Optimal Use of 99mTechnetium-Glucoheptonate Scintigraphy in the Detection of Pyelonephriticscarring in Children: A Preliminary Report

0022·-53,f.'7 /88/1405-1 J.7S$O:LOO/O THE JOURNAL

Vol. 140, November

UROLOGY

Printed in US.A.

Copyright© 1988 by 'fhe VV:lliams & VVilkins Cc.

OPTI1\/IAL USE OF 99 mTECHNETIUM-GLUCOHEPTONATE SCINTIGRAPHY IN THE DETECTION OF PYELONEPHRITIC SCARRING IN CHILDREN: A PRELIMINARY REPORT ELLEN SHAPIRO,* THOMAS L. SLOVIS, ALAND. PERLMUTTER

AND

LAWRENCE R. KUHNS

From the Departments of Pediatric Urology and Radiology, Children's Hospital of Michigan and Wayne State University School of Medicine, Detroit, Michigan

ABSTRACT

Renal scintigraphy represents the optimal modality for the detection of renal scars. mTechnetium-glucoheptonate is rapidly accumulated by the kidney through glomerular filtration and active transport by renal tubular cells. This permits rapid visualization of the renal parenchyma in the early phase (1 to 3-minute images) and subsequent imaging of the collecting system and ureters. About 10 to 15 per cent of the injected activity remains in the kidney, labeling the cells of the proximal convoluted tubules (late phase or 1 to 2-hour images). The late phase has been used more commonly to assess renal parenchymal damage. Early and late phase glucoheptonate scanning was performed in 42 children as part of the evaluation of recurrent febrile urinary tract infections with or without a history of vesicoureteral reflux. Inter-observer reliability to interpret glucoheptonate scans was good (early, 83 per cent agreement and late, 93 per cent agreement). The ability of glucoheptonate scanning to detect renal scarring in children with febrile urinary tract infections was equivalent with the early or late phase of the study. In 6 patients renal scarring was detected on only the early phase scan and in 7 scarring was detected only in the late phase. Although the detection rates are equivalent the over-all detection of scarring is improved by using both phases. Therefore, the early phase of the glucoheptonate scan may be a valuable adjunct to conventional glucoheptonate scan methodology used for the detection of renal scarring in children with recurrent urinary tract infections. (J. Ural., part 2, 140: 117599

1177, 1988) The accurate detection of pyelonephritic scarring is an integral part of the initial and subsequent investigations of children with vesicoureteral reflux. In the past the excretory urogram (IVP) has been used to visualize the renal contours but there is poor correlation between the extent and severity of caliceal distortion and the degree of cortical thinning. 1 One of the initial renal functions to be altered in the presence of a urinary tract infection is the concentrating mechanism of the kidney. 2 Since the scintigraphic methods used for the localization of upper urinary tract infection and the subsequent detection of renal scarring are primarily dependent upon tubular function, these images may demonstrate focal defects while the IVP will continue to appear normal. 3 • 4 For these reasons :radionuclide .:,;c111c,~u1,n1y has become the modality for the evaluation of pa:ren:chvm infection and subsequent damage. Scintigraphy with the dimercapto-succinic acid (DMSA) scan has been a valuable study for the identification of renal damage and it has been shown to be more sensitive than the IVP. 1 More recently, 99 mtechnetium (Tc)-glucoheptonate has been used as a radiotracer for the detection of renal scars. 5 • 6 Glucoheptonate is accumulated rapidly by the kidney through glomerular filtration and active transport by renal tubular cells. This permits excellent dynamic imaging with rapid visualization of the renal parenchyma in the early phase (1 to 3-minute images) and subsequent imaging of the collecting system and ureters. Since about 10 to 15 per cent of the injected activity remains in the kidney and labels the cells of the proximal convoluted tubules (later phase or 1 to 2-hour images), this phase has been used to assess renal damage. However, we have recently observed that renal scarring can be identified on the early as well as the * Current address: 400 S. Kingshighway, 5Wl4, St. Louis, Missouri 63110.

late phase of the glucoheptonate scan. This prospective study was performed to evaluate the usefulness of the early phase of the glucoheptonate scan in the detection of renal scars (see figure). MATERIALS AND METHODS

Between April 1986 and March 1987 glucoheptonate imaging was done in 42 patients 4 months to 16 years old (mean age 5.9 years) as part of the diagnostic evaluation of febrile urinary tract infections. Of the 42 children 32 (75 per cent) had a history of vesicoureteral reflux and 2 (5 per had a neurogenic bladder. There was no demonstrable etiology for the urinary tract infection in the remaining 8 patients (20 per cent). The ~ni.c,u•rno;.,c,,m:,c" scanning was not performed until at least 12 elapsed from an episode of acute pyelonephritis. A catheter was placed and left indwelling until completion of the study. 99 mTc-glucoheptonate was administered intravenously with the dose proportional to the body surface area. The maximum dose was 7.5 mCi. and the minimum dose was 2.0 mCi. A high resolution parallel hole collimator was used. Dynamic acquisition was obtained on the computer at 128 X 128 matrix, byte model as follows: ;20 images of 3-second duration followed by 58 images of 30-second duration. After data collection was terminated static-delayed 2-fold magnified views of the kidneys were obtained at 1 to 2 hours in the posterior, left posterior-oblique, right posterior-oblique and anterior views. The computer frames for the 1 to 3-minute acquisition and the 3 to 5-minute acquisition were summed and photographed. The early and late (1 to 2-hour) phases of the scans were randomly ordered and interpreted individually by 2 radiologists without access to additional clinical history or previous imaging studies. An abnormal scan was denoted by the demonstration of focal contour defect(s) in the cortical margin of the kidney

1175

1176

SHAPIRO AND ASSOCIATES

B

Cystogram in 3-month-old female infant after febrile urinary tract infection revealed right grade III/V and left grade II/V vesicoureteral reflux. IVP showed normal upper tracts without evidence of renal scarring. When she was 6 months old she had had 2 breakthrough infections while on preventive antibiotics. Glucoheptonate scintigraphy was performed before bilateral ureteral reimplantation. Posterior images demonstrate right upper pole scar (arrowhead) on early phase (A) but no scar was detected on late phase (B). TABLE

1. Detection of pyelonephritic scarring with glucoheptonate

scan in 42 cases Early Phase

Late Phase

Reader 2 Reader 1 No scar Scar

TABLE 2.

No Scar

Reader 2 Reader 1

Scar

25

4

3

10

No scar Scar

No scar Scar

Scar

25 1

2 14

Comparison of scar detection in early versus late phase glucoheptonate scan in 42 cases Reader 1

Early Phase

No Scar

Reader 2

Late Phase No Scar

Scar

25

6

4

9

Early Phase No scar Scar

Late Phase No Scar

Scar

22

6

4

10

compared to the normal reference distribution of the radiopharmaceutical. · The kappa statistic was used to measure the degree of agreement between the readers and the agreement between the early and late phases of the glucoheptonate scan. The significance of the agreement, beyond that expected by chance, was determined by using the standard error of kappa. 7 RESULTS

The readers' interpretations of the early and late phases of the glucoheptonate scan are shown in table 1. The inter-reader reliability to interpret the glucoheptonate scans was good. There was agreement in the detection of pyelonephritic scars on glucoheptonate scans in 83 per cent of the early phases reviewed (p = 0.0002) and 93 per cent of the late phases (p = 0.0001). The relative ability of the early and late phases, respectively, to detect renal scars is shown in table 2. The ability of the scan to detect renal scars in children with febrile urinary tract infection was equivalent with the early or late phase of the study (reader 1, 76 per cent agreement, p = 0.003 and reader 2, 76 per cent agreement, p = 0.002). In 6 patients renal scarring was detected on only the early phase scan. There was agreement on these findings in 2 of these 6 patients. DISCUSSION

It is important to identify reliably that subset of children with recurrent urinary tract infections and reflux who present with renal scarring at diagnosis or in whom new scarring develops during antibiotic therapy. Ransley noted that the over-

all prevalence of hypertension in all children with renal scarring was 7 per cent. 8 The prevalence of hypertension in patients with bilateral renal scarring was 18.5 per cent. Therefore, the identification of children with renal scarring is imperative for the successful long-term management of reflux and its sequelae since the detection of new scars will often change the therapeutic course from medical to surgical. The IVP has been used for the detection of renal scarring in children with recurrent febrile urinary tract infections. It serves as a baseline evaluation to demonstrate function and anatomical details of the cortex, calices and ureters. Studies performed several months after breakthrough episodes of pyelonephritis can be used to detect new renal scars. Radionuclide scintigraphy has recently become popular since no bowel preparation is necessary and no adverse reactions or toxic side effects develop. It has been shown that imaging with 99 mTc-DMSA facilitates detection of cortical scarring following pyelonephritis. Unlike the conventional IVP, cortical scanning can detect cortical defects that do not deform the renal outline or collecting system. Therefore, it is possible to detect renal scarring on scintigraphy that may not be detectable on an IVP. 3 •4 • 9 • 10 Merrick and associates found that the IVP had a sensitivity of 86 per cent and a specificity of 92 per cent for detection of renal scarring. 1 DMSA and glucoheptonate scintigraphy demonstrated a sensitivity of 96 per cent and a specificity of 98 per cent. 99 mTc-glucoheptonate has been used for the detection of renal scars. 6 Renal handling of this agent involves an early phase (glomerular filtration) and late phase (tubular labeling). The early phase is similar to the images obtained with a 99 mTcdiethylenetriaminepentaacetic acid scan. With 99 mTc-glucoheptonate the cortex, collecting system and ureters can be visualized, and the scan provides information on glomerular function. 99 mTc-DMSA does not undergo glomerular filtration, which is a limitation to its use when comparing these 2 agents. In our study 99 mTc-glucoheptonate was used to detect renal scarring in children with recurrent febrile urinary tract infections. The early and late phases were obtained to gain further functional and anatomical detail. It was observed that the early phase was accurate in detection of renal scars and it was further shown to be equivalent in its ability to demonstrate these defects. Although our study population was small, it is noteworthy that scars were detected in 6 patients on the early phase but not on the late phase scan. Therefore, the over-all detection rate of renal scarring is improved by interpreting both phases of the scan. It is possible that the delineation of these scars was improved when less tracer had accumulated in the kidney, as seen on the early phase. The readers agreed on the findings in 2 of these 6 patients. It will be important to continue to evaluate the findings on the early phase of the glucoheptonate scan. Also, in all children long-term followup with sequential scans will further confirm our initial observations on the early and late phase images. In summary, glucoheptonate scintigraphy can be used to detect accurately renal scarring. Optimal use of this isotope scanning technique should include the early phase of the study since useful information is provided about glomerular filtration, urinary tract anatomy and renal parenchymal damage. Drs. Michael Province assisted with the statistical analysis and Barry A. Siegel reviewed the manuscript. REFERENCES

1. Merrick, M. W., Uttley, W. S. and Wild, S. R.: The detection of pyelonephritic scarring in children by radioisotope imaging. Brit. J. Rad., 53: 544, 1980. 2. Winberg, J.: Renal concentration capacity during acute, nonobstructive urinary tract infections in infancy and early childhood. Acta Paed., 47: 635, 1958. 3. Gordon, I.: Indications for 99"'technetium dimercapto-succinic acid

GLUCOHEPTONATE SCINTIGRAPHY FOR DETECTION OF PYELONEPHRITIC SCARRING scan in children. J. Urol., 137: 464, 1987. 4. Traisman, E. S., Conway, J. J., Traisman, H. S., Yogev, R., Firlit, C., Shkolnik, A. and Weiss, S.: The localization of urinary tract infection with 99 mTc glucoheptonate scintigraphy. Ped. Rad., 16: 403, 1986. 5. McAfee, J. G.: Radionuclide imaging in the assessment of primary chronic pyelonephritis. Radiology, 133: 203, 1979. 6. Treves, S. T., Lebowitz, R. L., Kuruc, A., Heyman, S. and Rosen, P.: Kidneys. In: Pediatric Nuclear Medicine. Edited by S. T. Treves. New York: Springer-Verlag, chapt. 4, pp. 63-103, 1985.

1177

7. Fleiss, J. L.: Statistical Methods for Rates and Proportions. New York: John Wiley & Sons, pp. 140-153, 1973. 8. Ransley, P. G.: Discussion on complications. In: Reflux Nephropathy. Edited by C. J. Hodson and P. Kincaid-Smith. New York: Masson Publishing, p. 276, 1979. 9. Handmaker, H., Young, B. W. and Lowenstein, J. M.: Clinical experience with 99mTc-DMSA (dimercaptosuccinic acid), a new renal imaging agent. J. Nucl. Med., 16: 28, 1975. 10. Kahn, P. C.: Renal imaging with radionuclides, ultrasound, and computed tomography. Sem. Nucl. Med., 9: 43, 1979.