 copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
Appendix I Validation of ED physician US diagnosing hydronephrosis in ureteric colicCompleted forms to be placed in the marked box in the Resuscitation Room Date
Time
ED Physician name
Right Place patient details sticker here Location of
symptoms (Please circle or
comment)
Left
ULTRASOUND FINDINGS
Hydronephrosis Present
Estimate severity
NO MILD MODERATE SEVERE UNSURE
RIGHT
LEFT
Additional comments
Empty Half Full
Bladder size
Incidental Findings (eg Free fluidascites AAA effusion etc)
Study Group Use only CT scan
Performed Liverpool Hospital South West
RadiologyElsewhere
Date amp Time of scan
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
195
Appendix II Validation of ED physician US diagnosing hydronephrosis in ureteric colicReporting sheet for dr praneal sharma radiologist
Date Time
Place patient details sticker here
Diagnosis of Renal ureteric colic correct YES NO
Calculus Seen YES NO
Left Right
Position of Calculus
Size of Calculus
CT KUB FINDINGS
Hydronephrosis Present
Estimate severity
NO MILD MODERATE SEVERE UNSURE
RIGHT
LEFT
Additional Findings
S Watkins et al
190 copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
probe on abdominal preset investigators obtained andrecorded images and measurements of both kidneys(diameters of each renal pelvis and of each kidneyrsquoslongitudinal and transverse sections) Images weresaved digitally and printed Investigators completed areporting form that included demographic data thelocation of symptoms the presence or absence ofhydronephrosis for each kidney and the severity ofhydronephrosis (mild moderate or severe) if presentAn US diagnosis of hydronephrosis was made based onthe features listed in Table 1 Bladder size was esti-mated and documented (as empty half or full)
CT images were performed on a four-slice ToshibaAquilion (TSX-101A) scanner or a 16-slice ToshibaAquilion scanner Information recorded by the radiolo-gist included the presence or absence of hydronephrosisand its severity the presence of calculus and incidentalfindings
Data analysis
The accuracy of the EPU was determined by calculatingthe sensitivity specificity positive (PPV) and negativepredictive values (NPV) with 95 confidence intervals(CI) using the CT scan report as the reference standardAssuming a rate of hydronephrosis of 75 in patientswith renal colic19 we estimated approximately 50patients would be required to detect a sensitivity of90 with 95 CI of plusmn10
Microsoft Excel 2004 for Mac version 112 andVassarstats Statistical Computation website (httpfacultyvassaredulowryVassarStatshtml) were usedfor analysis
Results
Sixty-three patients were enrolled in the study Sixpatients did not complete the study because CT scanswere not performed or were not available for analysisOf the remaining 57 there were 48 men (84) and 9women (16) In the present study 34 (54) patientspresented with left-sided symptoms and 29 (46)
presented with right-sided symptoms The mean agewas 437 years (range 18ndash67) Two patients presentedmore than once during the study period and each pre-sentation was recorded separately
Of the 57 patients 48 had CT-confirmed diagnosis ofnephrolithiasis and 39 had CT-confirmed hydro-nephrosis Nine patients had negative CT scans fornephrolithiasis and hydronephrosis The prevalence ofhydronephrosis in the study population was 3957 or68 (CI 56ndash79)
Thirty-one of 39 patients with CT-proved hydroneph-rosis had positive EPU scans The main results aresummarized in Table 2 EPU demonstrated sensitivityof 80 (CI 65ndash89) and specificity of 83 (CI 61ndash94) a PPV of 91 (CI 75ndash98) and an NPV of 65(CI 43ndash83) Overall accuracy of EPU for the detectionof hydronephrosis was 81 (CI 69ndash89)
The study authors performed 60 of the scans12 other EP or trainees (registrars) performed theremainder
Discussion
The present study found that with minimal training EPand trainees were able to achieve a sensitivity of 80and a specificity of 83 for the diagnosis of hydroneph-rosis using bedside US in the setting of suspected acuterenal colic
Emergency physician US has potential advantages inthe diagnosis of hydronephrosis It can be performed atthe bedside using a portable machine is immediatelyavailable and repeatable 24 h a day 7 days a weekPatients do not leave the department to go to potentiallyless monitored areas which obviates the need for por-table monitoring andor nurse escort EP have been
Table 1 Grades of hydronephrosis
Grade I Grade II ndash mild Grade III ndash moderate Grade IV ndash severe
Slight blunting of calyceal fornices
Obvious blunting of calyceal fornices andenlargement of calices but intruding shadows of papillae are easily seen
Rounding of calices withobliteration of papillae
Extreme calyceal ballooning
Adapted from the study by Grainger and Allison18
Table 2 Comparison of emergency physician US (EPU) andCT in detection of hydronephrosis
CT ndash positive CT ndash negative
EPU ndash positive 31 3EPU ndash negative 8 15
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
191
using bedside US in the ED over recent years and inAustralia have training and accreditation proceduresfor its use in other conditions12ndash14
In the present study we investigated the ability of EPwith 1 h of focused training in renal US in addition toan ACEM-accredited US workshop to detect the pres-ence of hydronephrosis Previous studies have com-pared EPU with IVP215ndash17 or compared radiologist-performed US with CT59 This is the first study thatdirectly compares EPU detection of hydronephrosiswith the current lsquogold-standardrsquo of radiologist-reportedCT scan
Previous studies have demonstrated US sensitivity of85ndash94 and specificity of 100 in detection of hydro-nephrosis when performed by radiologists or sonogra-phers19102021 CT has proved a more accurate test withgreater sensitivity for hydronephrosis and calculusdetection5ndash10
The accuracy reported in our study is comparable tothat previously reported in most other EPU trials Rosenet al with 5 h training compared EPU diagnosis ofhydronephrosis with IVP and CT in 126 patients find-ing a sensitivity of 72 specificity of 73 PPV of85 NPV of 54 and accuracy of 7216 By contrastHenderson et al found a sensitivity of 97 for lsquopathol-ogy consistent with nephro-ureterolithiasis when com-pared to IVPrsquo in 108 patients but did not specificallyreport the detection of hydronephrosis as an outcome2
Lanoix et al reported an accuracy of 94 and sensitiv-ity of 96 after 4 h tuition based on 45 subjects and 39EPtrainees15 However the reference standard used inthat study is unclear
In an Australian study with 3 days of US trainingRowland et al reported 68 accuracy for EPU usingthree grades of hydronephrosis nil subtle or obvious17
They reported a sensitivity of 93 but only a specific-ity of 47 (PPV 59 NPV 89) and used IVP formalUS within 24 h or radiologist review of the EPU as theirlsquogold-standardrsquo Four investigators obtained images
from 31 subjects They reported more false-positivesthan false-negatives whereas our study reports theopposite In the above studies the difference betweenthe US and CT grading were in subjects with low-gradehydronephrosis A comparison of these studies isshown in Table 3
It is worth noting that the sonographic grading ofhydronephrosis into mild moderate or severe correlatespoorly with the clinical severity of disease2223 Hencefor the purposes of data analysis in our study hydro-nephrosis was reported simply as either present orabsent
To be effective as a screening test EPU wouldrequire a high sensitivity (ie few false-negatives) Ourfinding of a sensitivity of 79 and NPV of 65 sug-gests that EPU is currently not an acceptable screeningtest to rule out hydronephrosis However althougheight cases of CT-confirmed hydronephrosis were notdetected by EPU seven of these false-negative scanswere reported as mild hydronephrosis on CT Theeighth patient with false-negative EPU had moderatehydronephrosis demonstrated on CT but this scan wasperformed more than 24 h after the EPU hence it isuncertain whether this truly reflects the presence ofhydronephrosis at the time of the EPU but has beenincorporated into our results for completeness
Radiological diagnosis of hydronephrosis on CT issubjective with several studies reporting inter-observervariability between radiologists and between radiolo-gists trainees and urologists23ndash27 The amount of hydro-nephrosis shown by US varies dynamically with partialobstruction and with hydration status of the patient1125
as hydronephrosis can be induced in healthy volunteerswith forced fluid intake In serial US following hydra-tion mildndashmoderate hydronephrosis was induced in80 of subjects28 Repeating the US in dehydratedpatients following hydration might alter previouslyfalse-negative results11 Studies have demonstrated sim-ilar dynamic changes with CT Perinephric stranding
Table 3 Comparison of the previous studies of emergency physician US and detection of hydronephrosis
Sensitivity()
NPV () Comment
The present study (2005) (n = 57) 80 65 CT only 1 h + courseRosen et al16 (1998) (n = 126) 72 54 IVU + CT 5 h trainingHenderson et al2 (1998) (n = 108) 97 92 IVU unclear diagnostic criteriaLanoix et al15 (2000) (n = 45) 94 94 Multiple reference standards 4 h training 39 investigatorsRowland et al17 (2001) (n = 31) 93 89 Used IVUUSradiologist 3 days training 68 accuracy
NPV negative predictive value
S Watkins et al
192 copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
ureteral dilatation perinephric fluid and collecting sys-tem dilatation showed statistically significant changeover 8 h of study29 Therefore correlation discrepancy inour study might be explained by any of these factorsIt has also been reported that false-negatives on US areusually followed by uncomplicated spontaneous stoneemission3031
What level of minimum training is required to makeEPU an effective screening tool for hydronephrosisLanoix et al15 and Rosen et al16 trained EP for 4 h and5 h respectively with markedly different results asnoted earlier perhaps because of the very different ref-erence standards used in their studies Rowland et aldemonstrated an overall accuracy of 68 for EPU after3 days training whereas our study demonstrates anaccuracy of 81 after 1 h of focused training in renalscans following completion of an ACEM-accreditedworkshop17
From our results it would seem prudent to state thatalthough additional training and experience mightimprove the accuracy of EPU it will not supplant theuse of CT in the foreseeable future However despitesimilar results to ours for EPU accuracy previousauthors have suggested a place for EPU in the detectionof hydronephrosis21516
Australasian and American Colleges for EmergencyMedicine have published policies on the training andaccreditation of EP in focused assessment with sonog-raphy for trauma and abdominal aortic aneurysm12ndash14
However currently there are no guidelines for minimumtraining and accreditation in renal sonography for EP
Limitations
As patients were enrolled on a convenience basis dueto the presence or absence of an investigator to per-form the scans this might have introduced one ormore unknown biases A trend towards improvedinvestigator performance was noted as scan qualityimproved with experience this might have affectedthe detection rate in the earlier stages of the presentstudy Some EP performed less than three studiesothers more than 10 however the sample size was toosmall to afford meaningful subgroup analysis for indi-vidual EP Some of these limitations would be over-come by larger studies
It is worth noting that several of the EPU sono-graphers were relatively inexperienced in the use ofbedside US and were not yet accredited in otherEPU applications such as focused assessment by
sonography in trauma (FAST)abdominal aortic aneu-rysm (AAA) It is possible that accredited EPU sonog-raphers would be more accurate
Because of the dynamic nature of renal colic andurinary obstruction ideally all participants would havehad their EPU and CT scan performed within minimaltime delay to ensure an accurate assessment of EPU inone patient CT scanning was performed more than 24 hlater Finally our study did not include routine evalua-tion of renal resistive indexes that might improve detec-tion of early obstruction28
Conclusion
Using non-contrast CT as the gold standard we havefound EPU detection of hydronephrosis to have anaccuracy of 81 which is comparable to previousstudies However on the basis of the present studyEPU is probably not accurate enough to rule out hydro-nephrosis Further experience and training mightimprove the accuracy of EPU and allow its use as ascreening tool
Acknowledgements
The authors express their thanks to the staff specialistsand registrars from the Departments of EmergencyMedicine and Radiology for their assistance
Author contributions
SW contributed to study design ethics submission con-sent and patient information sheets reporting sheetsdata collation and analysis literature research andmanuscript preparation (85) JB contributed to origi-nal concept study design ethics submission investiga-tion manuscript preparation and supervision (40) PScontributed to investigator training study design andCT reporting (25) AH contributed to research meth-ods data analysis and manuscript preparation (20)AG contributed to study design investigation andmanuscript preparation (10) LC contributed to studydesign investigation and data collation (10)
Competing interests
Justin Bowra is a member of the Ultrasound Committeeof the Australasian College for Emergency MedicineAnna Holdgate holds the position of Section Editor of
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
193
Original Research for Emergency Medicine AustralasiaAll other authors declare no competing interests
Accepted 11 October 2006
References
1 Rosen P Barkin R Ling LJ Emergency Medicine Concepts andClinical Practice 4th edn St Louis Mosby 1998
2 Henderson SO Hoffner RJ Aragona JL Groth DE Esekogwu VIChan D Bedside emergency department ultrasonography plusradiography of the kidneys ureters and bladder versus intrave-nous pyelography in the evaluation of suspected ureteral colicAcad Emerg Med 1998 5 666ndash71
3 Eray O Cubuk MS Oktay CEM Yilmaz S Cete Y Ersoy FFThe efficacy of urinalysis plain films and spiral CT in EDpatients with suspected renal colic Am J Emerg Med 2003 21152ndash4
4 Brown DFM Rosen CL Sagarin M McCabe C Wolfe RE Impactof bedside ultrasonography by emergency physicians on theclinical likelihood of nephrolithiasis Ann Emerg Med 1996 27818
5 Fowler KAB Locken JA Duchesne JH Williamson MR Ultra-sound for detecting renal calculi with non-enhanced CT as areference standard Radiology 2002 222 109ndash13
6 Yilmaz S Sindel T Arslau G et al Renal colic comparison ofspiral CT US and IVU in the detection of ureteral calculi EurRadiol 1998 8 212ndash17
7 Spencer BA Wood BJ Dretler SP Helical CT and ureteric colicUrol Clin North Am 2000 27 231ndash41
8 Smith RC Verga M McCarthy S Rosenfield AT Diagnosis ofacute flank pain value of un-enhanced helical CT AJR Am JRoentgenol 1996 166 97ndash101
9 Sheafor DH Hertzberg BS Freed KS et al Non-enhanced helicalCT and US in the emergency evaluation of patients with renalcolic prospective comparison Radiology 2000 217 792ndash7
10 Patlas M Farkas A Fisher D Zaghal I Hadas-Halpern I Ultra-sound vs CT for the detection of ureteric stones in patients withrenal colic BJR 2001 74 901ndash4
11 Noble VE Brown DFM Renal ultrasound Emerg Med ClinNorth Am 2004 22 641ndash59
12 American College of Emergency Physicians (ACEP) Board ofDirectors Use of Ultrasound Imaging by Emergency PhysiciansPolicy Statement June 2001 Available from URL httpwwwaceporg16840html [Accessed August 2006]
13 Australasian College of Emergency Medicine (ACEM) CouncilCredentialling for ED Ultrasonography Policy Document P22July 2000 Available from URL httpwwwacemorgauinfo-centreaspxdocId=59 [Accessed August 2006]
14 Australasian College of Emergency Medicine (ACEM) CouncilUse of Bedside Ultrasound by Emergency Physicians Policy Doc-ument P21 July 1999 Available from URL httpwwwacemorgauinfocentreaspxdocId=59 [Accessed August 2006]
15 Lanoix R Leak LV Gaeta T Gernsheimer JR A preliminaryevaluation of emergency ultrasound in the setting of an emer-gency medicine training program Am J Emerg Med 2000 1841ndash5
16 Rosen CL Brown DFM Sagarin MJ Chang Y McCabe CJ WolfeRE Ultrasonography by emergency physicians in patients withsuspected renal colic J Emerg Med 1998 16 865ndash70
17 Rowland JL Kuhn M Bonnin RLL Davey MJ Langlois SLAccuracy of emergency department bedside ultrasonographyEmerg Med 2001 13 305ndash13
18 Grainger RG Allison DJ (eds) Diagnostic Radiology A Textbookof Medical Imaging 4th edn London Churchill Livingstone2001 p 1594
19 Kiely EA Hartnell GG Gibson RN Measurement of bladderVolume by real-time ultrasound Br J Urol 1987 60 33ndash5
20 Sinclair D Wilson S Toi A Greenspan L The evaluation ofsuspected renal colic ultrasound scan vs excretory urographyAnn Emerg Med 1989 18 556ndash9
21 Dalla Palma L Stacul F Bazzocchi M et al Ultrasonography andplain film versus intravenous urography in ureteric colic ClinRadiol 1993 47 333ndash6
22 Oumlzden E Karamuumlrsel T Gouml uuml Ccedil Yaman Ouml Inal T Gouml uuml ODetection rate of ureter stones with US relationship with gradeof hydronephrosis J Ankara Med Sch 2002 24 183ndash6
23 King L Hydronephrosis when is obstruction not obstructionCommon problems in paediatric urology Urol Clin North Am1995 22 31ndash42
24 Holdgate A Chan T How accurate are emergency clinicians atinterpreting non-contrast CT for suspected renal colic AcadEmerg Med 2003 10 315ndash19
25 Jeffrey RB Federle MP CT and ultrasonography of acute renalabnormalities Radiol Clin North Am 1983 21 515ndash25
26 Morse JW Saracino BS Melanson SW Arcona S Heller MBUltrasound interpretation of hydronephrosis is improved by abrief educational intervention Ann Emerg Med 1998 32(Suppl Pt 2) S27
27 Freed KS Paulson EK Frederick MG et al Interobserver vari-ability in the interpretation of unenhanced helical CT for thediagnosis of ureteral stones J Comput Assist Tomogr 1998 22732ndash7
28 Morse JW Hill R Greissinger WP Patterson JW Melanson SWHeller MB Rapid oral hydration results in hydronephrosis asdetermined by bedside ultrasound Ann Emerg Med 1999 34134ndash40
29 Varanelli MJ Coll DM Levine JA Rosenfield AT Smith RCRelationship between duration of pain and secondary signs ofobstruction of the urinary tract on unenhanced helical CT AJRAm J Roentgenol 2001 177 325ndash30
30 Haddad MC Sharif HS Abomelha MS Colour Doppler sonogra-phy and plain abdominal radiography in the management ofpatients with renal colic Eur Radiol 1994 4 529ndash32
31 Catalano O Nunziata A Altei F Siani A Suspected ureteralcolic primary helical CT versus selective helical CT after unen-hanced radiography and sonography AJR Am J Roentgenol2002 178 379ndash87
(g s
(g s
S Watkins et al
194 copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
Appendix I Validation of ED physician US diagnosing hydronephrosis in ureteric colicCompleted forms to be placed in the marked box in the Resuscitation Room Date
Time
ED Physician name
Right Place patient details sticker here Location of
symptoms (Please circle or
comment)
Left
ULTRASOUND FINDINGS
Hydronephrosis Present
Estimate severity
NO MILD MODERATE SEVERE UNSURE
RIGHT
LEFT
Additional comments
Empty Half Full
Bladder size
Incidental Findings (eg Free fluidascites AAA effusion etc)
Study Group Use only CT scan
Performed Liverpool Hospital South West
RadiologyElsewhere
Date amp Time of scan
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
195
Appendix II Validation of ED physician US diagnosing hydronephrosis in ureteric colicReporting sheet for dr praneal sharma radiologist
Date Time
Place patient details sticker here
Diagnosis of Renal ureteric colic correct YES NO
Calculus Seen YES NO
Left Right
Position of Calculus
Size of Calculus
CT KUB FINDINGS
Hydronephrosis Present
Estimate severity
NO MILD MODERATE SEVERE UNSURE
RIGHT
LEFT
Additional Findings
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
191
using bedside US in the ED over recent years and inAustralia have training and accreditation proceduresfor its use in other conditions12ndash14
In the present study we investigated the ability of EPwith 1 h of focused training in renal US in addition toan ACEM-accredited US workshop to detect the pres-ence of hydronephrosis Previous studies have com-pared EPU with IVP215ndash17 or compared radiologist-performed US with CT59 This is the first study thatdirectly compares EPU detection of hydronephrosiswith the current lsquogold-standardrsquo of radiologist-reportedCT scan
Previous studies have demonstrated US sensitivity of85ndash94 and specificity of 100 in detection of hydro-nephrosis when performed by radiologists or sonogra-phers19102021 CT has proved a more accurate test withgreater sensitivity for hydronephrosis and calculusdetection5ndash10
The accuracy reported in our study is comparable tothat previously reported in most other EPU trials Rosenet al with 5 h training compared EPU diagnosis ofhydronephrosis with IVP and CT in 126 patients find-ing a sensitivity of 72 specificity of 73 PPV of85 NPV of 54 and accuracy of 7216 By contrastHenderson et al found a sensitivity of 97 for lsquopathol-ogy consistent with nephro-ureterolithiasis when com-pared to IVPrsquo in 108 patients but did not specificallyreport the detection of hydronephrosis as an outcome2
Lanoix et al reported an accuracy of 94 and sensitiv-ity of 96 after 4 h tuition based on 45 subjects and 39EPtrainees15 However the reference standard used inthat study is unclear
In an Australian study with 3 days of US trainingRowland et al reported 68 accuracy for EPU usingthree grades of hydronephrosis nil subtle or obvious17
They reported a sensitivity of 93 but only a specific-ity of 47 (PPV 59 NPV 89) and used IVP formalUS within 24 h or radiologist review of the EPU as theirlsquogold-standardrsquo Four investigators obtained images
from 31 subjects They reported more false-positivesthan false-negatives whereas our study reports theopposite In the above studies the difference betweenthe US and CT grading were in subjects with low-gradehydronephrosis A comparison of these studies isshown in Table 3
It is worth noting that the sonographic grading ofhydronephrosis into mild moderate or severe correlatespoorly with the clinical severity of disease2223 Hencefor the purposes of data analysis in our study hydro-nephrosis was reported simply as either present orabsent
To be effective as a screening test EPU wouldrequire a high sensitivity (ie few false-negatives) Ourfinding of a sensitivity of 79 and NPV of 65 sug-gests that EPU is currently not an acceptable screeningtest to rule out hydronephrosis However althougheight cases of CT-confirmed hydronephrosis were notdetected by EPU seven of these false-negative scanswere reported as mild hydronephrosis on CT Theeighth patient with false-negative EPU had moderatehydronephrosis demonstrated on CT but this scan wasperformed more than 24 h after the EPU hence it isuncertain whether this truly reflects the presence ofhydronephrosis at the time of the EPU but has beenincorporated into our results for completeness
Radiological diagnosis of hydronephrosis on CT issubjective with several studies reporting inter-observervariability between radiologists and between radiolo-gists trainees and urologists23ndash27 The amount of hydro-nephrosis shown by US varies dynamically with partialobstruction and with hydration status of the patient1125
as hydronephrosis can be induced in healthy volunteerswith forced fluid intake In serial US following hydra-tion mildndashmoderate hydronephrosis was induced in80 of subjects28 Repeating the US in dehydratedpatients following hydration might alter previouslyfalse-negative results11 Studies have demonstrated sim-ilar dynamic changes with CT Perinephric stranding
Table 3 Comparison of the previous studies of emergency physician US and detection of hydronephrosis
Sensitivity()
NPV () Comment
The present study (2005) (n = 57) 80 65 CT only 1 h + courseRosen et al16 (1998) (n = 126) 72 54 IVU + CT 5 h trainingHenderson et al2 (1998) (n = 108) 97 92 IVU unclear diagnostic criteriaLanoix et al15 (2000) (n = 45) 94 94 Multiple reference standards 4 h training 39 investigatorsRowland et al17 (2001) (n = 31) 93 89 Used IVUUSradiologist 3 days training 68 accuracy
NPV negative predictive value
S Watkins et al
192 copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
ureteral dilatation perinephric fluid and collecting sys-tem dilatation showed statistically significant changeover 8 h of study29 Therefore correlation discrepancy inour study might be explained by any of these factorsIt has also been reported that false-negatives on US areusually followed by uncomplicated spontaneous stoneemission3031
What level of minimum training is required to makeEPU an effective screening tool for hydronephrosisLanoix et al15 and Rosen et al16 trained EP for 4 h and5 h respectively with markedly different results asnoted earlier perhaps because of the very different ref-erence standards used in their studies Rowland et aldemonstrated an overall accuracy of 68 for EPU after3 days training whereas our study demonstrates anaccuracy of 81 after 1 h of focused training in renalscans following completion of an ACEM-accreditedworkshop17
From our results it would seem prudent to state thatalthough additional training and experience mightimprove the accuracy of EPU it will not supplant theuse of CT in the foreseeable future However despitesimilar results to ours for EPU accuracy previousauthors have suggested a place for EPU in the detectionof hydronephrosis21516
Australasian and American Colleges for EmergencyMedicine have published policies on the training andaccreditation of EP in focused assessment with sonog-raphy for trauma and abdominal aortic aneurysm12ndash14
However currently there are no guidelines for minimumtraining and accreditation in renal sonography for EP
Limitations
As patients were enrolled on a convenience basis dueto the presence or absence of an investigator to per-form the scans this might have introduced one ormore unknown biases A trend towards improvedinvestigator performance was noted as scan qualityimproved with experience this might have affectedthe detection rate in the earlier stages of the presentstudy Some EP performed less than three studiesothers more than 10 however the sample size was toosmall to afford meaningful subgroup analysis for indi-vidual EP Some of these limitations would be over-come by larger studies
It is worth noting that several of the EPU sono-graphers were relatively inexperienced in the use ofbedside US and were not yet accredited in otherEPU applications such as focused assessment by
sonography in trauma (FAST)abdominal aortic aneu-rysm (AAA) It is possible that accredited EPU sonog-raphers would be more accurate
Because of the dynamic nature of renal colic andurinary obstruction ideally all participants would havehad their EPU and CT scan performed within minimaltime delay to ensure an accurate assessment of EPU inone patient CT scanning was performed more than 24 hlater Finally our study did not include routine evalua-tion of renal resistive indexes that might improve detec-tion of early obstruction28
Conclusion
Using non-contrast CT as the gold standard we havefound EPU detection of hydronephrosis to have anaccuracy of 81 which is comparable to previousstudies However on the basis of the present studyEPU is probably not accurate enough to rule out hydro-nephrosis Further experience and training mightimprove the accuracy of EPU and allow its use as ascreening tool
Acknowledgements
The authors express their thanks to the staff specialistsand registrars from the Departments of EmergencyMedicine and Radiology for their assistance
Author contributions
SW contributed to study design ethics submission con-sent and patient information sheets reporting sheetsdata collation and analysis literature research andmanuscript preparation (85) JB contributed to origi-nal concept study design ethics submission investiga-tion manuscript preparation and supervision (40) PScontributed to investigator training study design andCT reporting (25) AH contributed to research meth-ods data analysis and manuscript preparation (20)AG contributed to study design investigation andmanuscript preparation (10) LC contributed to studydesign investigation and data collation (10)
Competing interests
Justin Bowra is a member of the Ultrasound Committeeof the Australasian College for Emergency MedicineAnna Holdgate holds the position of Section Editor of
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
193
Original Research for Emergency Medicine AustralasiaAll other authors declare no competing interests
Accepted 11 October 2006
References
1 Rosen P Barkin R Ling LJ Emergency Medicine Concepts andClinical Practice 4th edn St Louis Mosby 1998
2 Henderson SO Hoffner RJ Aragona JL Groth DE Esekogwu VIChan D Bedside emergency department ultrasonography plusradiography of the kidneys ureters and bladder versus intrave-nous pyelography in the evaluation of suspected ureteral colicAcad Emerg Med 1998 5 666ndash71
3 Eray O Cubuk MS Oktay CEM Yilmaz S Cete Y Ersoy FFThe efficacy of urinalysis plain films and spiral CT in EDpatients with suspected renal colic Am J Emerg Med 2003 21152ndash4
4 Brown DFM Rosen CL Sagarin M McCabe C Wolfe RE Impactof bedside ultrasonography by emergency physicians on theclinical likelihood of nephrolithiasis Ann Emerg Med 1996 27818
5 Fowler KAB Locken JA Duchesne JH Williamson MR Ultra-sound for detecting renal calculi with non-enhanced CT as areference standard Radiology 2002 222 109ndash13
6 Yilmaz S Sindel T Arslau G et al Renal colic comparison ofspiral CT US and IVU in the detection of ureteral calculi EurRadiol 1998 8 212ndash17
7 Spencer BA Wood BJ Dretler SP Helical CT and ureteric colicUrol Clin North Am 2000 27 231ndash41
8 Smith RC Verga M McCarthy S Rosenfield AT Diagnosis ofacute flank pain value of un-enhanced helical CT AJR Am JRoentgenol 1996 166 97ndash101
9 Sheafor DH Hertzberg BS Freed KS et al Non-enhanced helicalCT and US in the emergency evaluation of patients with renalcolic prospective comparison Radiology 2000 217 792ndash7
10 Patlas M Farkas A Fisher D Zaghal I Hadas-Halpern I Ultra-sound vs CT for the detection of ureteric stones in patients withrenal colic BJR 2001 74 901ndash4
11 Noble VE Brown DFM Renal ultrasound Emerg Med ClinNorth Am 2004 22 641ndash59
12 American College of Emergency Physicians (ACEP) Board ofDirectors Use of Ultrasound Imaging by Emergency PhysiciansPolicy Statement June 2001 Available from URL httpwwwaceporg16840html [Accessed August 2006]
13 Australasian College of Emergency Medicine (ACEM) CouncilCredentialling for ED Ultrasonography Policy Document P22July 2000 Available from URL httpwwwacemorgauinfo-centreaspxdocId=59 [Accessed August 2006]
14 Australasian College of Emergency Medicine (ACEM) CouncilUse of Bedside Ultrasound by Emergency Physicians Policy Doc-ument P21 July 1999 Available from URL httpwwwacemorgauinfocentreaspxdocId=59 [Accessed August 2006]
15 Lanoix R Leak LV Gaeta T Gernsheimer JR A preliminaryevaluation of emergency ultrasound in the setting of an emer-gency medicine training program Am J Emerg Med 2000 1841ndash5
16 Rosen CL Brown DFM Sagarin MJ Chang Y McCabe CJ WolfeRE Ultrasonography by emergency physicians in patients withsuspected renal colic J Emerg Med 1998 16 865ndash70
17 Rowland JL Kuhn M Bonnin RLL Davey MJ Langlois SLAccuracy of emergency department bedside ultrasonographyEmerg Med 2001 13 305ndash13
18 Grainger RG Allison DJ (eds) Diagnostic Radiology A Textbookof Medical Imaging 4th edn London Churchill Livingstone2001 p 1594
19 Kiely EA Hartnell GG Gibson RN Measurement of bladderVolume by real-time ultrasound Br J Urol 1987 60 33ndash5
20 Sinclair D Wilson S Toi A Greenspan L The evaluation ofsuspected renal colic ultrasound scan vs excretory urographyAnn Emerg Med 1989 18 556ndash9
21 Dalla Palma L Stacul F Bazzocchi M et al Ultrasonography andplain film versus intravenous urography in ureteric colic ClinRadiol 1993 47 333ndash6
22 Oumlzden E Karamuumlrsel T Gouml uuml Ccedil Yaman Ouml Inal T Gouml uuml ODetection rate of ureter stones with US relationship with gradeof hydronephrosis J Ankara Med Sch 2002 24 183ndash6
23 King L Hydronephrosis when is obstruction not obstructionCommon problems in paediatric urology Urol Clin North Am1995 22 31ndash42
24 Holdgate A Chan T How accurate are emergency clinicians atinterpreting non-contrast CT for suspected renal colic AcadEmerg Med 2003 10 315ndash19
25 Jeffrey RB Federle MP CT and ultrasonography of acute renalabnormalities Radiol Clin North Am 1983 21 515ndash25
26 Morse JW Saracino BS Melanson SW Arcona S Heller MBUltrasound interpretation of hydronephrosis is improved by abrief educational intervention Ann Emerg Med 1998 32(Suppl Pt 2) S27
27 Freed KS Paulson EK Frederick MG et al Interobserver vari-ability in the interpretation of unenhanced helical CT for thediagnosis of ureteral stones J Comput Assist Tomogr 1998 22732ndash7
28 Morse JW Hill R Greissinger WP Patterson JW Melanson SWHeller MB Rapid oral hydration results in hydronephrosis asdetermined by bedside ultrasound Ann Emerg Med 1999 34134ndash40
29 Varanelli MJ Coll DM Levine JA Rosenfield AT Smith RCRelationship between duration of pain and secondary signs ofobstruction of the urinary tract on unenhanced helical CT AJRAm J Roentgenol 2001 177 325ndash30
30 Haddad MC Sharif HS Abomelha MS Colour Doppler sonogra-phy and plain abdominal radiography in the management ofpatients with renal colic Eur Radiol 1994 4 529ndash32
31 Catalano O Nunziata A Altei F Siani A Suspected ureteralcolic primary helical CT versus selective helical CT after unen-hanced radiography and sonography AJR Am J Roentgenol2002 178 379ndash87
(g s
(g s
S Watkins et al
194 copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
Appendix I Validation of ED physician US diagnosing hydronephrosis in ureteric colicCompleted forms to be placed in the marked box in the Resuscitation Room Date
Time
ED Physician name
Right Place patient details sticker here Location of
symptoms (Please circle or
comment)
Left
ULTRASOUND FINDINGS
Hydronephrosis Present
Estimate severity
NO MILD MODERATE SEVERE UNSURE
RIGHT
LEFT
Additional comments
Empty Half Full
Bladder size
Incidental Findings (eg Free fluidascites AAA effusion etc)
Study Group Use only CT scan
Performed Liverpool Hospital South West
RadiologyElsewhere
Date amp Time of scan
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
195
Appendix II Validation of ED physician US diagnosing hydronephrosis in ureteric colicReporting sheet for dr praneal sharma radiologist
Date Time
Place patient details sticker here
Diagnosis of Renal ureteric colic correct YES NO
Calculus Seen YES NO
Left Right
Position of Calculus
Size of Calculus
CT KUB FINDINGS
Hydronephrosis Present
Estimate severity
NO MILD MODERATE SEVERE UNSURE
RIGHT
LEFT
Additional Findings
S Watkins et al
192 copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
ureteral dilatation perinephric fluid and collecting sys-tem dilatation showed statistically significant changeover 8 h of study29 Therefore correlation discrepancy inour study might be explained by any of these factorsIt has also been reported that false-negatives on US areusually followed by uncomplicated spontaneous stoneemission3031
What level of minimum training is required to makeEPU an effective screening tool for hydronephrosisLanoix et al15 and Rosen et al16 trained EP for 4 h and5 h respectively with markedly different results asnoted earlier perhaps because of the very different ref-erence standards used in their studies Rowland et aldemonstrated an overall accuracy of 68 for EPU after3 days training whereas our study demonstrates anaccuracy of 81 after 1 h of focused training in renalscans following completion of an ACEM-accreditedworkshop17
From our results it would seem prudent to state thatalthough additional training and experience mightimprove the accuracy of EPU it will not supplant theuse of CT in the foreseeable future However despitesimilar results to ours for EPU accuracy previousauthors have suggested a place for EPU in the detectionof hydronephrosis21516
Australasian and American Colleges for EmergencyMedicine have published policies on the training andaccreditation of EP in focused assessment with sonog-raphy for trauma and abdominal aortic aneurysm12ndash14
However currently there are no guidelines for minimumtraining and accreditation in renal sonography for EP
Limitations
As patients were enrolled on a convenience basis dueto the presence or absence of an investigator to per-form the scans this might have introduced one ormore unknown biases A trend towards improvedinvestigator performance was noted as scan qualityimproved with experience this might have affectedthe detection rate in the earlier stages of the presentstudy Some EP performed less than three studiesothers more than 10 however the sample size was toosmall to afford meaningful subgroup analysis for indi-vidual EP Some of these limitations would be over-come by larger studies
It is worth noting that several of the EPU sono-graphers were relatively inexperienced in the use ofbedside US and were not yet accredited in otherEPU applications such as focused assessment by
sonography in trauma (FAST)abdominal aortic aneu-rysm (AAA) It is possible that accredited EPU sonog-raphers would be more accurate
Because of the dynamic nature of renal colic andurinary obstruction ideally all participants would havehad their EPU and CT scan performed within minimaltime delay to ensure an accurate assessment of EPU inone patient CT scanning was performed more than 24 hlater Finally our study did not include routine evalua-tion of renal resistive indexes that might improve detec-tion of early obstruction28
Conclusion
Using non-contrast CT as the gold standard we havefound EPU detection of hydronephrosis to have anaccuracy of 81 which is comparable to previousstudies However on the basis of the present studyEPU is probably not accurate enough to rule out hydro-nephrosis Further experience and training mightimprove the accuracy of EPU and allow its use as ascreening tool
Acknowledgements
The authors express their thanks to the staff specialistsand registrars from the Departments of EmergencyMedicine and Radiology for their assistance
Author contributions
SW contributed to study design ethics submission con-sent and patient information sheets reporting sheetsdata collation and analysis literature research andmanuscript preparation (85) JB contributed to origi-nal concept study design ethics submission investiga-tion manuscript preparation and supervision (40) PScontributed to investigator training study design andCT reporting (25) AH contributed to research meth-ods data analysis and manuscript preparation (20)AG contributed to study design investigation andmanuscript preparation (10) LC contributed to studydesign investigation and data collation (10)
Competing interests
Justin Bowra is a member of the Ultrasound Committeeof the Australasian College for Emergency MedicineAnna Holdgate holds the position of Section Editor of
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
193
Original Research for Emergency Medicine AustralasiaAll other authors declare no competing interests
Accepted 11 October 2006
References
1 Rosen P Barkin R Ling LJ Emergency Medicine Concepts andClinical Practice 4th edn St Louis Mosby 1998
2 Henderson SO Hoffner RJ Aragona JL Groth DE Esekogwu VIChan D Bedside emergency department ultrasonography plusradiography of the kidneys ureters and bladder versus intrave-nous pyelography in the evaluation of suspected ureteral colicAcad Emerg Med 1998 5 666ndash71
3 Eray O Cubuk MS Oktay CEM Yilmaz S Cete Y Ersoy FFThe efficacy of urinalysis plain films and spiral CT in EDpatients with suspected renal colic Am J Emerg Med 2003 21152ndash4
4 Brown DFM Rosen CL Sagarin M McCabe C Wolfe RE Impactof bedside ultrasonography by emergency physicians on theclinical likelihood of nephrolithiasis Ann Emerg Med 1996 27818
5 Fowler KAB Locken JA Duchesne JH Williamson MR Ultra-sound for detecting renal calculi with non-enhanced CT as areference standard Radiology 2002 222 109ndash13
6 Yilmaz S Sindel T Arslau G et al Renal colic comparison ofspiral CT US and IVU in the detection of ureteral calculi EurRadiol 1998 8 212ndash17
7 Spencer BA Wood BJ Dretler SP Helical CT and ureteric colicUrol Clin North Am 2000 27 231ndash41
8 Smith RC Verga M McCarthy S Rosenfield AT Diagnosis ofacute flank pain value of un-enhanced helical CT AJR Am JRoentgenol 1996 166 97ndash101
9 Sheafor DH Hertzberg BS Freed KS et al Non-enhanced helicalCT and US in the emergency evaluation of patients with renalcolic prospective comparison Radiology 2000 217 792ndash7
10 Patlas M Farkas A Fisher D Zaghal I Hadas-Halpern I Ultra-sound vs CT for the detection of ureteric stones in patients withrenal colic BJR 2001 74 901ndash4
11 Noble VE Brown DFM Renal ultrasound Emerg Med ClinNorth Am 2004 22 641ndash59
12 American College of Emergency Physicians (ACEP) Board ofDirectors Use of Ultrasound Imaging by Emergency PhysiciansPolicy Statement June 2001 Available from URL httpwwwaceporg16840html [Accessed August 2006]
13 Australasian College of Emergency Medicine (ACEM) CouncilCredentialling for ED Ultrasonography Policy Document P22July 2000 Available from URL httpwwwacemorgauinfo-centreaspxdocId=59 [Accessed August 2006]
14 Australasian College of Emergency Medicine (ACEM) CouncilUse of Bedside Ultrasound by Emergency Physicians Policy Doc-ument P21 July 1999 Available from URL httpwwwacemorgauinfocentreaspxdocId=59 [Accessed August 2006]
15 Lanoix R Leak LV Gaeta T Gernsheimer JR A preliminaryevaluation of emergency ultrasound in the setting of an emer-gency medicine training program Am J Emerg Med 2000 1841ndash5
16 Rosen CL Brown DFM Sagarin MJ Chang Y McCabe CJ WolfeRE Ultrasonography by emergency physicians in patients withsuspected renal colic J Emerg Med 1998 16 865ndash70
17 Rowland JL Kuhn M Bonnin RLL Davey MJ Langlois SLAccuracy of emergency department bedside ultrasonographyEmerg Med 2001 13 305ndash13
18 Grainger RG Allison DJ (eds) Diagnostic Radiology A Textbookof Medical Imaging 4th edn London Churchill Livingstone2001 p 1594
19 Kiely EA Hartnell GG Gibson RN Measurement of bladderVolume by real-time ultrasound Br J Urol 1987 60 33ndash5
20 Sinclair D Wilson S Toi A Greenspan L The evaluation ofsuspected renal colic ultrasound scan vs excretory urographyAnn Emerg Med 1989 18 556ndash9
21 Dalla Palma L Stacul F Bazzocchi M et al Ultrasonography andplain film versus intravenous urography in ureteric colic ClinRadiol 1993 47 333ndash6
22 Oumlzden E Karamuumlrsel T Gouml uuml Ccedil Yaman Ouml Inal T Gouml uuml ODetection rate of ureter stones with US relationship with gradeof hydronephrosis J Ankara Med Sch 2002 24 183ndash6
23 King L Hydronephrosis when is obstruction not obstructionCommon problems in paediatric urology Urol Clin North Am1995 22 31ndash42
24 Holdgate A Chan T How accurate are emergency clinicians atinterpreting non-contrast CT for suspected renal colic AcadEmerg Med 2003 10 315ndash19
25 Jeffrey RB Federle MP CT and ultrasonography of acute renalabnormalities Radiol Clin North Am 1983 21 515ndash25
26 Morse JW Saracino BS Melanson SW Arcona S Heller MBUltrasound interpretation of hydronephrosis is improved by abrief educational intervention Ann Emerg Med 1998 32(Suppl Pt 2) S27
27 Freed KS Paulson EK Frederick MG et al Interobserver vari-ability in the interpretation of unenhanced helical CT for thediagnosis of ureteral stones J Comput Assist Tomogr 1998 22732ndash7
28 Morse JW Hill R Greissinger WP Patterson JW Melanson SWHeller MB Rapid oral hydration results in hydronephrosis asdetermined by bedside ultrasound Ann Emerg Med 1999 34134ndash40
29 Varanelli MJ Coll DM Levine JA Rosenfield AT Smith RCRelationship between duration of pain and secondary signs ofobstruction of the urinary tract on unenhanced helical CT AJRAm J Roentgenol 2001 177 325ndash30
30 Haddad MC Sharif HS Abomelha MS Colour Doppler sonogra-phy and plain abdominal radiography in the management ofpatients with renal colic Eur Radiol 1994 4 529ndash32
31 Catalano O Nunziata A Altei F Siani A Suspected ureteralcolic primary helical CT versus selective helical CT after unen-hanced radiography and sonography AJR Am J Roentgenol2002 178 379ndash87
(g s
(g s
S Watkins et al
194 copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
Appendix I Validation of ED physician US diagnosing hydronephrosis in ureteric colicCompleted forms to be placed in the marked box in the Resuscitation Room Date
Time
ED Physician name
Right Place patient details sticker here Location of
symptoms (Please circle or
comment)
Left
ULTRASOUND FINDINGS
Hydronephrosis Present
Estimate severity
NO MILD MODERATE SEVERE UNSURE
RIGHT
LEFT
Additional comments
Empty Half Full
Bladder size
Incidental Findings (eg Free fluidascites AAA effusion etc)
Study Group Use only CT scan
Performed Liverpool Hospital South West
RadiologyElsewhere
Date amp Time of scan
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
195
Appendix II Validation of ED physician US diagnosing hydronephrosis in ureteric colicReporting sheet for dr praneal sharma radiologist
Date Time
Place patient details sticker here
Diagnosis of Renal ureteric colic correct YES NO
Calculus Seen YES NO
Left Right
Position of Calculus
Size of Calculus
CT KUB FINDINGS
Hydronephrosis Present
Estimate severity
NO MILD MODERATE SEVERE UNSURE
RIGHT
LEFT
Additional Findings
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
193
Original Research for Emergency Medicine AustralasiaAll other authors declare no competing interests
Accepted 11 October 2006
References
1 Rosen P Barkin R Ling LJ Emergency Medicine Concepts andClinical Practice 4th edn St Louis Mosby 1998
2 Henderson SO Hoffner RJ Aragona JL Groth DE Esekogwu VIChan D Bedside emergency department ultrasonography plusradiography of the kidneys ureters and bladder versus intrave-nous pyelography in the evaluation of suspected ureteral colicAcad Emerg Med 1998 5 666ndash71
3 Eray O Cubuk MS Oktay CEM Yilmaz S Cete Y Ersoy FFThe efficacy of urinalysis plain films and spiral CT in EDpatients with suspected renal colic Am J Emerg Med 2003 21152ndash4
4 Brown DFM Rosen CL Sagarin M McCabe C Wolfe RE Impactof bedside ultrasonography by emergency physicians on theclinical likelihood of nephrolithiasis Ann Emerg Med 1996 27818
5 Fowler KAB Locken JA Duchesne JH Williamson MR Ultra-sound for detecting renal calculi with non-enhanced CT as areference standard Radiology 2002 222 109ndash13
6 Yilmaz S Sindel T Arslau G et al Renal colic comparison ofspiral CT US and IVU in the detection of ureteral calculi EurRadiol 1998 8 212ndash17
7 Spencer BA Wood BJ Dretler SP Helical CT and ureteric colicUrol Clin North Am 2000 27 231ndash41
8 Smith RC Verga M McCarthy S Rosenfield AT Diagnosis ofacute flank pain value of un-enhanced helical CT AJR Am JRoentgenol 1996 166 97ndash101
9 Sheafor DH Hertzberg BS Freed KS et al Non-enhanced helicalCT and US in the emergency evaluation of patients with renalcolic prospective comparison Radiology 2000 217 792ndash7
10 Patlas M Farkas A Fisher D Zaghal I Hadas-Halpern I Ultra-sound vs CT for the detection of ureteric stones in patients withrenal colic BJR 2001 74 901ndash4
11 Noble VE Brown DFM Renal ultrasound Emerg Med ClinNorth Am 2004 22 641ndash59
12 American College of Emergency Physicians (ACEP) Board ofDirectors Use of Ultrasound Imaging by Emergency PhysiciansPolicy Statement June 2001 Available from URL httpwwwaceporg16840html [Accessed August 2006]
13 Australasian College of Emergency Medicine (ACEM) CouncilCredentialling for ED Ultrasonography Policy Document P22July 2000 Available from URL httpwwwacemorgauinfo-centreaspxdocId=59 [Accessed August 2006]
14 Australasian College of Emergency Medicine (ACEM) CouncilUse of Bedside Ultrasound by Emergency Physicians Policy Doc-ument P21 July 1999 Available from URL httpwwwacemorgauinfocentreaspxdocId=59 [Accessed August 2006]
15 Lanoix R Leak LV Gaeta T Gernsheimer JR A preliminaryevaluation of emergency ultrasound in the setting of an emer-gency medicine training program Am J Emerg Med 2000 1841ndash5
16 Rosen CL Brown DFM Sagarin MJ Chang Y McCabe CJ WolfeRE Ultrasonography by emergency physicians in patients withsuspected renal colic J Emerg Med 1998 16 865ndash70
17 Rowland JL Kuhn M Bonnin RLL Davey MJ Langlois SLAccuracy of emergency department bedside ultrasonographyEmerg Med 2001 13 305ndash13
18 Grainger RG Allison DJ (eds) Diagnostic Radiology A Textbookof Medical Imaging 4th edn London Churchill Livingstone2001 p 1594
19 Kiely EA Hartnell GG Gibson RN Measurement of bladderVolume by real-time ultrasound Br J Urol 1987 60 33ndash5
20 Sinclair D Wilson S Toi A Greenspan L The evaluation ofsuspected renal colic ultrasound scan vs excretory urographyAnn Emerg Med 1989 18 556ndash9
21 Dalla Palma L Stacul F Bazzocchi M et al Ultrasonography andplain film versus intravenous urography in ureteric colic ClinRadiol 1993 47 333ndash6
22 Oumlzden E Karamuumlrsel T Gouml uuml Ccedil Yaman Ouml Inal T Gouml uuml ODetection rate of ureter stones with US relationship with gradeof hydronephrosis J Ankara Med Sch 2002 24 183ndash6
23 King L Hydronephrosis when is obstruction not obstructionCommon problems in paediatric urology Urol Clin North Am1995 22 31ndash42
24 Holdgate A Chan T How accurate are emergency clinicians atinterpreting non-contrast CT for suspected renal colic AcadEmerg Med 2003 10 315ndash19
25 Jeffrey RB Federle MP CT and ultrasonography of acute renalabnormalities Radiol Clin North Am 1983 21 515ndash25
26 Morse JW Saracino BS Melanson SW Arcona S Heller MBUltrasound interpretation of hydronephrosis is improved by abrief educational intervention Ann Emerg Med 1998 32(Suppl Pt 2) S27
27 Freed KS Paulson EK Frederick MG et al Interobserver vari-ability in the interpretation of unenhanced helical CT for thediagnosis of ureteral stones J Comput Assist Tomogr 1998 22732ndash7
28 Morse JW Hill R Greissinger WP Patterson JW Melanson SWHeller MB Rapid oral hydration results in hydronephrosis asdetermined by bedside ultrasound Ann Emerg Med 1999 34134ndash40
29 Varanelli MJ Coll DM Levine JA Rosenfield AT Smith RCRelationship between duration of pain and secondary signs ofobstruction of the urinary tract on unenhanced helical CT AJRAm J Roentgenol 2001 177 325ndash30
30 Haddad MC Sharif HS Abomelha MS Colour Doppler sonogra-phy and plain abdominal radiography in the management ofpatients with renal colic Eur Radiol 1994 4 529ndash32
31 Catalano O Nunziata A Altei F Siani A Suspected ureteralcolic primary helical CT versus selective helical CT after unen-hanced radiography and sonography AJR Am J Roentgenol2002 178 379ndash87
(g s
(g s
S Watkins et al
194 copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
Appendix I Validation of ED physician US diagnosing hydronephrosis in ureteric colicCompleted forms to be placed in the marked box in the Resuscitation Room Date
Time
ED Physician name
Right Place patient details sticker here Location of
symptoms (Please circle or
comment)
Left
ULTRASOUND FINDINGS
Hydronephrosis Present
Estimate severity
NO MILD MODERATE SEVERE UNSURE
RIGHT
LEFT
Additional comments
Empty Half Full
Bladder size
Incidental Findings (eg Free fluidascites AAA effusion etc)
Study Group Use only CT scan
Performed Liverpool Hospital South West
RadiologyElsewhere
Date amp Time of scan
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
195
Appendix II Validation of ED physician US diagnosing hydronephrosis in ureteric colicReporting sheet for dr praneal sharma radiologist
Date Time
Place patient details sticker here
Diagnosis of Renal ureteric colic correct YES NO
Calculus Seen YES NO
Left Right
Position of Calculus
Size of Calculus
CT KUB FINDINGS
Hydronephrosis Present
Estimate severity
NO MILD MODERATE SEVERE UNSURE
RIGHT
LEFT
Additional Findings
S Watkins et al
194 copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
Appendix I Validation of ED physician US diagnosing hydronephrosis in ureteric colicCompleted forms to be placed in the marked box in the Resuscitation Room Date
Time
ED Physician name
Right Place patient details sticker here Location of
symptoms (Please circle or
comment)
Left
ULTRASOUND FINDINGS
Hydronephrosis Present
Estimate severity
NO MILD MODERATE SEVERE UNSURE
RIGHT
LEFT
Additional comments
Empty Half Full
Bladder size
Incidental Findings (eg Free fluidascites AAA effusion etc)
Study Group Use only CT scan
Performed Liverpool Hospital South West
RadiologyElsewhere
Date amp Time of scan
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
195
Appendix II Validation of ED physician US diagnosing hydronephrosis in ureteric colicReporting sheet for dr praneal sharma radiologist
Date Time
Place patient details sticker here
Diagnosis of Renal ureteric colic correct YES NO
Calculus Seen YES NO
Left Right
Position of Calculus
Size of Calculus
CT KUB FINDINGS
Hydronephrosis Present
Estimate severity
NO MILD MODERATE SEVERE UNSURE
RIGHT
LEFT
Additional Findings
Hydronephrosis study
copy 2007 The AuthorsJournal compilation copy 2007 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine
195
Appendix II Validation of ED physician US diagnosing hydronephrosis in ureteric colicReporting sheet for dr praneal sharma radiologist
Date Time
Place patient details sticker here
Diagnosis of Renal ureteric colic correct YES NO
Calculus Seen YES NO
Left Right
Position of Calculus
Size of Calculus
CT KUB FINDINGS
Hydronephrosis Present
Estimate severity
NO MILD MODERATE SEVERE UNSURE
RIGHT
LEFT
Additional Findings