- Email: [email protected]
Comparison of MRI and renal cortical scintigraphy findings in childhood acute pyelonephritis: preliminary experience
European Journal of Radiology 49 (2004) 76–80
Comparison of MRI and renal cortical scintigraphy findings in childhood acute pyelonephritis: preliminary experience Arzu Kovanlikaya a , Nese Okkay a , Handan Cakmakci a,∗ , Özhan Özdo˘gan b , Berna Degirmenci b , Salih Kavukcu c a
b
Department of Radiology, Dokuz Eylül University School of Medicine, Inciralti, Izmir, Turkey Department of Nuclear Medicine, Dokuz Eylül University School of Medicine, Inciralti, Izmir, Turkey c Department of Pediatrics, Dokuz Eylül University School of Medicine, Inciralti, Izmir, Turkey Received 19 July 2002; received in revised form 28 October 2002; accepted 29 October 2002
Abstract Objective: The diagnosis of acute pyelonephritis in children remains a clinical challenge. It may cause permanent renal scar formation and results in the chronic renal failure if prompt diagnosis and treatment are delayed. The purpose of this study is to compare magnetic resonance imaging (MRI) and renal cortical scintigraphy (RCS) findings in childhood acute pyelonephritis and to determine pyelonephritic foci in the acute phase. Materials and method: Twenty children (15 females and five males) with symptoms dysuria, enuresis, costovertebral pain, fever of 37.5 ◦ C or more and/or positive urine culture were imaged by unenhanced turbo spin echo T2, spin echo T1-weighted, preand post-gadolinium inversion recovery MRI and RCS. Both imaging techniques were read independently by two radiologists and nuclear medicine specialists. Sensitivity and specificity of MRI in detecting acute pyelonephritic foci and scar lesions were calculated. Furthermore, in order to calculate the reliability of MRI over RCS in differentiating scar tissue and acute pyelonephritic foci, follow-up MRI studies were done in six patients after treatment of acute pyelonephritis. Results: Sensitivity and specificity of MRI in the detection of pyelonephritic lesions were found to be 90.9 and 88.8%, respectively. There is no statistically significant difference in lesion detection between the two diagnostic modalities (P > 0.05). Conclusion: Post-gadolinium MR images show significant correlation with RCS in the determination of renal pathology. Moreover, the ability of discriminating acute pyelonephritic foci and renal scar in early stages of disease is the superiority of MRI. © 2002 Elsevier Ireland Ltd. All rights reserved. Keywords: Pyelonephritis; Renal MRI; Renal scintigraphy; Renal diseases; Renal scar
1. Introduction Acute pyelonephritis is a critical disease, which may cause permanent renal scar formation and result in chronic renal failure if prompt diagnosis and treatment are delayed. Currently, technetium (Tc)-99m dimercaptosuccinate (DMSA) is accepted as the gold standard in detecting pyelonephritic scar. The sensitivity of DMSA has been reported to range between 80 and 100% [1,2,9,10]. However, DMSA is unable to differentiate pyelonephritic foci and permanent renal scar. The final diagnosis is usually reached 6 months later by follow-up scintigraphy. In the mean time, the patient is assumed to have acute pyelonephritis and receives intravenous antibiotic treatment. ∗
Corresponding author. E-mail address: [email protected] (H. Cakmakci).
In an animal study, magnetic resonance imaging (MRI) has been reported to be an alternative diagnostic modality in detecting pyelonephritic foci in the acute phase with a sensitivity of 91% and specificity of 93% [3]. In this study, children with suspected acute pyelonephritis were imaged by MRI after intravenous gadolinium injection and the results were compared with DMSA in a double-blinded fashion.
2. Materials and methods Twenty children (15 females and five males) with symptoms of dysuria, enuresis, costovertebral pain, fever of 37.5 ◦ C or more, and/or a positive urine culture were imaged by MRI and renal cortical scintigraphy (RCS). The ages of the patients ranged between 2 and 14 years with a mean age of 7.3 ± 3.4 years. Patients underwent the two
0720-048X/$ – see front matter © 2002 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/S0720-048X(02)00350-9
A. Kovanlikaya et al. / European Journal of Radiology 49 (2004) 76–80
imaging studies, MRI and RCS, within a week in either order. Exclusion criteria from the study included elevated levels of serum creatinine, allergy to gadopentate dimeglumine, history of haemolysis and claustrophobia. MRI was performed with a 1.0 T magnet (Siemens, Magnetom) using the body coil. After obtaining T2-weighted turbo spin echo, T1-weighted coronal and turbo inversion recovery (TIR), coronal images 0.1 mmol/kg Gd-DTPA was administered intravenously. Post-gadolinium images were performed in the coronal plane utilising the TIR sequence. In three patients, additional axial images were obtained. RCS was performed with Tc-99m DMSA. Intravenously administered Tc-99m DMSA was given at a dose of 50 Ci/kg ranging between 1.5 and 4 mCi. Approximately 4 h after the injection, patients underwent imaging with a gamma camera (GE XR/T and GE XC/T). Anterior, posterior, left posterior oblique and right posterior oblique views of the kidneys were obtained. Sedation was used in four patients for MRI. Midazolom hydrochloride at an intravenous dose of 0.1–0.2 mg/kg was used for sedation. Sedated patients were monitored continuously with a pulse oximeter for heart rate and tissue oxygenation. Blood pressure was recorded at every 15 min. Vital signs were recorded for the next 4 h following the completion of MRI. All patients tolerated sedation well with no adverse effects. Each imaging study was read independently by two radiologists and nuclear medicine specialists who were blinded to the results of the other imaging study. All images were read as hard copies, without window manipulation. The kidneys were divided into three zones (upper, mid and lower). Acute pyelonephritis was seen as increased signal areas on enhanced TIR MR images. Loss of cortical thickness and contour irregularity without increased signal was described as scar formation. The sensitivity and specificity of MRI in detecting acute pyelonephritic foci and scar lesions were calculated. The McNemar statistic was used to test the null hypothesis that there is no difference between RCS and gadolinium-enhanced TIR MRI for the detection of pyelonephritic scars. Furthermore, in order to calculate the reliability of MRI over RCS in differentiating scar tissue and acute pyelonephritic foci, follow-up MRI studies were done in six patients after treatment for acute pyelonephritis.
3. Results Both MRI and RCS demonstrated evidence of lesions in 11 patients (55%). Both showed correspondence as being positive in 10 and negative in 8 (Table 1). While six (54.5%) out of 11 patients were diagnosed as acute pyelonephritis by MRI, only two patients (18.2%) were read as most likely acute pyelonephritis by RCS. In five patients (45.5%), scar tissue was detected by MRI, only two
77
Table 1 MRI and RCS findings of the patients RCS+
RCS−
Total
MR+ MR−
10 1
1 8
11 9
Total
11
9
20
patients (18.2%) were read as most likely scar tissue by RCS (Table 2). In seven patients (63.6%), suspected acute pyelonephritis was read and a 6-month follow-up RCS was necessary in order to differentiate acute pyelonephritic foci and scar tissue. In one patient, whose RCS was negative for scar but demonstrated a dilated pyelocalyxeal system on mid–upper zone, MRI revealed a small millimetric scar on mid-zone and hydronephrosis on upper zone. In another patient whose MRI was negative RCS has shown a prominent right upper pole lesion probably due to a pyelonephritic foci. In a child with a lower pole scar on MRI, Tc-99m DMSA scintigraphy demonstrated a lower pole lesion, which was proved to be due to the infection on control scintigraphy. After complete clinical and laboratory recovery, follow-up MRI was done in six patients who were previously diagnosed to have acute pyelonephritis by MRI. The second study revealed disappearance of the lesions in five of these patients confirming that these lesions do represent acute inflammation and if treated early in the course of the disease, complete recovery is possible without scar formation (Fig. 1). In one patient, who has vesicoureteral reflux and recurrent urinary tract infections, although follow-up RCS revealed scar tissue formation, the hyperintense signal of the inflammatory foci persisted in the follow-up MRI examination. According to our data, we found the sensitivity and the specificity of MRI in the detection of pyelonephritic lesions as 90.9 and 88.8%, respectively. There is no statistically significant difference in lesion detection according to zones between the two diagnostic modalities (Table 3; P>0.05). However, in the acute phase, post-gadolinium MRI is superior to RCS in detecting the pyelonephritic foci. Table 2 Comparison of MRI and RCS (Tc-99m DMSA) findings of patients on the base of pyelonephritis–scar discrimination MRI
RCS (Tc-99m DMSA)
Total
AP
SAP
Scar
Normal
AP Scar Normal
2 – –
4 2 1
– 2 –
– 1 8
6 5 9
Total
2
7
2
9
20
AP (RCS): most likely acute pyelonephritis, SAP (RCS): suspected acute pyelonephritis, scar (RCS): most likely scar.
78
A. Kovanlikaya et al. / European Journal of Radiology 49 (2004) 76–80
Fig. 1. (A) Post-gadolinium axial IR MR images show an acute pyelonephritic focus in the right kidney (arrow). (B) Follow-up axial gadolinium-enhanced IR MR images reveal disappearance of the prior high signal intensity region in the right kidney.
Table 3 Comparison of MRI and RCS findings of the renal zones Zones
MRI Positive
Right upper Right middle Right lower Left upper Left middle Left lower Total
RCS Negative
Positive
Negative
5 3 3 5 5 5
15 17 17 15 15 15
6 2 2 5 3 4
14 18 18 15 17 16
26
94
22
98
4. Discussion The correct and early diagnosis of childhood acute pyelonephritis is important for initiation of prompt treatment. Patients with the diagnosis of acute pyelonephritis
undergo intravenous antibiotic treatment and are hospitalised [4]. Therefore, early identification of pyelonephritic foci and differentiation from scar tissue are important for the management of treatment and prognosis. Tc-99m DMSA is accepted as the gold standard in the diagnosis of acute pyelonephritis [5–9]. Previous studies demonstrated that the sensitivity of Tc-99m DMSA is between 80 and 100% [1]. Foci of pyelonephritis are seen as hypoactive photopenic regions in RCS secondary to focal ischaemia and tubular dysfunction. However, these hypoactive regions are not specific for acute pyelonephritis [10]. Renal cysts, renal abscesses, hydronephrosis, calculi and chronic scar tissue appear as hypointense areas too. Therefore, RCS can only be diagnostic if these findings are supported by clinical signs and by comparison with prior RCS findings [9,11,12]. If there is not enough clinical support and in the absence of prior RCS, patients are recommended to have a follow-up scintigraphy 6 months later. Recent studies with MRI have suggested it to be an alternative to Tc-99m DMSA in the diagnosis of acute pyelonephritis [2,3]. Experimental studies done on piglets showed that TIR MRI has a sensitivity of 91% and a specificity of 93% in the diagnosis of acute pyelonephritis [3,13,14]. The basic mechanism used for imaging of acute pyelonephritic foci on MRI is the effect of shortening of T2 created by high concentrations of paramagnetic contrast media which is eliminated via the kidneys [15,16]. Normal renal parenchyma on post-gadolinium IR MR images is seen as hypointense dark areas due to shortening of T2. In the regions of infection, perfusion is defective and gadolinium concentration is low. Thus, the affected regions remain as bright areas (Fig. 2). In TIR sequences, the TI is short which creates images similar to T2W SE images, but contrast resolution is more in TIR compared with that of SE images enabling detection of acute pyelonephritic foci. In this study, there is no statistically significant difference between the two techniques in the detection of pathology. On the other hand, four zones, which were positive on MRI, were found to be negative on RCS (Table 3). When these lesions were examined retrospectively on RCS, millimetric parenchymal defects on scarred renal tissue were overlooked. The scars of two zones, which were detected by RCS but not detected by MRI, were attributed to the low SNR of the inversion recovery sequence. The most important point in imaging pathology is the ability to differentiate scar tissue and pyelonephritic focus in the acute phase. This ability would lead to the selection of patients carrying the risk of scar formation and start their early treatment. On RCS scar lesions are seen as evident photopenic areas or parenchymal retractions. On scintigraphy, both scar and severe pyelonephritic foci are seen as photopenic areas (Fig. 2). Scar lesions in MRI are parenchyma defects in the form of depression and contour deformity without any signal change (Fig. 3). However, active inflamed renal foci are seen as hyperintense regions.
A. Kovanlikaya et al. / European Journal of Radiology 49 (2004) 76–80
79
Fig. 3. Coronal T1-weighted MR image demonstrates scar lesions as a parenchymal defect in the lower zone of the left kidney (arrow).
In conclusion, post-gadolinium inversion recovery MR images show significant correlation with RCS in the determination of pathology. On the other hand, the ability of discriminating acute pyelonephritic foci and renal scar is the superiority of MRI. MRI is a reliable diagnostic method, which can be used as a complementary method to RCS in the determination of patients with acute pyelonephritis.
References
Fig. 2. (A) Coronal gadolinium-enhanced IR MR images show acute pyelonephritic foci in right kidney lower zone (arrowhead) and left kidney upper, mid and lower zones (arrows). (B) Tc-99m DMSA demonstrates hypoactive regions in lower zone of the right kidney and upper, mid and lower zones of the left kidney (arrows). (C) Follow-up Tc-99m DMSA revealed disappearance of the hypoactive regions with persistence of contour irregularities as sequelae (arrow).
[1] Lavocat MP, Granjon D, Allard D, Gay C, Dubois F, et al. Imaging of pyelonephritis, Pediatr. Radiol. 1997;27:159–65. [2] Lonergan JG, Pennington DJ, Morrison JC, et al. Childhood pyelonephritis: comparison of gadolinium-enhanced MR imaging and renal cortical scintigraphy for diagnosis. Radiology 1998;207:377– 84. [3] Pennington DJ, Lonergan JG, Flack EC, et al. Experimental pyelonephritis in piglets: diagnosis with MR imaging. Radiology 1996;201:199–205. [4] Bjorgvinsson E, Majd M, Eggli KD. Diagnosis of acute pyelonephritis in children: comparison of sonography and Tc-99m DMSA scintigraphy. Am J Roentgenol 1991;157:539–43. [5] Jakobson B, Nolstedt L, Svensson L, Soderlundh S, Berg U. Tc-99m DMSA scan in the diagnosis of acute pyelonephritis in children: relation to clinical and radiological findings. Pediatr Nephrol 1992;6:328– 34. [6] De Sadeleer C, De Boe V, Keuppens F, Pintelon H, Verboven M, Peipsz A. Can the outcome of renal abnormalities be predicted on the basis of the initial Tc-99m DMSA scintigraphy? Eur J Nucl Med 1993;20:867. [7] Maisey MN, Britton KE, Gilday DL. Clinical Nuclear Medicine. 2nd ed., Vol. 110. Philadelphia, PA: Lippincott, 1991. p. 488. [8] Henkin RE, Boles MA, Dillehay CL. Nuclear Medicine1st ed 1996;:118–20MosbyNew York. [9] Majd M, Rushton HG. Renal cortical scintigraphy in the diagnosis of acute pyelonephritis. Semin Nucl Med 1992;22:98–111. [10] Haraoka M, Matsumuto T, Koichi T, Kubo S, Tanaka M, Kumazawa J. Suppression of renal scarring by prednisolone combined with ciprofloxacin in ascending pyelonephritis in rats. J Urol 1994;151:1078–80.
80
A. Kovanlikaya et al. / European Journal of Radiology 49 (2004) 76–80
[11] (a) Benador D, Benador N, Slosman DO, Nusslé D, Mermillod B, Girardin E. Cortical scintigraphy in the evaluation of renal parenchymal changes in children with pyelonephritis. J Pediatr 1994;124:17– 2020; (b) Bradley WG, Bydder GM. Advanced MR Imaging Techniques 1996;:250–5MosbyNew York. [12] Björgvinsson E, Majd M, Eggli KD. Diagnosis of acute pyelonephritis in children: comparison of sonography and Tc-99m DMSA scintigraphy. Am J Roentgenol 1991;157:539–43. [13] Majd M, Shalaby-Rana E, Blask A, Markle BM, Pohl H, Park J. Diagnosis of experimental pyelonephritis in piglets: comparison
of Tc-99m DMSA SPECT, spiral CT, MRI and power Doppler sonography. Radiology 1997;205(P):348. [14] Giblin JG, O’Connor KP, Fildes KR, et al. The diagnosis of acute pyelonephritis in the piglet using single photon emission computed tomography dimercaptosuccinic acid scintigraphy: a pathological correlation. J Urol 1993;150:759–62. [15] Edelman, Hesselink, Zlatkin. Clinical Magnetic Resonance Imaging. 2nd ed., Vol. 171. Philadelphia, PA: Saunders, 1996. [16] Spielman DM, Sidhu MK, Shotrliffe LM, Herfkens RJ. Development of a comprehensive MR imaging evaluation of the pediatric urinary tract. Radiology 1995;197(P):288.
Comparison of MRI and renal cortical scintigraphy findings in childhood acute pyelonephritis: preliminary experience Arzu Kovanlikaya a , Nese Okkay a , Handan Cakmakci a,∗ , Özhan Özdo˘gan b , Berna Degirmenci b , Salih Kavukcu c a
b
Department of Radiology, Dokuz Eylül University School of Medicine, Inciralti, Izmir, Turkey Department of Nuclear Medicine, Dokuz Eylül University School of Medicine, Inciralti, Izmir, Turkey c Department of Pediatrics, Dokuz Eylül University School of Medicine, Inciralti, Izmir, Turkey Received 19 July 2002; received in revised form 28 October 2002; accepted 29 October 2002
Abstract Objective: The diagnosis of acute pyelonephritis in children remains a clinical challenge. It may cause permanent renal scar formation and results in the chronic renal failure if prompt diagnosis and treatment are delayed. The purpose of this study is to compare magnetic resonance imaging (MRI) and renal cortical scintigraphy (RCS) findings in childhood acute pyelonephritis and to determine pyelonephritic foci in the acute phase. Materials and method: Twenty children (15 females and five males) with symptoms dysuria, enuresis, costovertebral pain, fever of 37.5 ◦ C or more and/or positive urine culture were imaged by unenhanced turbo spin echo T2, spin echo T1-weighted, preand post-gadolinium inversion recovery MRI and RCS. Both imaging techniques were read independently by two radiologists and nuclear medicine specialists. Sensitivity and specificity of MRI in detecting acute pyelonephritic foci and scar lesions were calculated. Furthermore, in order to calculate the reliability of MRI over RCS in differentiating scar tissue and acute pyelonephritic foci, follow-up MRI studies were done in six patients after treatment of acute pyelonephritis. Results: Sensitivity and specificity of MRI in the detection of pyelonephritic lesions were found to be 90.9 and 88.8%, respectively. There is no statistically significant difference in lesion detection between the two diagnostic modalities (P > 0.05). Conclusion: Post-gadolinium MR images show significant correlation with RCS in the determination of renal pathology. Moreover, the ability of discriminating acute pyelonephritic foci and renal scar in early stages of disease is the superiority of MRI. © 2002 Elsevier Ireland Ltd. All rights reserved. Keywords: Pyelonephritis; Renal MRI; Renal scintigraphy; Renal diseases; Renal scar
1. Introduction Acute pyelonephritis is a critical disease, which may cause permanent renal scar formation and result in chronic renal failure if prompt diagnosis and treatment are delayed. Currently, technetium (Tc)-99m dimercaptosuccinate (DMSA) is accepted as the gold standard in detecting pyelonephritic scar. The sensitivity of DMSA has been reported to range between 80 and 100% [1,2,9,10]. However, DMSA is unable to differentiate pyelonephritic foci and permanent renal scar. The final diagnosis is usually reached 6 months later by follow-up scintigraphy. In the mean time, the patient is assumed to have acute pyelonephritis and receives intravenous antibiotic treatment. ∗
Corresponding author. E-mail address: [email protected] (H. Cakmakci).
In an animal study, magnetic resonance imaging (MRI) has been reported to be an alternative diagnostic modality in detecting pyelonephritic foci in the acute phase with a sensitivity of 91% and specificity of 93% [3]. In this study, children with suspected acute pyelonephritis were imaged by MRI after intravenous gadolinium injection and the results were compared with DMSA in a double-blinded fashion.
2. Materials and methods Twenty children (15 females and five males) with symptoms of dysuria, enuresis, costovertebral pain, fever of 37.5 ◦ C or more, and/or a positive urine culture were imaged by MRI and renal cortical scintigraphy (RCS). The ages of the patients ranged between 2 and 14 years with a mean age of 7.3 ± 3.4 years. Patients underwent the two
0720-048X/$ – see front matter © 2002 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/S0720-048X(02)00350-9
A. Kovanlikaya et al. / European Journal of Radiology 49 (2004) 76–80
imaging studies, MRI and RCS, within a week in either order. Exclusion criteria from the study included elevated levels of serum creatinine, allergy to gadopentate dimeglumine, history of haemolysis and claustrophobia. MRI was performed with a 1.0 T magnet (Siemens, Magnetom) using the body coil. After obtaining T2-weighted turbo spin echo, T1-weighted coronal and turbo inversion recovery (TIR), coronal images 0.1 mmol/kg Gd-DTPA was administered intravenously. Post-gadolinium images were performed in the coronal plane utilising the TIR sequence. In three patients, additional axial images were obtained. RCS was performed with Tc-99m DMSA. Intravenously administered Tc-99m DMSA was given at a dose of 50 Ci/kg ranging between 1.5 and 4 mCi. Approximately 4 h after the injection, patients underwent imaging with a gamma camera (GE XR/T and GE XC/T). Anterior, posterior, left posterior oblique and right posterior oblique views of the kidneys were obtained. Sedation was used in four patients for MRI. Midazolom hydrochloride at an intravenous dose of 0.1–0.2 mg/kg was used for sedation. Sedated patients were monitored continuously with a pulse oximeter for heart rate and tissue oxygenation. Blood pressure was recorded at every 15 min. Vital signs were recorded for the next 4 h following the completion of MRI. All patients tolerated sedation well with no adverse effects. Each imaging study was read independently by two radiologists and nuclear medicine specialists who were blinded to the results of the other imaging study. All images were read as hard copies, without window manipulation. The kidneys were divided into three zones (upper, mid and lower). Acute pyelonephritis was seen as increased signal areas on enhanced TIR MR images. Loss of cortical thickness and contour irregularity without increased signal was described as scar formation. The sensitivity and specificity of MRI in detecting acute pyelonephritic foci and scar lesions were calculated. The McNemar statistic was used to test the null hypothesis that there is no difference between RCS and gadolinium-enhanced TIR MRI for the detection of pyelonephritic scars. Furthermore, in order to calculate the reliability of MRI over RCS in differentiating scar tissue and acute pyelonephritic foci, follow-up MRI studies were done in six patients after treatment for acute pyelonephritis.
3. Results Both MRI and RCS demonstrated evidence of lesions in 11 patients (55%). Both showed correspondence as being positive in 10 and negative in 8 (Table 1). While six (54.5%) out of 11 patients were diagnosed as acute pyelonephritis by MRI, only two patients (18.2%) were read as most likely acute pyelonephritis by RCS. In five patients (45.5%), scar tissue was detected by MRI, only two
77
Table 1 MRI and RCS findings of the patients RCS+
RCS−
Total
MR+ MR−
10 1
1 8
11 9
Total
11
9
20
patients (18.2%) were read as most likely scar tissue by RCS (Table 2). In seven patients (63.6%), suspected acute pyelonephritis was read and a 6-month follow-up RCS was necessary in order to differentiate acute pyelonephritic foci and scar tissue. In one patient, whose RCS was negative for scar but demonstrated a dilated pyelocalyxeal system on mid–upper zone, MRI revealed a small millimetric scar on mid-zone and hydronephrosis on upper zone. In another patient whose MRI was negative RCS has shown a prominent right upper pole lesion probably due to a pyelonephritic foci. In a child with a lower pole scar on MRI, Tc-99m DMSA scintigraphy demonstrated a lower pole lesion, which was proved to be due to the infection on control scintigraphy. After complete clinical and laboratory recovery, follow-up MRI was done in six patients who were previously diagnosed to have acute pyelonephritis by MRI. The second study revealed disappearance of the lesions in five of these patients confirming that these lesions do represent acute inflammation and if treated early in the course of the disease, complete recovery is possible without scar formation (Fig. 1). In one patient, who has vesicoureteral reflux and recurrent urinary tract infections, although follow-up RCS revealed scar tissue formation, the hyperintense signal of the inflammatory foci persisted in the follow-up MRI examination. According to our data, we found the sensitivity and the specificity of MRI in the detection of pyelonephritic lesions as 90.9 and 88.8%, respectively. There is no statistically significant difference in lesion detection according to zones between the two diagnostic modalities (Table 3; P>0.05). However, in the acute phase, post-gadolinium MRI is superior to RCS in detecting the pyelonephritic foci. Table 2 Comparison of MRI and RCS (Tc-99m DMSA) findings of patients on the base of pyelonephritis–scar discrimination MRI
RCS (Tc-99m DMSA)
Total
AP
SAP
Scar
Normal
AP Scar Normal
2 – –
4 2 1
– 2 –
– 1 8
6 5 9
Total
2
7
2
9
20
AP (RCS): most likely acute pyelonephritis, SAP (RCS): suspected acute pyelonephritis, scar (RCS): most likely scar.
78
A. Kovanlikaya et al. / European Journal of Radiology 49 (2004) 76–80
Fig. 1. (A) Post-gadolinium axial IR MR images show an acute pyelonephritic focus in the right kidney (arrow). (B) Follow-up axial gadolinium-enhanced IR MR images reveal disappearance of the prior high signal intensity region in the right kidney.
Table 3 Comparison of MRI and RCS findings of the renal zones Zones
MRI Positive
Right upper Right middle Right lower Left upper Left middle Left lower Total
RCS Negative
Positive
Negative
5 3 3 5 5 5
15 17 17 15 15 15
6 2 2 5 3 4
14 18 18 15 17 16
26
94
22
98
4. Discussion The correct and early diagnosis of childhood acute pyelonephritis is important for initiation of prompt treatment. Patients with the diagnosis of acute pyelonephritis
undergo intravenous antibiotic treatment and are hospitalised [4]. Therefore, early identification of pyelonephritic foci and differentiation from scar tissue are important for the management of treatment and prognosis. Tc-99m DMSA is accepted as the gold standard in the diagnosis of acute pyelonephritis [5–9]. Previous studies demonstrated that the sensitivity of Tc-99m DMSA is between 80 and 100% [1]. Foci of pyelonephritis are seen as hypoactive photopenic regions in RCS secondary to focal ischaemia and tubular dysfunction. However, these hypoactive regions are not specific for acute pyelonephritis [10]. Renal cysts, renal abscesses, hydronephrosis, calculi and chronic scar tissue appear as hypointense areas too. Therefore, RCS can only be diagnostic if these findings are supported by clinical signs and by comparison with prior RCS findings [9,11,12]. If there is not enough clinical support and in the absence of prior RCS, patients are recommended to have a follow-up scintigraphy 6 months later. Recent studies with MRI have suggested it to be an alternative to Tc-99m DMSA in the diagnosis of acute pyelonephritis [2,3]. Experimental studies done on piglets showed that TIR MRI has a sensitivity of 91% and a specificity of 93% in the diagnosis of acute pyelonephritis [3,13,14]. The basic mechanism used for imaging of acute pyelonephritic foci on MRI is the effect of shortening of T2 created by high concentrations of paramagnetic contrast media which is eliminated via the kidneys [15,16]. Normal renal parenchyma on post-gadolinium IR MR images is seen as hypointense dark areas due to shortening of T2. In the regions of infection, perfusion is defective and gadolinium concentration is low. Thus, the affected regions remain as bright areas (Fig. 2). In TIR sequences, the TI is short which creates images similar to T2W SE images, but contrast resolution is more in TIR compared with that of SE images enabling detection of acute pyelonephritic foci. In this study, there is no statistically significant difference between the two techniques in the detection of pathology. On the other hand, four zones, which were positive on MRI, were found to be negative on RCS (Table 3). When these lesions were examined retrospectively on RCS, millimetric parenchymal defects on scarred renal tissue were overlooked. The scars of two zones, which were detected by RCS but not detected by MRI, were attributed to the low SNR of the inversion recovery sequence. The most important point in imaging pathology is the ability to differentiate scar tissue and pyelonephritic focus in the acute phase. This ability would lead to the selection of patients carrying the risk of scar formation and start their early treatment. On RCS scar lesions are seen as evident photopenic areas or parenchymal retractions. On scintigraphy, both scar and severe pyelonephritic foci are seen as photopenic areas (Fig. 2). Scar lesions in MRI are parenchyma defects in the form of depression and contour deformity without any signal change (Fig. 3). However, active inflamed renal foci are seen as hyperintense regions.
A. Kovanlikaya et al. / European Journal of Radiology 49 (2004) 76–80
79
Fig. 3. Coronal T1-weighted MR image demonstrates scar lesions as a parenchymal defect in the lower zone of the left kidney (arrow).
In conclusion, post-gadolinium inversion recovery MR images show significant correlation with RCS in the determination of pathology. On the other hand, the ability of discriminating acute pyelonephritic foci and renal scar is the superiority of MRI. MRI is a reliable diagnostic method, which can be used as a complementary method to RCS in the determination of patients with acute pyelonephritis.
References
Fig. 2. (A) Coronal gadolinium-enhanced IR MR images show acute pyelonephritic foci in right kidney lower zone (arrowhead) and left kidney upper, mid and lower zones (arrows). (B) Tc-99m DMSA demonstrates hypoactive regions in lower zone of the right kidney and upper, mid and lower zones of the left kidney (arrows). (C) Follow-up Tc-99m DMSA revealed disappearance of the hypoactive regions with persistence of contour irregularities as sequelae (arrow).
[1] Lavocat MP, Granjon D, Allard D, Gay C, Dubois F, et al. Imaging of pyelonephritis, Pediatr. Radiol. 1997;27:159–65. [2] Lonergan JG, Pennington DJ, Morrison JC, et al. Childhood pyelonephritis: comparison of gadolinium-enhanced MR imaging and renal cortical scintigraphy for diagnosis. Radiology 1998;207:377– 84. [3] Pennington DJ, Lonergan JG, Flack EC, et al. Experimental pyelonephritis in piglets: diagnosis with MR imaging. Radiology 1996;201:199–205. [4] Bjorgvinsson E, Majd M, Eggli KD. Diagnosis of acute pyelonephritis in children: comparison of sonography and Tc-99m DMSA scintigraphy. Am J Roentgenol 1991;157:539–43. [5] Jakobson B, Nolstedt L, Svensson L, Soderlundh S, Berg U. Tc-99m DMSA scan in the diagnosis of acute pyelonephritis in children: relation to clinical and radiological findings. Pediatr Nephrol 1992;6:328– 34. [6] De Sadeleer C, De Boe V, Keuppens F, Pintelon H, Verboven M, Peipsz A. Can the outcome of renal abnormalities be predicted on the basis of the initial Tc-99m DMSA scintigraphy? Eur J Nucl Med 1993;20:867. [7] Maisey MN, Britton KE, Gilday DL. Clinical Nuclear Medicine. 2nd ed., Vol. 110. Philadelphia, PA: Lippincott, 1991. p. 488. [8] Henkin RE, Boles MA, Dillehay CL. Nuclear Medicine1st ed 1996;:118–20MosbyNew York. [9] Majd M, Rushton HG. Renal cortical scintigraphy in the diagnosis of acute pyelonephritis. Semin Nucl Med 1992;22:98–111. [10] Haraoka M, Matsumuto T, Koichi T, Kubo S, Tanaka M, Kumazawa J. Suppression of renal scarring by prednisolone combined with ciprofloxacin in ascending pyelonephritis in rats. J Urol 1994;151:1078–80.
80
A. Kovanlikaya et al. / European Journal of Radiology 49 (2004) 76–80
[11] (a) Benador D, Benador N, Slosman DO, Nusslé D, Mermillod B, Girardin E. Cortical scintigraphy in the evaluation of renal parenchymal changes in children with pyelonephritis. J Pediatr 1994;124:17– 2020; (b) Bradley WG, Bydder GM. Advanced MR Imaging Techniques 1996;:250–5MosbyNew York. [12] Björgvinsson E, Majd M, Eggli KD. Diagnosis of acute pyelonephritis in children: comparison of sonography and Tc-99m DMSA scintigraphy. Am J Roentgenol 1991;157:539–43. [13] Majd M, Shalaby-Rana E, Blask A, Markle BM, Pohl H, Park J. Diagnosis of experimental pyelonephritis in piglets: comparison
of Tc-99m DMSA SPECT, spiral CT, MRI and power Doppler sonography. Radiology 1997;205(P):348. [14] Giblin JG, O’Connor KP, Fildes KR, et al. The diagnosis of acute pyelonephritis in the piglet using single photon emission computed tomography dimercaptosuccinic acid scintigraphy: a pathological correlation. J Urol 1993;150:759–62. [15] Edelman, Hesselink, Zlatkin. Clinical Magnetic Resonance Imaging. 2nd ed., Vol. 171. Philadelphia, PA: Saunders, 1996. [16] Spielman DM, Sidhu MK, Shotrliffe LM, Herfkens RJ. Development of a comprehensive MR imaging evaluation of the pediatric urinary tract. Radiology 1995;197(P):288.