|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Pediatric Imaging |
1 From the Department of Pediatrics and Institute of Clinical Medical Sciences (Y.Y.C.), Public Health (S.T.W.), Physiology (M.J.T.), and Nuclear Medicine (B.F.L., N.T.C.), National Cheng Kung University Medical Center and College of Medicine, 138 Sheng-Li Rd, Tainan, Taiwan 704, Republic of China. Received December 13, 2000; revision requested January 16, 2001; revision received April 2; accepted April 9. Supported in part by grants from the National Science Council, Republic of China (NSC 88-2314-B-006-071 and 87-2314-B-006-001). Address correspondence to N.T.C. (e-mail: ntchiu@mail.ncku.edu.tw).
| ABSTRACT |
|---|
|
|
|---|
MATERIALS AND METHODS: Children with acute pyelonephritis underwent 99mTc DMSA renal SPECT during acute infection and 610 months later. At quantitative analysis, the volume of photopenic lesions and the ratio of radioactivity in the photopenic lesion to that in normal renal tissue were calculated. Sensitivity, specificity, and positive and negative predictive values were determined.
RESULTS: Sixty-nine acute pyelonephritis foci in 44 children were analyzed. Thirty-seven (54%) of these lesions were normal on follow-up renal scans, while 32 (46%) developed scars. Significant differences in the photopenic lesion volume were found between the two groups (P < .001). When photopenic lesion volume indicated a positive diagnosis (
4.6-cm3 lesion volume), sensitivity, specificity, positive predictive, and negative predictive values were 96.7%, 92.3%, 90.6%, and 97.3%, respectively.
CONCLUSION: Quantitative analysis of acute DMSA renal SPECT findings is valuable in predicting renal fibrosis. The volume of an acute pyelonephritis lesion is useful in predicting the development of fibrosis.
Index terms: Kidney, infection, 81.2121 Kidney, SPECT, 81.12162 Nephritis, 81.2121 Single photon emission computed tomography (SPECT), in infants and children, 81.12162
| INTRODUCTION |
|---|
|
|
|---|
The authors of several studies have reported that technetium-99m (99mTc) dimercaptosuccinic acid (DMSA) scintigraphy is a highly sensitive and specific noninvasive imaging modality for detection of renal inflammation, because it can demonstrate radiotracer uptake defects in APN foci (8,9). In one report (10), the authors said it had a specificity of 100% and a sensitivity of 87% for prospective detection of APN, when compared with histopathologic criteria. The application of single photon emission computed tomography (SPECT) together with DMSA scintigraphy further enhanced the sensitivity of DMSA in the detection of APN to 96% (11). DMSA scintigraphy is more effective and has better sensitivity for the detection of permanent renal damage (scarring) than do urography and ultrasonography (US) (12,13). It may become even more valuable if acute DMSA scintigraphy can provide prognostic information that helps determine whether APN lesions are at risk of developing into scars, thereby requiring intensive treatment.
With the assistance of SPECT, we can accurately assess the volume of an APN lesion. The issue of whether volume is associated with the process of scar formation has not, to our knowledge, been determined. Thus, the purpose of our study was to assess the value of the volume of an APN lesion derived from acute DMSA renal SPECT images as a predictor for the development of subsequent renal fibrosis.
| MATERIALS AND METHODS |
|---|
|
|
|---|
38°C; positive urine culture,
105 colony-forming units of bacteria of a given species per milliliter; leukocyturia, more than five cells per high power field); and (c) presence of focal or multifocal photon defects on a DMSA renal SPECT scintigram obtained during acute infection. The exclusion criteria were (a) history of UTI; (b) concurrent congenital urogenital abnormality or uropathy, except VUR; (c) recurrent UTI infection between the initial and subsequent scintigrams; (d) diffuse hypoactive kidney on DMSA renal SPECT scintigram; and (e) space-occupying lesion at US. The study protocol was approved by our institutional review board. Informed consent was obtained from the patients parents after a full explanation. After bacteriologic samples had been obtained from urine and blood, empiric parenteral treatment with ampicillin (Ancillina; China Chemical & Pharmaceutical, Taipei, Taiwan) and gentamicin (Yung Shin Pharmaceutical Industrial, Taichung, Taiwan) was administered for 35 days until the fever had subsided for 48 hours. This was followed by an approximately 10-day course of oral antibiotics according to the sensitivity of each drug to the bacterial infection. All patients received treatment within 7 days of fever occurrence. A low dose of trimethoprim or cephalothin was applied until voiding cystourethrography was performed 24 weeks later. Presence of VUR was graded according to the system of the International Reflux Study in Children (14). US of the urinary tract was performed at the time of admission to exclude congenital uropathy and space-occupying lesions.
99mTc DMSA Renal SPECT
Initial 99mTc DMSA renal SPECT was performed within 1 week after admission. The results were used as the reference standard for determination of APN. Those who had APN underwent repeat DMSA renal SPECT more than 6 months later. In general, an APN lesion takes at least 5 months to stabilize (15). If a renal cortical defect persisted, a renal scar was diagnosed. According to these results, we (Y.Y.C., B.F.L., N.T.C.) together classified the children into scar and nonscar groups and the lesions into sequelae and nonsequelae groups. Lesions that were persistently present on follow-up DMSA renal SPECT scans were assigned to the sequelae group. Those that resolved completely were classified into the nonsequelae group. Children with any sequelae-group lesion were assigned to the scar group, and those in whom all lesions resolved completely were assigned to the nonscar group.
For renal SPECT imaging, each subject was injected intravenously with 2.5 MBq per kilogram of body weight (minimal dose, 20 MBq) of 99mTc DMSA. Imaging was initiated approximately 23 hours later. We used a triple-headed rotating gamma camera (Multispect3; Siemens Medical Systems, Hoffman Estates, Ill) with high-resolution collimators. The SPECT data were acquired over a circular 360° rotation, 120 steps, and 35 seconds per step in a 128 x 128 x 16 matrix. Reconstruction was performed with filtered back projection by using a Butterworth filter (cut-off frequency, 0.55 of the Nyquist frequency; power factor, 7) with attenuation correction according to the Chang method (16). After reconstruction, each image was sectioned at 1-pixel (0.89-mm) intervals in the transverse, coronal, and sagittal planes. Three-dimensional images were also constructed. Any photopenic area in the renal cortex was considered abnormal. The number and location of photopenic areas, homogeneity of uptake, presence of renal volume loss, changes in renal contour, and differential renal function were recorded. Two experienced nuclear medicine physicians (B.F.L., N.T.C.) who were unaware of the patients clinical signs and symptoms visually assessed the SPECT images independently to render their diagnosis. Disagreements were resolved in discussion to reach a consensus interpretation. In a previous analysis, the agreement between these two physicians in the interpretation of DMSA renal SPECT was substantial (
= 0.78 for normal-abnormal DMSA scan dichotomy for a kidney; unpublished data, 1999).
For quantitative analysis of the acute DMSA renal SPECT images, we adopted the empiric-threshold method for volume determination (17,18). According to our previous phantom study results (unpublished data, 1996), the optimal threshold value for our gamma camera system was 41%. We analyzed the volumes and radioactivity of normal renal tissue and acute inflammatory foci from the acute DMSA renal SPECT images. We then calculated the photopenic lesion volume and radioactivity change (the ratio of radioactivity in the photopenic lesion to that in remaining normal renal tissue of the ipsilateral kidney) of the photopenic lesion to compare the sequelae and nonsequelae groups.
Statistical Analysis
Sex distribution between the scar and nonscar groups was compared with the Fisher exact test; and age distribution, with the Fisher-Freeman-Halton test (19). Mann-Whitney statistics were used to compare photopenic lesion volume and radioactivity change of the photopenic area between the sequelae and nonsequelae groups. Multiple logistic regression was used to determine the significance of age, sex, volume, and radioactivity change in predicting renal fibrosis. Because some patients had multiple foci and because these foci could be positively correlated, we used generalized estimating equations to calculate these effects (20). The radioactivity change of the photopenic area was correlated with the volume of the photopenic lesion by using the Spearman rank-order correlation. The cutoff value with maximum efficiency (ie, maximal sum of sensitivity and specificity) for a diagnostic measure was chosen for its positive predictive value for renal scarring. Unless indicated otherwise, continuous data were expressed as the mean plus or minus standard error. A P value of less than .05 was considered to indicate a significant difference.
| RESULTS |
|---|
|
|
|---|
|
|
In both the sequelae and nonsequelae groups, radioactivity change in the photopenic area was not linearly correlated with volume of the photopenic lesion (Spearman rank-order correlation coefficient = 0.5, P = .1). The means of lesion volume and radioactivity ratio between abnormal and normal renal tissue in the nonsequelae group were 2.5 cm3 ± 0.2 and 45% ± 1.2, respectively; in the sequelae group these values were 15.0 cm3 ± 2.0 and 39% ± 1.4, respectively (Fig 1). There was a significant difference in lesion volume between these two groups (P < .001). Although the nonsequelae group had a larger mean radioactivity ratio, the difference was not statistically significant (P = .33). The significance of lesion volume for prediction of renal fibrosis still held after adjustment for age, sex, and radioactivity change and was the only significant predictor in a multiple logistic regression (P < .001). A lesion volume 4.6 cm3 or larger was chosen as an indicator for subsequent development of scar; the sensitivity, specificity, positive predictive value, and negative predictive value were 96.7%, 92.3%, 90.6%, and 97.3%, respectively. Furthermore, there were six children (three boys, three girls) with 18 foci in whom eight foci developed into permanent renal scarring; the other 10 foci resolved. The lesion volumes of the eight foci that developed into permanent scar were higher than those of the latter 10 foci (12.1 cm3 ± 2.8 vs 2.4 cm3 ± 0.4, P = .001) (Fig 2).
|
|
| DISCUSSION |
|---|
|
|
|---|
After intravenous injection, 99mTc DMSA accumulates in the proximal tubular cells. Cortical uptake of this radiotracer depends on renal blood flow and proximal tubular cell membrane transport function (23). In pyelonephritic areas, the low uptake of DMSA will be confined to photopenic areas, due to locally reduced blood flow and/or a disturbed transport mechanism (13). The reduced blood flow is attributed to intravascular granulocyte aggregation leading to arteriolar or capillary occlusion, and the disturbed transport to the intratubular neutrophils that release toxic enzymes and produce superoxide (24). The combination of host immunologic response and bacterial virulence factors could further enhance tissue inflammation. With tissue ischemia and necrosis with edema change, the range of parenchymal injury would be more extensive, ultimately resulting in extended tissue volume and irreversible renal scarring (23). Therefore, an analysis of the lesion region and intensity with acute DMSA SPECT is a plausible method for predicting subsequent renal scarring. We believe that changes in radioactivity levels shown on DMSA scintigrams would be a reversible process in a limited region if these were caused solely by intravascular granulocyte aggregation, capillary occlusion, and/or local inflammation. On the contrary, if this process leads to tubular dysfunction, tubular cell death, and extensions into the interstitium, it would result in an imbalance of the repair process and a more extensive volume, both leading to irreversible renal scarring.
Renal scarring is also closely related to the normal process by which tissue recovers from damage. The failure of beneficial clearance of renal cells by means of apoptosis and/or the removal of apoptotic leukocytes before they lose membrane integrity will inevitably lead to the uncontrolled release of noxious contents. Undesirable disintegration of apoptotic cells prior to phagocytic clearance may not only exacerbate inflammation but also increase the risk of further rounds of immune-mediated tissue injury due to the release of potential intracellular autoantigens (2528). These hypotheses remain to be verified, but they all describe possible sources of the exacerbation of local tissue injury. We believe that any lesion that undergoes the above-mentioned sequence and becomes a scar will cause a more extensive photopenic focus at DMSA scintigraphy. Therefore, in addition to the severity of regional function changes, we think it is necessary to consider the involved inflammatory photopenic lesion volume when analyzing nephropathy.
Various factors have been implicated as causes of renal scarring in children with UTI. The reported rate of renal scarring in children after APN varies between 37% and 80% (2931). Duration of symptoms prior to initiation of therapy has been shown to be a factor in the development of renal fibrosis after APN (32,33). However, study results have suggested that there would be no difference in the frequency of renal scarring if treatment were to be started in the 1st week (7,29). Although a higher percentage of VUR was found in the sequelae group, renal damage could occur in the absence of reflux (34). Of 17 patients in the scar group in our study, nine (53%) had VUR and eight (47%) did not. Among the 26 patients who did not have reflux, nine (35%) had fibrosis. The evidence indicates that all children, regardless of age and sex, run a risk of renal scarring after APN. Thus, if one could successfully modulate the inflammatory response, postpyelonephritic renal scarring might be lessened. An immune modulator, oral prednisolone, was used as adjunctive therapy in an experimental piglet model (22) of severe reflux and APN, and it appeared to be effective in diminishing renal scarring. Our results emphasize the importance of inflamed photopenic lesion volume during an acute episode of infection; our findings may also provide evidence for the use of an immune modulator agent with more adequate selective criteria.
We have demonstrated that measurement of the volume of an APN focus with 99mTc DMSA renal SPECT is valuable for prediction of the occurrence of renal scarring and identification of those at high risk for scarring. Furthermore, the volume of the renal injury in the pathogenesis of renal fibrosis may be a more important and more accurate predictor than in situ inflammation.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, Y.Y.C., N.T.C.; study concepts, Y.Y.C., N.T.C.; study design, Y.Y.C., S.T.W., M.J.T., N.T.C.; literature research, Y.Y.C., N.T.C.; clinical studies, Y.Y.C., B.F.L., N.T.C.; data acquisition, Y.Y.C., B.F.L., N.T.C.; data analysis/interpretation, Y.Y.C., S.T.W., B.F.L., N.T.C.; statistical analysis, S.T.W.; manuscript preparation, Y.Y.C., S.T.W., N.T.C.; manuscript definition of intellectual content, all authors; manuscript editing, Y.Y.C., N.T.C.; manuscript revision/review and final version approval, all authors.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. G. Flohr, S. Schaller, K. Stierstorfer, H. Bruder, B. M. Ohnesorge, and U. J. Schoepf Multi-Detector Row CT Systems and Image-Reconstruction Techniques Radiology, June 1, 2005; 235(3): 756 - 773. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Imperiale, C. Olianti, S. Sestini, M. Materassi, S. Daniela, R. Ienuso, and G. La Cava 123I-Hippuran Renal Scintigraphy with Evaluation of Single-Kidney Clearance for Predicting Renal Scarring After Acute Urinary Tract Infection: Comparison with 99mTc-DMSA Scanning J. Nucl. Med., November 1, 2003; 44(11): 1755 - 1760. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |