Arts & Culture

Diagnostic value of renal parenchymal density difference on unenhanced helical computed tomography scan in acutely obstructing ureteral stone disease

Description
Diagnostic value of renal parenchymal density difference on unenhanced helical computed tomography scan in acutely obstructing ureteral stone disease
Categories
Published
of 4
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  DIAGNOSTIC VALUE OF RENAL PARENCHYMAL DENSITY DIFFERENCE ON UNENHANCED HELICAL COMPUTEDTOMOGRAPHY SCAN IN ACUTELY OBSTRUCTINGURETERAL STONE DISEASE CANER O¨ZER, ESIN YENCILEK, F. DEMIR APAYDIN, MELTEM NASS DUCE, ALTAN YILDIZ,ERIM ERDEM, ARZU KANIK,  AND  SELAHITTIN C¸AYAN  ABSTRACT Objectives.  To evaluate the diagnostic value of the renal parenchymal density difference, detected usingunenhanced helical computed tomography, as a secondary sign of acute obstruction due to a ureteral stone. Methods.  Fifty-five patients with acute flank pain, in whom a ureteral stone was detected on the symp-tomatic side using unenhanced helical computed tomography, and 22 control subjects with no urinary stonedisease were included in this prospective study. Computed tomography was performed, starting from theupper poles of the kidneys down to the base of the urinary bladder. The renal parenchymal density wasmeasured in the upper, middle, and lower portions of each kidney, and a mean value was calculated. Thedifference between the mean values of the two kidneys was used to predict the presence of an acutelyobstructing ureteral stone. Results.  In 49 patients with a ureteral stone (89.1%), the difference between the parenchymal densities of the obstructed and nonobstructed kidneys was 5 Hounsfield units (HU) or greater and was lower on theobstructed side. In the remaining 6 patients (10.9%), the density difference was less than 5 HU but was stilllower on the obstructed side. All subjects in the control group had a density difference of less than 5 HU. A renal parenchymal density difference of 5.0 HU or greater had 89.1% sensitivity, 100% specificity, 100%positive predictive value, 85.7% negative predictive value, and 93.4% accuracy in predicting the presenceof an acute obstructing ureteral stone. Conclusions.  These data suggest that the renal parenchymal density difference may be a valuable second-ary sign of acute obstructing ureteral stone disease.  UROLOGY  64:  223–227, 2004. © 2004 Elsevier Inc. A cute flank pain is a common abdominal emer-gency. Several recent studies have shown thatunenhancedhelicalcomputedtomography(CT)isa safe, rapid, useful, and highly accurate techniquein the evaluation of patients with acute flank pain,sparing the patients the risk or discomfort associ-atedwithintravenouscontrastmedia. 1–4 Thistech-nique has a sensitivity of 95%, specificity of 98%,and accuracy of 97% for the diagnosis of ureteralstone disease. 5 The stones can be visualized di-rectly on CT most of the time. However, if a stoneisnotreadilyidentified,secondaryCTsignsofure-teralobstructioncanbeimportantdiagnosticcluesfor the presence of stone disease. The secondarysigns of ureterolithiasis consist of hydroureter(67% to 92%), hydronephrosis (69% to 86%), per-inephric stranding (65% to 83%), ipsilateralnephromegaly(69%),andvisibleureteralwallsur-roundingthecalculus,theso-calledtissue-rimsign(69%). These signs have been reported to havehigh sensitivity, specificity, positive and negativepredictive values, and accuracy in predicting ure-teral stone disease. 2,6,7 The difference in the renal parenchymal densitybetweentheacutelyobstructedandnonobstructedkidneyhasbeensuggestedasanadditionalsecond-arysignandhasrecentlybeendescribed. 8 Thepur-pose of this study was to investigate whether therenal parenchymal density difference can be a reli- From the Departments of Radiology, Urology, and Biostatistics,Mersin University Faculty of Medicine, Mersin, TurkeyReprint requests: Caner O¨ zer, M.D., Department of Radiology,Mersin U ¨ niversitesi Tıp Faku¨ltesi Hastanesi, I ˙hsaniye Mah., 123Cad., 4931 Sok, Mersin TR-33079, TurkeySubmitted:January22,2004,accepted(withrevisions):March16, 2004  ADULT UROLOGY  © 2004 E LSEVIER   I NC . 0090-4295/04/$30.00  ALL RIGHTS RESERVED  doi:10.1016/j.urology.2004.03.058  223  able secondary sign of acute obstruction due toureteral stone disease. MATERIAL AND METHODS From April 2002 to August 2003, patients with a clinicalsuspicion of an acutely obstructing ureteral stone were exam-ined with unenhanced helical CT, and the findings were eval-uated prospectively. Patients with a condition that mightasymmetrically affect one kidney, such as sarcoidosis, renalcarcinoma, or a history of surgery or radiotherapy were ex-cluded. A total of 55 patients (40 men and 15 women), inwhom a ureteral stone was detected on the symptomatic side,constituted the study group. The mean age was 41.8 years(range 8 to 74). Twenty-two subjects, who had been admittedto the emergency department and eventually diagnosed withacute appendicitis, acute cholecystitis, or acute pancreatitis,servedasthecontrolgroup.Noneofthesubjectsinthecontrolgroup had urinary complaints.HelicalCTexaminationswereperformedwiththeSomatomAR Star 40 scanner (Siemens Medical Systems, Erlangen, Ger-many) using 5-mm collimation, an incremental table speed of 5 mm/s, pitch of 1, 130 kVp, and 80 mA. The data were ac-quiredfromthetopofthekidneystothebottomofthebladderduring two or three breath holds, without using intravenousor oral contrast media. All images were evaluated at a soft-tissue setting with window width of 300 Hounsfield units(HU) and window level of 40 HU. The scans were examinedforurinarystonedisease,aswellasforanyotherextra-urinaryabnormality. Then the parenchymal densities of both kidneyswere measured. The measurements were performed from theupper poles, middle portions, and lower poles of each kidney,sparing the renal sinus complex and using the same region of interest value (area 0.4 cm 2 ). After obtaining the mean paren-chymal density of each kidney, the density difference betweenthe obstructed and nonobstructed kidney was calculated.The patients were informed about the procedure before theexamination and all provided written informed consent. Thesame radiologists performed the examinations, measure-ments, and interpretations.The statistical analyses were performed using the indepen-dent sample  t  test to compare the renal parenchymal densitydifferences of the two kidneys between the study and controlgroups. The paired  t  test was also used to compare the renalparenchymal density differences of the two kidneys of thesamepatientinbothgroups.Thereceiveroperatingcharacter-isticcurvewasusedtoshowthepredictiveperformanceoftheabsolute values of the renal parenchymal density differences.The sensitivity, specificity, positive and negative predictivevalues, and accuracy of the parenchymal density differencebetween the obstructed and nonobstructed kidneys of the twogroups were calculated. The parenchymal density values aregiven as the mean  the standard deviation. RESULTS The descriptive statistics of this study are givenin Table I. The mean renal parenchymal density difference is given as the average of the right andleftkidneysofthecontrolgroup.Inpatientswithaureteral stone, the mean parenchymal density of the obstructed kidneys was 27.7  2.9 HU (range18.3 to 32.6) and that of the nonobstructed kid-neys was 34.4  2.5 HU (range 28.3 to 38.9). Themean renal parenchymal density difference be-tween the obstructed and nonobstructed kidneywas6.7HU(range0.6to13.2HU).Thisdifferencewas statistically significant ( P  0.0001). The pa-renchymal density of the obstructed kidney wasalways lower than that of the nonobstructed kid-ney. In 49 (89.1%) of 55 patients with ureteralstone disease, the difference between the paren-chymal densities was 5 HU or more (Fig. 1), al- though the difference was less than 5 HU in 6 pa-tients (10.1%).The mean parenchymal density of the right andleft kidneys of the control group was 33.9    2.7HU (range 27.2 to 38.1) and 34.0  2.9 HU (range27.9to38.1),respectively,withtheaverage34.0  2.7 HU (range 27.2 to 38.1). The renal parenchy-mal density difference was 0.7 HU (range 0.1 to1.5;Fig.2).Thedifferencewasnotstatisticallysig-nificant ( P  0.506).Themeanrenalparenchymaldensitiesoftheob-structed kidneys, nonobstructed kidneys, and kid-neysofthecontrolgrouparepresentedinFigure3.The difference in the parenchymal densities be-tween the obstructed kidneys and the kidneys in  TABLE I.  Descriptive statistics of renal parenchymal densities Group n MeanStandardDeviation95% ConfidenceInterval for Mean Minimum Maximum Controls 22 34.0 2.7 32.8–35.2 27.20 38.13Nonobstructed 55 34.4 2.5 33.7–35.1 28.30 38.90Obstructed 55 27.7 2.9 26.9–28.5 18.30 32.60 FIGURE 1.  Axial unenhanced helical CT scan obtainedat middle portion of kidneys of 52-year old patient with left proximal ureteral stone. Attenuation of right and left  kidney 36.7 HU and 27.0 HU, respectively. Nephro- megaly,perinephricstranding,andhydronephrosisalso seen on left side. 224  UROLOGY 64  (2), 2004  the control group was statistically significant ( P  0.0001). However, no statistically significant dif-ference was found in the parenchymal densitiesbetween the nonobstructed kidneys of the patientsand the kidneys in the control group ( P  0.523).The receiver operating characteristic curve anal-ysis showed that the renal parenchymal densitydifferencehadhighaccuracy(areaunderthecurve0.995,  P  0.0001; Fig. 4). The overall statisticalanalysis revealed that, with a cutoff value of 5 HU,the renal parenchymal density difference had asensitivity of 89.1%, specificity of 100%, positivepredictivevalueof100%,negativepredictivevalueof 85.7%, and accuracy of 93.4% in predicting thepresence of an acute obstructing ureteral stone. COMMENT Secondary CT signs of ureteral obstruction areoftenpresentandusefulincasesinwhichastoneisnot readily identified. 3 The results of the presentstudy have shown that the renal parenchymal den-sity difference might be an additional and compli-mentary sign to the previously accepted ones. Thisnew sign was first suggested by Georgidas  et al. 8 Itis based on the anatomic changes that occur in therenal parenchyma in the presence of acute uret-erolithiasis. The physiologic adaptation process of a kidney that is subjected to acute obstruction isthe reason for these anatomic changes. Whenacutely obstructed, the kidney responds to the in-creased ureteral pressure by the diffusion of urinetotherenalinterstitiumthroughseveralpathways,including pyelosinus, pyelotubular, pyelolymphatic,and pyelovenous back flow mechanisms. 1,9–14 Thedensity difference between the renal parenchymaof an obstructed and nonobstructed kidney is theresultofthisacuteadaptationanddefensivemech-anism and is reported to occur more frequentlywith acute and complete obstruction than with in-complete obstruction. 13 In our study, the diagnostic value of the renalparenchymaldensitydifferenceasasecondarysignof calculus-induced acute ureteral obstruction wasassessed. In the patients with an acutely obstruct-ing ureteral stone, the renal parenchymal densityof the symptomatic side was always less than thatof the asymptomatic side. This difference, with acutoff value of 5 HU, was statistically significant in49patients(89.1%)andwasnotstatisticallysignif-icant in 6 (10.9%). This nonstatistically significantdifferenceinthe6patientscouldhavebeenaresultof either the timing of the examination or individ-ual differences among the patients. In the study byVaranelli  et al., 13 all secondary CT signs of ureteral FIGURE 2.  Axial unenhanced helical CT scan of 32- year old patient from the control group with final diag- nosis of acute appendicitis. Parenchymal densities of  kidneys were not significantly different (36.2 HU on right and 35.2 HU on left). FIGURE 3.  Mean renal parenchymal densities of ob- structed kidneys, nonobstructed kidneys, and kidneysof control group. FIGURE 4.  Receiver operating characteristic curve of  renal parenchymal density differences for obstructed kidneys. UROLOGY  64  (2), 2004  225  obstruction, except for nephromegaly, showed astatistically significant increase in frequency as theduration of flank pain increased. Additional stud-ies, especially concerning the time between the ex-amination and the beginning of flank pain, areneeded to explain this controversy.Acute ureteral obstruction is not the only reasonfor the density difference between kidneys. In ad-dition to obstructive conditions, nonobstructiveconditions, such as acute pyelonephritis and renalvein thrombosis, can also result in decreased renalparenchymaldensity.Moreimportantly,arecentlypassed calculus must be considered in the evalua-tion of patients with a statistically significant den-sity difference if a calculus cannot be demonstrat-ed. 8 No such patient was encountered in our studygroup.Our study was the second study to assess thediagnostic value of renal parenchymal differencesin obstructed and nonobstructed kidneys after thestudyofGeorgidas et al. 8 Thefindingsweresimilarinbothstudies.However,becauseourstudygroupwas larger, we were also able to evaluate the sensi-tivity and specificity. In the control group, no sta-tistically significant difference (less than 5 HU)was found between the parenchymal densities of the kidneys. Therefore, the specificity and positivepredictive value were both 100%. However, thesensitivity and negative predictive value were rela-tively low because of the false-negative results duetothecutoffvalueof5HUorgreater.Toovercomethisproblem,alowercutoffvalue(eg,3HU)couldbe selected. However, in that case, the false-posi-tiveresultswouldincrease,negativelyaffectingthespecificity and positive predictive value. CONCLUSIONS The results of this study suggest that a renal pa-renchymal density difference of 5 HU or greatercan be a useful secondary sign to differentiate thecause of an acutely obstructed kidney from thenonobstructedone.Inaminorityofpatientswithaureteral stone, in whom the renal parenchymaldensitydifferenceislessthan5HU,combiningthissign with other secondary signs may be helpful toovercome this handicap. REFERENCES1. Boridy IC, Kawashima A, Goldman SM,  et al:  Acuteureterolithiasis: nonenhanced helical CT findings of peri-nephricedemaforpredictionofdegreeofureteralobstruction.Radiology  213:  663–667, 1999.2. Katz DS, Lane MJ, and Sommer FG: Unenhanced heli-cal CT of ureteral stones: incidence of associated urinary tractfindings. AJR Am J Roentgenol  166:  1319–1322, 1996.3. Smith RC, Verga M, McCarty SM,  et al:  Diagnosis of acute flank pain: value of unenhanced helical CT. AJR Am JRoentgenol  166:  97–101, 1996.4. Yilmaz S, Sindel T, Arslan G,  et al:  Renal colic: compar-ison of spiral CT, US and IVU in the detection of ureteralcalculi. Eur Radiol  8:  212–217, 1998.5. Smith RC, Dalrymple NC, and Neitlich J: Noncontrasthelical CT in the evaluation of acute flank pain. Abdom Imag-ing  23:  10–16, 1998.6. Lanoue MZ, and Mindell HJ: The use of unenhancedhelical CT to evaluate suspected renal colic. AJR Am J Roent-genol  169:  1579–1584, 1997.7. Smith RC, Verga M, Dalrymple N,  et al:  Acute ureteralobstruction: value of secondary signs on helical unenhancedCT. AJR Am J Roentgenol  167:  1109–1113, 1996.8. Georgidas CS, Moore CJ, and Smith DP: Differences of renal parenchymal attenuation for acutely obstructed and un-obstructed kidneys on unenhanced helical CT: a useful sec-ondary sign? AJR Am J Roentgenol  176:  965–968, 2001.9. Gillenwater JY: The pathophysiology of urinary tractobstruction, in Walsh PC, Retik AB, Stamey TA,  et al  (Eds): Campbell’s Urology , 6th ed. Philadelphia, WB Saunders, 1992,vol 1, pp 499–532.10. Rose JG, and Gillenwater JY: Pathophysiology of ure-teral obstruction. Am J Physiol  225:  830–837, 1973.11. Smith RC, Levine J, and Rosenfield AT: Helical CT of urinary tract stones epidemiology, srcin, pathophysiology,diagnosis, and management. Radiol Clin North Am  37:  911–952, 1999.12. Takahashi N, Kawashima A, Ernst RD,  et al:  Ureteroli-thiasis: can clinical outcome be predicted with unenhancedhelical CT? Radiology  208:  97–102, 1998.13. Varanelli MJ, Coll DM, Levine JA,  et al:  Relationshipbetween duration of pain and secondary signs of the urinarytract on unenhanced helical CT. AJR Am J Roentgenol  177: 325–330, 2001.14. Vaughan ED, Shenasky JH, and Gillenwater JY: Mech-anism of acute hemodynamic response to ureteral occlusion.Invest Urol  9:  109–118, 1971.EDITORIAL COMMENTNoncontrast enhanced helical CT scanning will detect uri-nary tract stones in roughly 95% to 98% of patients who havethem. 1,2 Making an accurate diagnosis in the remaining 2% to5% of patients with undetected urinary tract stones representsa diagnostic challenge. In these uncommon cases, urologistsand radiologists must rely on “secondary” signs of ureteralobstruction. Historically, these signs have included hydrone-phrosis, hydroureter, unilateral renal enlargement, perineph-ric stranding, periureteral stranding, and the ureteral tissuerim sign. Of these signs, however, only hydronephrosis andhydroureter occur with relative frequency (70% to 90% of cases) and, hence, have a reasonably high sensitivity. 3 Theother signs occur with less frequency and are, therefore, onlyhelpful if they are present (high specificity but low sensitiv-ity).The authors should be commended for confirming the va-lidity of, perhaps, the most sensitive, as well as specific, sec-ondary sign for acute ureteral obstruction, a renal parenchy-maldensitydifferenceofmorethan5HUonnonenhancedCT(sensitivity 89.1% and specificity 100%). This concept wassrcinally described by Georgidas  et al. 4 who studied 92 pa-tients with a presumptive diagnosis of acute renal colic andwhohadundergonehelicalCTastheirprimaryimagingstudy.Of this group, 30 patients (33%) had an obstructive ureteralcalculus and 62 patients (67%) had no visible stone. Of thepatients with an obstructing stone, all but one had lower den-sities in the obstructed kidney compared with the nonob-structed kidney, with an average difference in attenuation of 14.5  10 HU. Furthermore, in the 25% of patients who un-derwent additional imaging studies because of the nonvisual- 226  UROLOGY 64  (2), 2004
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks