ORIGINAL ARTICLE 
Annals of Nuclear Medicine Vol. 8, No. 2, 147-152, 1994 

A comparative study of evaluating renal scars by 99mTC-DMSA planar and SPECT renal scans, intravenous urography, and ultrasonography 

Tzu-Chen YEN,* Wei-Perng CHEN,** Shu-Lian CHANG,* Yan-Cherng HUANG,* Chung-Pin HSIEH,*** Shih-Hwa YEH* and Ching-Yuang LIN** 

Departments of *Nuclear Medicine and **Pediatrics, Veterans General Hospital-Taipei and National Yang-Ming Medical College 
***Department of Pediatrics Taipei Municipal Women's and Children's Hospital 

The purpose of this prospective study is to compare 3 types of 99mTc-DMSA renal scan [(a) planar, (b) x-ray type film static SPECT presentation (SPECT-1 ) and (c) dynamic three-view display of SPECT slices (SPECT-2)], intravenous urography, and ultrasonography in the diagnosis of renal scars. All these studies were performed in 130 pediatric patients, with urinary tract infection (42 patients), vesicoureteral reflux (37), and unilateral or bilateral small kidney(s) (51). The number of renal scars detected was highest with the 99mTc-DMSA renal SPECT-1 scan and next came the 99mTc-DMSA renal SPECT-2 studies. There is a significant difference (p < 0.05) between the ability of planar and SPECT-1 to recognize renal defects. However, SPECT-2 may provide the best stereotactic localization and image quality of all the methods. 

Key words: Tc-99m DMSA, SPECT, IVU (Intravenous urography), ultrasonography, renal scar 

INTRODUCTION 
THE DIAGNOSIS of renal scarring is very important for the pediatric patient because it has been associated with the later development of hypertension, kidney failure, and end-stage kidney disease. 1 Although the etiology of renal scarring is still not clear, some reports show that it bears on recurrent urinary tract infection and vesicoureteral reflux. 2-4 For many years intravenous urography (IVU) was considered the best method to use in diagnosing renal scar, however, recent studies indicate that the sensitivity and specificity of the 99mTc-dimercaptosuccinic acid (DMSA) renal scan are better than IVU, especially in detecting early stage renal scar. 5-8 Although ultrasonography is noninvasive, nonionizing, relatively inexpensive and very convenient, its role in the early diagnosis of renal scar is limited. 6,9 
Various studies have compared the sensitivity of the 99mTc-DMSA renal scan, IVU, and ultrasonography in detecting renal scarring. 6,10-16 Although these studies usu-ally showed that 99mTc-DMSA was the most sensitive procedure, and emphasized their complementary nature, 

Received November 29, 1993 , revision accepted February 23 , 1994. 
For reprint contact: Dr. Tzu-Chen Yen, Department of Nuclear Medicine, Veterans General Hospital-Taipei, Taipei, Taiwan 11217, REPUBLIC OF CHINA. 

some included the following conditions or limitations which may have modified the results. These were avoided in the present work. Dillon et al. 10 studied selected subjects who had a small and contracted kidney and renal hypertension. The results of Verber et al. 11 were based on opinions from different pediatricians and hence were perhaps inconsistent. The 99mTc-DMSA renal scan used in these three studies was planar only. These limitations were avoided in the present work. Although the role of x-ray film type single photon emission computed tomography (SPECT-1) in the 99mTc-DMSA renal scan has been described, 17 few applications have been reported, although SPECT has been shown to detect children's renal cortical defects not diagnosed by planar scintigraphy. 18 It is believed that dynamic three-view display SPECT slices (SPECT-2) have not previously been used with 99mTc-DMSA renal scans, although they have proved to be a useful diagnostic method 19 in detecting liver hemangiomas. 
The aim of our research is to extend the previous studies 6,10-16 and we assess and discuss the value of 99mTc-DMSA renal scans using SPECT-1 and SPECT-2. 

MATERIALS AND METHODS 
Patients: This study was done from Aug. 1 , 1992 to July 30, 1993. One hundred and thirty patients were referred by the pediatric nephrologists of the Veterans General Hospital-Taipei and Taipei Municipal Women's and Children's Hospital with the following diagnoses: 1) urinary tract infection (UTI) (42 patients), 2) vesicoureteral reflux (37), and 3) unilateral or bilateral small kidney(s) (51) (Table 1 ). The patients with UTI were imaged within 3 days of the diagnosis being confirmed by positive urine culture. All patients underwent 99mTc-DMSA renal scans (planar, SPECT-1, and SPECT-2), IVU, and ultrasonography and all these examinations were completed within one week. Sedation was required very rarely, for the radiographers were experienced in handling pediatric patients. 

Intravenous urography: The IVU procedure was done without any prior bowel preparation or fluid restriction. All studies were performed with non-ionic contrast medium (Omnipaque300 ; Winthrop Laboratories, New York, N.Y.) at a dose of 2 ml/kg with a minimum dose of 10 ml and a maximum dose of 40 ml. Each study comprised 4 plain films with no additional tomography of the kidneys. A scar was defined as a focal area of narrowing, deformity, or both, of the renal parenchyma, usually associated with deformity of an adjacent calix. 

Ultrasonography: The renal images were obtained from an ATL Ultra Mark 8 machine with a 5 MHz transducer. At least nine images were obtained of different portions of each kidney including three parasagittal views, three transverse views and three coronal views. A scar was defined as an area of cortical thinning with or without depression of the cortical margin. 

DMSA scan method: The 99mTc-DMSA was prepared from a commercial kit (Daiichi, Japan) and the 99mTc-DMSA dose was based on the adult dose of 100 MBq and corrected for the patient's age in years by means of the following formula: 
The minimum dose was 40 MBq. 19 
Imaging was performed 3 h after injection of 99mTc-DMSA. All studies were performed on an APEX SP-4HR (Elscint, Haifa, Israel) digital gamma camera with interfaced computer. Planar images were acquired first, in the anterior and posterior projections with a total of 300000 counts per image using a low-energy, all-purpose collimator. SPECT images were obtained immediately after planar imaging and used 60 projections with 20 s per stop. Following the application of a Matz pre-filter, transverse images were reconstructed with backprojection and a ramp filter. Coronal and sagittal views with respect to the patient were reconstructed from transverse slices without any additional filtering. The SPECT-1 method comprised evaluation of a side by side static display of image slices in transverse, sagittal and coronal directions. This is also known as the "conventional static method". The dynamic display method (SPECT-2) also displays images in the same three orthogonal planes, side by side, one cross section per view, but allows for cine display and sequential step by step viewing through the slice set. The position of each slice of the group of three viewed simultaneously is identifiable by a marker on the other two. These studies were reviewed by three radiologists and nuclear physicians in random order without knowledge of the diagnosis or lesion location. The diagnosis and degree of scar severity were confirmed by at least two of the three interpreters. The degree of severity for all methods was on the scale 0-8 according to the criteria of Monsour et al., 20 which included combinations of the presence of cortical flattening, definite localized defect (one or multiple) in one or both kidneys and shrunken kidneys. 

Statistical analysis: All statistical analyses were performed by ANOVA-square analysis for categorical variables and the t test for the continuous variable. P < 0.05 level was regarded as statistically significant. 

RESULTS 
One hundred and thirty children were enrolled into this study and 14 had only one kidney, giving a total of 246 kidneys studied. The patients (Table 1 ) ranged in age from one month to 14 years (median, 5 years) and the male to female ratio was 1.2 : 1. A total of 473 renal scars were identified. 
The sensitivity and numbers of renal scars detected by the different methods are given in Table 2. The 99mTc-DMSA SPECT-1 scan identified most scars and ultrasonography fewest. Only ten scars were missed by the 99mTc-DMSA SPECT-1 , but these were detected by 99mTc-DMSA SPECT-2 and IVU. These ten defects were in ten different patients. Although the effectiveness in scar recognition of the 99smTc-DMSA SPECT-2 renal scan is equal to or slightly less than that of the 99mTc-DMSA SPECT-1 renal scan, SPECT-2 can provide the best stereotactic localization and imaging quality among these methods (Figs. 1A, 1B, and 1C). 
The results of these studies were reviewed again one week later by these radiologists and nuclear physicians. Seventeen false positives were detected, which proved to be congenital renal cysts. There were no clear correlations between the type of patient or lesion and the findings by the different scintigraphic methods. The intraobserver reliabilities for 99mTc-DMSA renal scans (planar, SPECT-1 , SPECT-2), IVU, and ultrasonography were 96%, 94%, 94%, 92%, and 88%, respectively. The interobserver reliabilities for 99mTc-DMSA renal scans (planar, SPECT-1 , SPECT-2). IVU, and ultrasonography were 88%, 90%, 92%, 95%, and 81%, respectively. 

DISCUSSION 
Although the etiology of renal scarring is still not clear, the association with UTI and vesicoureteral reflux and renal scarring is widely recognized. 2-4 Although in our studies only 42 patients (32.3%) had a UTI during the examination, when we traced the history of all the children, there were 87 (67%) who had previous UTIs. Also, 21 children (57%) had vesicoureteral reflux and 24 (47%) had unilateral or bilateral small kidneys. These results are very similar to those of Dwoskin et al. 21 and Smellie et al. 22 
For many years, IVU was considered to be the best method to use diagnosing renal scar. However, the quality of IVU images in children is not always optimal because of overlying ribs or bowel gas. Besides, such problems as allergic reactions to contrast medium in some patients and the kidney concentration power for contrast medium in children younger than two years old still have to be solved. Although tomography can help to overcome some deficiencies in image quality, the higher doses of radiation involved are unjustified for children, because of the proximity of the gonads and their increased radiosensitivity before puberty. Ultrasonography is most satisfactory both for physicians and patients because it is not invasive and uses nonionizing radiation. It can provide reliable information about the size and shape of the kidney and is accurate in diagnosing nephrocalcinosis and renal cysts. 23,24 However, our results demonstrated that its ability to detect renal scarring, especially small defects, is not satisfactory. 
Recently, the 99mTc-DMSA renal scan has replaced IVU in detecting renal scarring. 5-8 The diagnostic radio-pharmaceutical 99mTc-DMSA exhibits superior imaging characteristics to other radiopharmaceuticals for the clinical assessment of kidney morphology. The absolute uptake via glomerular filtration and peritubular capillary uptake following administration of the compound has been used to evaluate renal function. 25 Although a number of factors such as the chemical formulation, 26 binding to plasma macromolecules, 27 biochemical and physiological alterations and the presence of unlabeled DMSA 28 can alter the uptake, biodistribution and excretion of DMSA, most of these flaws can be overcome by strictly following the manufacturer's instructions during preparation of the commercial kit. Because DMSA can offer a biologic sense of kidney function, it can demonstrate functional damage, whereas IVU and ultrasonography can only exhibit anatomical damage which is the end result of functional disturbance. 
With advances in nuclear medicine instrumentation, the resolution of scintigraphic images has improved much in the last ten years. All over the world SPECT is now a very common and popular procedure. The 99mTc-DMSA renal SPECT kidney uptake ratio and its functional volume determination have been studied, 29,30 but the role of SPECT in detecting renal cortical defects has seldom been reported. 30,31 Mouratidis et al. 18 used 99mTc-DMSA renal SPECT-1 and high-resolution planar scintigraphy to compare their sensitivity and specificity in detecting renal cortical defects in children. Although they found four defects detected by SPECT-1 (out of a total of 24) not revealed by planar scintigraphy, there was no statistically significant difference between these two methods. In our studies there were 14 scars detected by SPECT-1 (out of a total of 473), which were not observed by planar scintigraphy and this shows a significant difference (P < 0.05) between 99mTc-DMSA renal SPECT-1 and planar scintigraphy. The difference between the study by Mouratidis et al. 18 and this study in results may be due to more subjects being included in the present work ( 130 vs. 41) and different collimators used in these two examinations (low-energy, general purpose collimator vs. high-resolution collimator). Theoretically, the assessment of dynamically displayed SPECT studies (SPECT-2) allows tracing of areas of decreased activity simultaneously in three views and therefore may allow better recognition of the sites of photopenia, for example differentiation of renal medulla tissues from focal lesions, even in small scars. However, we used a low-energy, general purpose one-headed SPECT system with dynamic SPECT display software supplied by the manufacturer and recognized some deeply seated lesions with the SPECT-2 method extending into the medulla even though uptake is lower there. Although the SPECT-2 method provides the best imaging quality and stereotactic localization among these examinations, we believe that with a high-resolution triple-headed SPECT system and improved dynamic SPECT display software, we can achieve better even results. However, a small cortical abnormality detected by a planar DMSA scan during an acute episode of UTI may resolve and further work is required to ascertain clinical significance of such findings. 
In conclusion, we show that 99mTc-DMSA renal SPECT-1 studies are superior to IVU, sonography, 99mTc-DMSA renal planar and SPECT-2 studies in detecting renal scarring. SPECT is a very common and popular procedure and dynamic SPECT display software is included in most currently available computer software systems, so, it is not difficult to perform 99mTc-DMSA renal SPECT-1 and SPECT-2 scintigraphy for a given patient in many nuclear medicine departments. Although much time is necessary for data acquisition and thus may limit its utilization, we still believe that the role of SPECT-2 will become increasingly important in detecting renal scarring. 

ACKNOWLEDGMENT 
This work was supported by grant No. NSC 83-0412-B-075-076 from the National Science Council, Taiwan. The authors thank Dr. Ren-Shyan Liu and Dr. Bing-Fu Shih of the Department of Nuclear Medicine for interpreting these data and Dr. Sinclair Wynchank for many helpful suggestions. This study was presented in part at the European Association of Nuclear Medicine, Congress 1993, Lausanne, Switzerland, October 10-14, 1993. 

REFERENCES 
1. Savage JM, Dillon MJ, Shah V. Barratt TM, Williams DI. Renin and blood pressure in children with renal scarring and vesicoureteral reflux. Lancet 2: 411-414, 1978. 
2. Bernstein J, Arant BS Jr. Morphological characteristics of segmental renal scarring in vesicoureteral reflux. J Urol 148: 1712-1714, 1992. 
3. Berg UB. Long-term followup of renal morphology and function in children with recurrent pyelonephritis. J Urol 148: 1715-1720, 1992. 
4. Olbing H, Claesson I, Ebel KD, et al. Renal scars and parenchymal thinning in children with vesicoureteral reflux : a 5-year report of the International Reflux Study in Children (European Branch). J Urol 148: 1653-1656, 1992. 
5. Merrick MV. Uttley WS, Wild SR. The detection of pyelonephritic scaming in children by radioisotope imaging. Br J Radiol 53: 544-546, 1980. 
6. Shanon A, Feldman W, McDonald P, et al. Evaluation of renal scars by technetium-labeled dimercaptosuccinic acid scan, intravenous urography, and ultrasonography: A comparative study. J Pediatr 120: 399-403, 1992. 
7. Rushton HG, Majd M, Jantausch B, Wiedermann BL, Belman AB. Renal scarring following reflux and nonreflux pyelonephritis in children: evaluation with 99mtechnetium-dimercaptosuccinic acid scintigraphy. J Urol 147: 1327-1332, 1992. 
8. Arant BS Jr. Medical management of mild and moderate vesicoureteral reflux : followup studies of infants and young children. A preliminary report of the Southwest Pediatric Nephrology Study Group. J Urol 148: 1683-1687, 1992. 
9. Kangarloo H, Gold RH, Fine RN, Diament MJ, Boechat MI. Urinary tract infections in infants and children evaluated by ultrasound. Radiology 154: 367-373, 1985. 
10. Dillon MJ. Gordon I, Shah V. 99mTc-DMSA scarring and segmental renal vein renin estimates in children with renal scarring. Contrib Nephrol 39: 20-27, 1984. 
11. Verber IG, Studley MR, Meller ST. 99mTc-dimercaptosuccinic acid scan as the first investigation of urinary tract infections. Arch Dis Child 63: 1320-1325, 1988. 
12. Smellie JM, Shaw PJ, Prescod NP, Bantock HM. 99mTc dimercaptosuccinic acid (DMSA) scan in patients with established radiological renal scarring. Arch Dis Child 63: 1315-1319, 1988. 1
3. Whitear P, Shaw P, Gordon I. Comparison of 99mTc dimercaptosuccinic acid scans and intravenous urography in children. Br J Radiol 63: 438-443, 1990. 
14. Ozen HA, Basar I, Erbas B, et al. DMSA renal scanning versus urography for detecting renal scars in vesicoureteral reflux. Eur Urol 17: 47-50, 1990. 
15. Melis K, Vandevivere J, Hoskens C, Vervaet A, Sand A, van Acker KJ. Involvement of the renal parenchyma in acute urinary tract infection: the contribution of 99mTc dimercaptosuccinic acid scan. Eur J Pediatr 151 : 536-539, 1992. 
16. Rosenberg AR, Rossleigh MA, Brydon MP, Bass SJ, Leighton DM, Farnsworth RH. Evaluation of acute urinary tract infection in children by dimercaptosuccinic acid scintigraphy: a prospective study. J Urol 148: 1746-1749, 1992. 
17. Williams ED. Renal single photon emission tomography: should we do it? Semin Nucl Med 22: 112-121, 1992. 
18. Mouratidis B, Ash JM, Gilday DL. Comparison of planar and SPECT 99mTc-DMSA scintigraphy for the detection of renal cortical defects in children. Nucl Med Commun 14: 82-86, 1993. 
19. Elison BS, Taylor D, van der Wall H, et al. Comparison of DMSA scintigraphy with intravenous urography for the detection of renal scarring and its correlation with vesicoureteral reflux. Br J Urol 69: 294-302, 1992. 
20. Monsour M, Azmy AF, MacKenzy R. Renal scarring secondary to vesicoureteric reflux : critical assessment and new grading. Br J Urol 60: 320-324, 1987. 
21. Dwoskin JY, Perlmutter AD. Vesicoureteral reflux in children: a computerized review. J Urol 109: 888-890, 1973. 
22. Smellie J, Edwards D, Hunter N. Vesicoureteral reflux and renal scarring. Kidney Int 8: S65-72, 1979. 
23. Alon U, Pery M, Davidi G, Berant M. Ultrasonography in radiologic evaluation of children with urinary tract infection. Pediatrics 78: 58-64, 1986. 
24. Johnson CE, DeBaz BP. Shurin PA. DeBartolomeo R. Renal ultrasound evaluation of urinary tract infections in children. Pediatrics 78: 871-878, 1986. 
25. Maneval DC, D'Argenio DZ, Wolf W. A kinetic model for 99mTc-DMSA in the rat. Eur J Nucl Med 16: 29-34, 1990. 
26. Taylor A, Lallone RL, Hagan PL. Optimal handling of dimercaptosuccinic acid for quantitative renal scanning. J Nucl Med 21: 1190-1193, 1980. 
27. Goldraich NP, Alvarenga AR, Goldraich IH, Ramos OL, Sigulem D. Renal accumulation of Tc-99m DMSA in the artificially perfused isolated rat kidney. J Urol 134: 1282-1286, 1985. 
28. Provoost AP, Van Aken M. The effect of DMSA loading on the renal handling of technetium-99m in rats. Uremia Invest 9: 147-150, 1986.
29. Murase K. Tanada S, Ishine M, Yokoyama M, Hamamoto K. Methods for measuring the renal uptake rate of 99mTc-dimercaptosuccinic acid (DMSA): a comparative study. J Nucl Med 6: 725-731, 1990. 
30. Groshar D, Embon OM, Frenkel A, Front D. Renal function and technetium-99m-dimercaptosuccinic acid uptake in single kidney: the value of in vivo SPECT quantitation. J Nucl Med 32: 766-768, 1991. 
31. Itoh K. Asano Y, Tsukamoto E, et al. Single photon emission computed tomography with 99mTc-dimercaptosuccinic acid in patients with upper urinary infection and/or vesicoureteral reflux . Ann Nucl Med 5: 29-34, 1991.