1Slobodanka Lj. Beatovi� MD, PhD,

2Marija Radulovi� MD, MSc, 3Milica M. Jankovi� PhD,

1Vera M. Artiko MD, PhD, 2Boris Ajdinovi� MD, PhD,

1Dragana P. �obi�-�aranovi� MD, PhD

1. University of Belgrade, Faculty of Medicine, Center for Nuclear Medicine, Clinical Center of Serbia2. Military Medical Academy, Institute of Nuclear Medicine, Belgrade, Serbia 3. University of Belgrade, Faculty of Electrical Engineering, Belgrade, Serbia

Keywords: Radionuclide renography 99m- Tc-DTPA -Output e�ciency

-Normalized residual activity

-Conventional interpretation of the

renogram

Corresponding author: Dr Slobodanka Beatovi� MD, PhD,University of Belgrade Faculty of Medicine, Center for Nuclear Medicine Clinical Center of Serbia, Vi�egradska 26, 11000 Belgrade, SerbiaTel: +381637019989 beatovic.boba@gmail.com

Rece�ved: 12 April 2018 Accepted revised: 21 June 2018

Renal output efficiency and normalized residual activity

examined by technetium-99m-DTPA renography have

by far greater specificity to diagnose obstructive disease

as compared to other conventional parameters of the

renogram. First such study of output efficiency

AbstractObjective: The aim of this study was to assess the diagnostic ability of renal output e�ciency (OE), norma-lized residual activity (NORA) and conventional interpretation of the diethylene triamine pentaacetic acid

99m( Tc-DTPA) diuretic renogram (CIR) in diagnosing urine �ow obstruction. Subjects and Methods: Group A consisted of 73 obstructed kidneys and Group B of 80 kidneys with dilated upper urinary tract. Forty eight kidneys were examined as the control Group C. A 40min acquisition with 240 of 10sec images was applied. Furosemide was administered after 20min (F+20). Post-void image was acquired at 50min. Output e�ci-ency was calculated at 20min (OE ) and 20min after furosemide test (OE ) and NORA at 20min (NORA ) 20 40 20

and on the post-micturition acquisition (NORA ). Results: Both the above parameters (OE and NORA ) PM 40 PM

had high sensitivity (92% and 97%), speci�city (99% for both) and accuracy (95% and 98%) in di�erenti-ating between obstructed and dilated unobstructed kidneys. The test of NORA correctly reclassi�ed 40 PM

out of 41 indeterminate or false positive results of the conventional interpretations of renograms (CIR) as obstructed (12 cases) or dilated (28 cases). Cut-o� values for obstruction were <80% and �0.23 for OE and 40

NORA , respectively. Compared to CIR, both parameters showed better speci�city, especially NORA . PM PM99mConclusion: Factors of OE and especially NORA of the renogram by Tc-DTPA were able to diagnose 40 PM

urine out�ow obstruction better than CIR. Speci�cally, these factors well di�erentiated obstruction from urinary tract dilatation.

Hell J Nucl Med 2018; 21(2): 140-144 Epub ahead of print: 12 July 2018 Published online: 10 August 2018

Introduction

Radionuclide diuretic renography is a non-invasive test for the assessment of ob-structed or equivocally obstructed upper urinary tract. After the anatomic diag-nosis of a dilated urinary tract has been made, renography may diagnose or ex-

clude the presence of renal obstruction and suggest the kind of treatment [1]. However, reduced renal function can make it di�cult to distinguish between obstruction and uri-nary tract dilatation with false-positive (FP) or indeterminate results on the usual re-nogram analysis [2]. Several methods have been proposed for overcoming this di�culty. One of them is the renal output e�ciency (OE) and of the normalized residual kidney ac-tivity NORA [3, 4]. These parameters have been reported in several studies, mainly using

99m 99mtechnetium-99m mercaptoacetyltriglycine ( Tc-MAG ) and not Tc-diethylene tria-3

mine pentaacetic acid (DTPA) [5-7]. Furthermore, the time of injecting the diuretic is not 99mthe same in all studies. In this study in a large number of patients we used Tc-DTPA and

a speci�c time for injecting furosemide and studied the diagnostic ability of OE, NORA and of the other interpretation of the conventional renogram (CIR) data in diagnosing urine �ow obstruction.

Patients and Methods

Between November 2011 and July 2014, 143 patients (age range: 18-82), 71 males and 72 females with signs and/or symptoms of renal obstruction were enrolled in this study. The clinical suspicion of obstruction was based on symptoms of mild to severe pain and/

Research Article

93Hellenic Journal of Nuclear Medicine May-August 2018• www.nuclmed.gr 140

or �nding of a dilated renal collecting system on ultrasound (US) examination and/or intravenous (i.v.) urography (IVU) or computerized tomography (CT) of abdomen or magnetic re-sonance urography (MRU) and renal function tests including serum creatinine level in all cases, glomerular �ltration rate estimated by Gates method in patients with chronic kidney disease (CKD). The exclusion criteria were: recent renal or ure-teric surgery, vesicoureteric re�ux (VUR) grade III-V and the persistence of nephrostomy tube.

For the control Group, 48 healthy subjects who were candi-dates for kidney donation were examined (age range: 31-67), 23 males and 25 females. They had no structural abnormality of their kidneys on the US examination, biochemical tests were normal and had no history of renal disease or hyperten-sion.

Each subject received 500-1000mL water 60 minutes befo-re the study and emptied the urine bladder just before the acquisition. A large �eld of view �-camera (Siemens Orbiter 7500) was set up with a low-energy all-purpose collimator. The patients were studied in the supine position, so that the left ventricle and both kidneys were scanned. A 40min acqu-isition protocol with 240 of 10-sec each images in a 128x128

99mmatrix size was applied. The i.v. dose of Tc-DTPA was bet-ween 74-200MBq adjusted for body weight [2]. Furosemide in a dose of 40mg was administered i.v. after twenty minutes (F+20 protocol), and a post-void static image of 1 minute du-ration was acquired 50 minutes after the beginning of the study.

Regions of interest (ROI) were drawn over the left ventricle and both kidneys. For background correction, a perirenal area was used. From the generated renograms, the time to maximum activity (T ) and to half maximum (T ) were max 1/2

calculated. Di�erential renal function (DRF) was determined using the Rutland-Patlak plot (RP) method [8].

Two experienced nuclear medicine physicians evaluated the conventional renogram diuretic study to assess urine drainage. They analyzed visually and/or calculated the dyna-mic images, the post-void images, the renogram curve, the T and T values and reported each renogram as obstruc-max 1/2

tive, non-obstructive and indeterminate, if needed by con-sensus. The indeterminate renograms were characterized by the absence of signi�cant drainage on the dynamic and post-void images after furosemide that is the slow and incomplete downslope phase of the renogram curve [5].

For the OE and NORA, the International Atomic Energy Agency (IAEA) software package for analysis of scintigraphic renal dynamic studies was used [9]. This calculation was per-formed without knowing the report of the renogram. Both parameters were determined at 2 time points: OE at 20min (OE ) and at 40min (20min after furosemide injection, OE ), 20 40

NORA at 20min (NORA ) and on the post-micturition acqu-20

isition (NORA ). PM

The urologists determined the diagnosis by using data from surgery, histopathology (HP), invasive diagnostic pro-cedures (cystoscopy, retrograde pyelography) and other imaging tests (US, IVU, CT, MRU). The diagnosis was obstruc-tion (Group A) and dilatation without obstruction (Group B). The results of OE and NORA were compared to the CIR. Pati-ents were followed-up for at least 6 months.

Statistical analysisDescriptive and analytical statistics (SPSS version 20) were used. The default level of signi�cance has been put below 0.05. We used the one-way ANOVA to evaluate the di�eren-ces between Groups. The unpaired t-test was used to com-pare the values between Groups A and B.

To determine the e�cacy of the parameters of renal was-hout and distinguish between obstruction and dilatation, receiver operating characteristics (ROC) curve analysis was performed. The sensitivity, speci�city, the area under the curve (AUC) with 95% con�dence interval, and the optimal cut-o� values were analyzed.

For validation of OE , NORA and CIR McNemar test was 40 PM

used to examine the di�erence between respective �n-dings, whereas kappa (�) coe�cient was used to determine the agreement between the 2 methods. The sensitivity, spe-ci�city, accuracy, positive predictive value (PPV) and nega-tive predictive value (NPV) were calculated using the stan-dard formulas.

Results

SubjectsThe control Group C consisted of 48 subjects with 96 normal renal units (RU) (Table 1). A hundred and forty three patients had 281 functioning RU. In 5 patients, nephrectomy of one of the kidneys had been previously done due to obstructive renal pathology. Out of 281RU, 153 kidneys presented with suspected upper out�ow obstruction. The underlying pat-hology were: renal calculosis (54%), pelviureteric junction (PUJ) narrowing (40%) and obstructive megaureters (5%). Seventy three kidneys had obstruction and were classi�ed into Group A, while Group B consisted of 80 kidneys with di-latation without obstruction (Table 2).

In total, 191 subjects and 249RU were studied.

Parameters of renal washout in control subjectsThe results for OE , OE , NORA and NORA are shown in 20 40 20 PM

Table 1.

Research Article

993 Hellenic Journal of Nuclear Medicine May-August 2018• www.nuclmed.gr141

Table 1 . OE and NORA in normal kidneys.

Oe 20

(%)Oe 40

(%)

NORA20

NORAPM

DRF (%)

N 96 96 96 96 96

Mean 87.85 94.09 0.63 0.07 50.00

SD 2.27 1.04 0.10 0.02 3.22

Mini-mum

83 91 0.44 0.04 43

Maxi-mum

93 96 0.90 0.11 57

Parameters of renal washout in patientsIn the patients' Groups (A and B), the parameters analyzed were OE and NORA , which were representative of the 40 PM

kidney response to forced diuresis after administration of furosemide (Table 2).

The one-way ANOVA comparison between the Groups, taking normal Group C as reference, showed signi�cant di�e-rence for the OE and NORA (P<0.001).40 PM

Conventional interpretation of the renogram The obstructive pattern was observed on CIR in 73 curves. Documented clinical diagnosis (DCD) con�rmed obstruction in 61 cases, whereas 12 CIR were FP for obstruction. The FP �ndings were in cases of CKD associated with low DRF, soli-tary kidney, or pelvicalyceal dilatation. Non-obstructive pat-tern was seen on CIR in 51 curves.

There were 41 indeterminate �ndings by CIR. In 12 of them NORA revealed obstruction. All kidneys were with severely PM

reduced DRF and in 6 cases CKD was also present. In 28 inde-terminate CIR �ndings, our �ndings from NORA showed PM

patent urinary tract. Most of these kidneys had grossly dila-ted pelvis but preserved DRF.

One hundred and twelve kidneys (73%) were correctly diagnosed on CIR. The di�erence between CIR and DCD was signi�cant (P=0.012), with the � coe�cient for agreement of 0.468.

Conventional interpretation of the renogram The ability of the OE and NORA to discriminate between 40 PM

obstructed and patent urinary tract were analyzed by ROC curve analysis (Table 3).

The results of comparison between OE /DCD and NORA 40 PM

/DCD are shown in Table 4. OE correctly discriminated bet-40

ween obstruction/non-obstruction in 95% RU. The di�eren-ce between OE /DCD was not signi�cant (P=0.453), with the 40

� coe�cient for agreement of 0.908. The results of NORA PM

showed 99% of accurate diagnoses. The di�erence between NORA /DCD was not signi�cant (P=1.000; � coe�cient 0.9-PM

74).

Comparison of OE, NORA and CIR �ndingsBoth OE and NORA were superior in comparison with CIR 40 PM

in discriminating obstruction from dilated urinary tract (Tab-le 5).

Output e�ciency correctly reclassi�ed 36 out of 41 inde-terminate or FP renograms (88%), 28 dilated and 8 obstruc-ted, whereas NORA identi�ed correctly 40 intermediate re-PM

nograms (99%), 28 dilated and 12 obstructed (Figure 1). The characteristics of kidneys in which OE or NORA were dis-40 PM

cordant with DCD are summarized in Table 6.

Figure 1. Technetium-99m-DTPA diuretic study in a patient aged 22 years, with rig-ht loin pain. US examination and IVU showed a calculus of 33mm in the lower caly-ces of the right kidney, with pelvicalyceal dilatation and reduced parenchymal thic-kness. The conventional interpreterion of the renogram could not exclude the right sided obstruction. Di�erential renal function was: left 54%, right 46%. The OE and 40

NORA of the right kidney were 88% and 0.15, respectively. According to these PM

results, the study was reported as dilated, but patent upper urinary tract on the right side. The negative for obstruction �nding was con�rmed by the extraction of a calculus by percutaneous nephrolithotomy, 2 weeks after renography.

Discussion

Although OE and NORA have been proposed as objective se-mi-quantitative parameters of kidney drainage are not ex-tensively used and the data for their calculation are not incor-porated in many software packages for the analysis of renal dynamic scintigraphy. The previously reported studies do not all use the same time for furosemide administration and for the calculation of the washout parameters [3-6]. This ma-kes it di�cult to compare the results between studies. The International Scienti�c Committee of Radionuclides in Neph-

Table 2 . OE a nd NORA i n o bstructed o r dilated u rinary t ract.

OE (%)40 NORAPM

Group A (obstructed)

N 73 73

Mean 69.71 0.38

SD 8.34 0.12

Min 37 0.20

Max 85 0.73

Group B (dilated)

N 80 80

Mean 89.04 0.11

SD 3.82 0.04

Min 79 0.04

Max 94 0.23

P value <0.001 <0.001

93Hellenic Journal of Nuclear Medicine May-August 2018• www.nuclmed.gr 142

Research Article

Left Right

NORA20 0.79 1.34

Oe (%)20 86 75

Oe (%)40 92 88

NORAPM 0.12 0.15

DRF (%) 54 46

ro-urology (ISCORN) proposed that the time of the calcula-tion be speci�ed at 20 minutes from the start point of the re-nogram and at 20 minutes after furosemide (OE , OE ) [10]. 20 40

These data were included in the non-commercial software developed under the auspices of International Atomic Ener-gy Agency. The current version of the software suggests the calculation of OE , OE , NORA and NORA as we calcula-20 40 20 PM

ted the present results [9, 11]. Our results remarkably reduced the number of indetermi-

nate �ndings of the conventional renogram.Not many studies reported the results of OE and NORA in

99mTc-DTPA renography. This radiopharmaceutical is not re-commended as the agent of choice for diuresis renography due to its low extraction fraction in comparison with tubular tracers [1]. Nevertheless, it is still used for dynamic scintigra-

phy in adults in many parts of the world, due to the availabi-lity, low price and the possibility to measure glomerular �ltra-tion rate together with imaging the kidney [12, 13]. Of cour-se, the unfavorable pharmacokinetics of DTPA increase the number of indeterminate interpretations of furosemide res-ponse in comparison with the �tubular agents�, particularly in cases of reduced renal function [14, 15]. This emphasizes the need to study the most suitable parameters for the interpre-

99mtation of Tc-DTPA renogram. Other studies on NORA �n-PM

dings in di�erent diagnostic conditions used much fewer subjects [16, 17].

99mThe initial study of the validation of OE in Tc-MAG reno-3

graphy indicated its usefulness in patients with impaired re-nal function and suggested the assessment of the validity of this parameter on a larger cohort of patients [3]. Britton K. E. (2001) emphasized that OE aids the interpretation of the equivocal furosemide response, by compensating for the e�ect of reduced renal function [18]. After the evaluation of

Research Article

993 Hellenic Journal of Nuclear Medicine May-August 2018• www.nuclmed.gr143

Table 3 . Analysis o f t he a rea u nder t he ROC c urve f or OE a nd NORA .40 PM

Predictor variables

N AUC 95% CI P Optimal cut-off value

Sensitivity (%)

Specificity (%)

OE (%)40 153 0.992 0.984-1.000 <0.001 <80 92 99

NORAPM 153 1.00 0.999-1.000 <0.001 ≥0.23 97 99

Table 4 . Comparison between OE , NORA a nd DCD. 40 PM

Obstructed Dilated Total

OE 40

(%)Obstructed 67 6 73

Dilated 1 79 80

Total 68 85 153

NORAPM

Obstructed 72 1 73

Dilated 1 79 80

Total 73 80 153

Table 5 . Comparison b etween OE, NORA a nd CIR.

OE (%)40 NORAPM CIR

Sensitivity (%) 92 97 83

Specificity (%) 99 99 64

Accuracy (%) 95 98 73

PPV (%) 98 98 68

NPV (%) 93 98 81

Table 6 . Characteristics o f k idneys w ith d iscordant r esults between

OE /DCD a nd/or NORA /DCD.40 PM

Pati-ent no.

DRF (%)

DCD OE 40

(%)NORAPM CIR

1 30* Obst-ruction

83† 0.28 Indeter-minate

2 31* -II- 82† 0.38 Obstruc-tive

3 25* -II- 85† 0.23 Indeter-minate

4 20* -II- 80† 0.33 -II-

5 30* -II- 85† 0.30 -II-

6 53 -II- 82† 0.20† Obst-ructive

7 52 Dilata-tion

79† 0.22 -II-

8 40 -II- 86 0.23† Indeter-minate

� *Patients w ith c hronic k idney d isease, Result d iscordant t o DCD

the OE in 162 patients with urological disorders, Jain et al. (2003) recommended this test as useful for resolving inde-terminate renograms [5]. However, we were unable to �nd in medical literature available, a paper about the study of OE in 99mTc-DTPA diuretic nephrogram.

On the other hand, the NORA test was proposed as a sim-pler parameter of kidney washout, which well correlated with the more sophisticated OE [7]. The calculation of NORA on postmicturition acquisition has been suggested as an ac-curate test for the assessment of kidney drainage, since it ta-kes into account the e�ect of gravity and voiding [19]. Piepsz A. (2011) in a systematic review of the management of hyd-ronephrosis in children, pointed out that the calculation of NORA on postmicturition acquisition helps interpretation of the diuretic test in cases of reduced renal function and very distensible renal pelvis [20].

In the present study, the � coe�cient between the OE re-sults and DCD fell within the range of substantial agreement [21]. The high sensitivity and speci�city of output e�ciency was better, and, due to the higher number of subjects stu-died [3].

Normalized residual activity aided the interpretation of PM

furosemide response in all except one incorrect result of OE. It helped to establish the accurate diagnosis not only in the cases of grossly dilated collective system, but also when renal function was reduced. This result a�rms the statement that the values calculated on the PM acquisition o�er the best possible assessment of kidney washout, since they take into account almost all variables of drainage [22]. Only one patient, with huge dilatation of renal pelvis due to UPJ ste-nosis, presented with NORA of 0.23, which was FP for ob-PM

struction (patient 8 in Table 6).In conclusion, the calculation of output e�ciency (OE) and

normalized residual activity test (NORA) provided accurate 99manalysis of furosemide response in the Tc-DTPA dynamic

renogram showed by far greater speci�city to diagnose ob-structive disease as compared to CIR and in particular in ca-ses of reduced renal function or grossly dilated urinary col-lecting system. Thus we may suggest that these two para-meters should be routinely used in the analysis of renogra-phy in adults. The NORA test was better than the OE test.PM 40

The authors declare that they have no con�icts of interest.

Bibliography1. Taylor AT, Blaufox DM, David C et al. SNMMI Procedure Standard/EANM

Practice Guideline for Diuretic Renal Scintigraphy in Adults with Sus-pected Upper Urinary Tract Obstruction 1.0. J Nucl Med 2017; 58. DOI: 10.2967/jnumed. 117.19 4191.

2. Taylor AT. Radionuclides in nephrourology, part 1: Pitfalls and Diag-nostic Applications. J Nucl Med 2014; 55: 786-98.

3. Chaiwatanarat T, Padhy AK, Bomanji JB et al. Validation of renal output e�ciency as an objective quantitative parameter in the evaluation of upper urinary tract obstruction. J Nucl Med 1993; 34: 845-8.

4. Piepsz A, Tondeur M, Ham H. NORA: a simple and reliable parameter for estimating renal output with or without furosemide challenge. Nucl Med Commun 2000; 21: 317-23.

5. Jain S, Cosgri� PS, Turner DT et al. Calculating the renal output e�ciency as a method for clarifying equivocal renograms in adults with suspected upper urinary tract obstruction. BJU Int 2003; 92: 485-7.

6. Piepsz A, Kuyvenhoven JD, Tondeur M et al. Normalized residual ac-tivity: usual values and robustness of the method. J Nucl Med 2002; 43: 33-8.

7. Piepsz A, Nogarède C, Tondeur M. Is normalized residual activity a good marker of renal output e�cinency? Nucl Med Commun 2011; 32: 824-8.

8. Rutland MD. A comprehensive analysis of renal DTPA studies. I. Theory and normal values. Nucl Med Commun 1985; 6: 11-20.

9. Zaknun JJ, Rajabi H, Piepsz A et al. The international atomic energy agency software package for the analysis of scintigraphic renal dyna-mic studies: a tool for clinician, teacher and researcher. Semin Nucl Med 2011; 41: 73-80.

10. Durand E, Blaufox MD, Briton KE et al. International Scienti�c Comitee of Radionuclides in Nephrourology (ISCORN) Consensus of Renal Tran-sit Time Measurements. Semin Nucl Med 2008; 38: 82-102.

11. Beatovic S, Sobic-Saranovic D, Jaksic E et al. Validation of numerical outputs of IAEA software by the analysis of duretic nephrogram in children with antenatally detected hydronephrosis. Hell J Nucl Med 20-15; Supp l1: 143.

99m12. Cheng B, Ding X, Du X et al. The half maximum time of Tc-DTPA reno-131graphy measured in healthy kidney donors, compared to I-OIH. Hell

J Nucl Med 2011; 14: 322-3.13. Zhao X, Shao Y, Wang Y et al. New normal values not related to age and

99msex, of glomerular �ltration rate by Tc-DTPA renal dynamic imaging, for the evaluation of living kidney graft donors. Hell J Nucl Med 2012; 15: 210-4.

14. Lima MC, de Lima ML, Pepe CF et al. Technetium-99m-L, L-ethylene-dicysteine is more e�ective than technetium-99m diethylenetriamine penta-acetic acid for excluding obstruction in patients with pyelocali-cinal dilation. Urology 2010; 76: 283-8.

15. Taylor AT, Schuster DM. Re: �Cost-savings analysis of renal scintigraphy, strati�ed by renal function thresholds: mercaptoacetyltriglycine ver-sus diethylene triamine penta-acetic acid.� J Am Coll Radiol 2017; 14: 146.

16. Jung HS, Chung YA, Kim EN et al. In�uence of hydration status in nor-mal subjects: fractional analysis of parameters of Tc-99m DTPA and Tc-99m MAG renography. Ann Nucl Med 2005; 19: 1-7.3

99m17. Rehling M, Møller ML, Lund JO et al. Tc-DTPA gamma-camera reno-graphy: normal values and rapid determination of single-kidney glo-merular �ltration rate. Eur J Nucl Med 1985; 11: 1-6.

18. Britton KE. Factors in�uencing the accuracy of renal output e�ciency. Nucl Med Commun 2001; 22: 1164-5.

19. Ismaili K, Piepsz A. The antenatally detected pelvi-ureteric junction ste-nosis: advances in renography and strategy of management. Pediatr Radiol 2013; 43: 428-35.

20. Piepsz A. Antenatal detection of pelviureteic junction stenosis: main Controversies. Semin Nucl Med 2011; 41: 11-9.

21. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977; 33: 159-74.

22. Gordon I, Piepsz A, Sixt R. Guidelines for standard and diuretic reno-gram in children. Eur J Nucl Med Mol Imaging 2011; 38: 1175-88.

93Hellenic Journal of Nuclear Medicine May-August 2018• www.nuclmed.gr 144

Research Article