Long-Term Follow-Up of Reflux Nephropathy in Adults with Vesicoureteral Reflux - Radiological and Pathoanatomical Analysis

Transcription

1 Acta Radiologica ISSN: (Print) (Online) Journal homepage: Long-Term Follow-Up of Reflux Nephropathy in Adults with Vesicoureteral Reflux - Radiological and Pathoanatomical Analysis J. Köhler, H. Thysell, J. Tencer, L. Forsberg & M. Hellström To cite this article: J. Köhler, H. Thysell, J. Tencer, L. Forsberg & M. Hellström (2001) Long- Term Follow-Up of Reflux Nephropathy in Adults with Vesicoureteral Reflux - Radiological and Pathoanatomical Analysis, Acta Radiologica, 42:4, To link to this article: Published online: 09 Jul Submit your article to this journal Article views: 203 View related articles Full Terms & Conditions of access and use can be found at

2 Acta Radiologica 42 (2001) Copyright C Acta Radiologica 2001 Printed in Denmark All rights reserved ACTA RADIOLOGICA ISSN LONG-TERM FOLLOW-UP OF REFLUX NEPHROPATHY IN ADULTS WITH VESICOURETERAL REFLUX RADIOLOGICAL AND PATHOANATOMICAL ANALYSIS J. KⁿHLER 1,H.THYSELL 1,J.TENCER 1,L.FORSBERG 2 and M. HELLSTRⁿM 2 1 Department of Nephrology, University Hospital, Lund and 2 Department of Diagnostic Radiology, Sahlgrenska University Hospital, Göteborg, Sweden. Abstract Purpose: To study the long-term development of urographic renal mor- Key words: Kidney, reflux phology in adults with vesicoureteral reflux, to investigate the relationship be- nephropathy; renal dysplasia; tween renal damage and reflux grade, and to analyse the association between urography; urinary tract infection; the long-term urographic outcome and the occurrence of acute pyelonephritis vesicoureteral reflux. and reflux during follow-up. The purpose was also to try to distinguish between acquired and developmental renal damage, based on analyses of renal histologi- Correspondence: Jan Köhler, cal specimens and urographic features, and to analyse associated congenital Department of Nephrology, urogenital abnormalities and family history of reflux, reflux nephropathy, uro- University Hospital, logical malformation or death from end-stage renal disease. SE Lund, Sweden. Material and Methods: Renal damage was identified in 100 (83 women) of FAX π adults, selected because of documented reflux. Eighty-seven patients had two urographies done (median interval 14.3 years). The extent and progression Accepted for publication 26 February of renal damage were assessed and features of developmental renal damage were determined. Histological renal specimens were available in 23 patients with renal damage. Results and Conclusions: The extent of renal damage correlated positively with the severity of reflux. No renal damage developed during the follow-up in 45 previously undamaged kidneys and progression of renal damage was rare (4 of 120 previously damaged kidneys), despite persisting reflux in half of the cases and episodes of acute pyelonephritis during follow-up. Thus, repeated renal imaging is rarely justified in adults with reflux nephropathy. Histological examination showed chronic pyelonephritis in all 23 cases and co-existing renal dysplasia in 1 case. The detailed urographic analysis did not reveal support for developmental renal damage. High frequencies of associated congenital urogenital abnormalities and of a positive family history were found. Thus, congenital and/or hereditary factors cannot be discarded as background factors for the development of renal damage. Primary vesicoureteral reflux occurs with about equal frequency in boys and girls (4) and is recognized to occur in families (24). The long-term prognosis for reflux is determined mainly by the presence and severity of associated renal damage, i.e. reflux nephropathy (RN). It is the end result of a spectrum of abnormalities from prenatal renal maldevelopment (6, 26, 27, 35) to postnatally acquired postinfectious renal damage (26 28). Whether the term RN should be used when renal damage is caused by urinary tract infection (UTI) per se, in the absence of reflux, is questionable (12). Important complications of RN are hypertension and renal failure in later childhood or adulthood, as well as complications of pregnancy. RN accounts for 7% to 17% of all cases of end-stage renal disease (ESRD) in adults (7, 14, 34). Interestingly, a recent report (3) suggests that despite con- 355

3 J. KÖHLER ET AL. siderable efforts after 1960 to reduce the long-term complications of UTI and reflux by chemoprophylaxis and anti-reflux surgery, there has been no significant reduction in the frequency of ESRD from RN in Australia and New Zealand after the introduction of these preventive measures, in comparison with earlier results. The authors (3) advocate that this is plausible, if the primary pathogenic event in individuals who develop ESRD from RN, is a congenital or developmental anomaly rather than an acquired event involving acute pyelonephritis (APN). RN is primarily a disease of childhood. Therefore, the role of reflux and APN for the development of new or progressive renal damage in adults has rarely been investigated. Dimercaptosuccinic acid (DMSA)-scintigraphy and ultrasonography are currently preferred to urography to detect and monitor renal damage. Renal parenchymal damage and calyceal deformity constitute the basis for diagnosis of RN. Neither DMSA-scintigraphy nor ultrasonography (32) allow detailed visualization of the calyceal configuration, and detailed analysis of RN is therefore limited. As DMSA-scintigraphy and ultrasonography are relatively new methods, they have not yet been used in long-term studies. Thus, patients originally examined by urography are often examined with other techniques at follow-up. The limited storage time of radiographs is another limiting factor in long-term studies. The present material provided a long-term study opportunity, as urography was used for the detection as well as for the follow-up of renal damage. The aims of the study were to analyse the longterm development of urographic renal morphology in adults with vesicoureteral reflux (with or without RN), to investigate the relationship between renal damage and reflux grade, and to analyse the association between the long-term urographic outcome and the occurrence of APN and reflux during follow-up. The purpose was also to try to distinguish between acquired and developmental renal damage, based on analyses of renal histological specimens and urographic features, and to analyse associated congenital urogenital abnormalities and family history. Material and Methods Between 1967 and 1984, reflux was diagnosed in 125 patients admitted to the Departments of Nephrology or Urology at the University Hospital in Lund, Sweden. The following criteria were used for inclusion in the study: Primary reflux demonstrated at MUCG performed at the age of 16 years or later; urography of adequate quality performed in association with MUCG; no evidence of urinary tract obstruction and no anti-reflux surgery performed before entry to the study. Ten patients did not fulfill the criteria and were excluded. Thus, 115 (98 women) patients were included in the study. The patients were recruited from the primary uptake area of the hospital (58 patients) or by referral from hospitals in the southern part of Sweden (57 patients). The majority of patients (91%) were born before 1960 (19). Anti-reflux surgery was performed in adulthood in 57 patients (20). Radiology: All 115 patients had urography (inclusion urography) performed in association with the diagnosis of reflux at MUCG. The median patient age at inclusion urography was 28 years (range, years). Nineteen patients had an additional urography (pre-inclusion urography) performed earlier. At least two urographies were performed in 87 patients. The first (pre-inclusion or inclusion urography) and the last (inclusion or follow-up urography) of these were used for a longterm study of the development of renal morphology. Table 1 shows patient ages at the time of urography and the time intervals between urographies in the 87 patients who had at least two urographies performed. The median patient age at first urography in the 87 patients was 26 years (range, years) and at last urography 41 years (range, years). The median radiological follow-up time was 14.3 years (mean 14.5; range, years). Table 1 Age at urography among the 115 patients and time intervals between urographies in the 87 patients who had at least two urographies performed Type of urography(ies) Median age at first urography Median age at last urography Time interval, median years range years range years range Only inclusion, nω ª60 ª ª ª ª Pre-inclusion and inclusion, nω ª ª ª29 Inclusion and follow-up, nω ª ª ª24.2 Pre-inclusion, (inclusion) and follow-up, nω ª ª ª

4 RADIOLOGICAL FINDINGS IN ADULTS WITH VESICOURETERAL REFLUX The urographies were performed without standardization of the technique at different hospitals. Tomography was commonly used in combination with urography. One radiologist (L.F.) had read all inclusion urographies during the 1980s. The radiological reports of all 115 patients and the films from at least one urography from 93 (81%) of the 115 patients were saved. All saved urographies (including the saved first and last urographies in 65 of the 87 patients who had at least two urographies performed) were evaluated blindly and independently by two experienced radiologists (L.F. and M.H.) in 1998 and 1999, including a re-evaluation of the inclusion urographies. They had no information about the clinical history. In the case of divergent opinions on renal classification, the films were re-evaluated jointly and a consensus was reached. The original radiological reports by L.F. were used in cases with lost urographies. Urographic evaluation: A standardized protocol was used for evaluation of the renal morphology. Renal outline was classified as irregular or smooth. The renal parenchymal thickness was assessed by inspection and measurements (mm) were performed at sites of injury in each zone (cranial, middle, caudal) of the kidney for comparison with corresponding measurements at follow-up. In addition, standardized measurements were made at defined sites and the measurements were related to the height of the first three lumbar vertebrae (L1 L3 distance) (2). When the L1 3 distance exceeded the maximum range value of 115 mm (22 patients) in the nomogram, the largest reference value (115 mm) was used. To avoid errors of measurement due to different film-focus distances at repeated urographies, correction with a factor of magnification or diminuition was made using the inclusion urography as the standard with regard to the L1 3 distance. A presence of calyceal deformity was classified as follows: Mild (blunting-flattening with no or minimal dilatation), moderate (moderate dilatation and/or deformation) or severe (marked dilatation and/or deformation). The number of deformed and normal calyces in each kidney was recorded. The renal collecting system (calyces and renal pelvis) was subjectively assessed regarding its size (normal or subnormal). It was considered subnormal if it appeared smaller than the normal appearing collecting system in the contralateral kidney. We considered a kidney as normal if overall size, renal parenchyma and calyces appeared normal by inspection and by objective parenchymal measurements applying the reference nomogram of CLAËSSON et al. (2) at defined parenchymal measurement sites. A kidney was considered damaged if reduced parenchymal thickness ( ª2 SD or corresponding reduction outside the defined measurement sites at visual assessment) with calyceal deformity, or reduced parenchymal thickness ( ª2.5 SD or corresponding reduction outside the defined measurement sites at visual assessment) with normal calyces (thinning), or normal parenchymal thickness with moderate or severe calyceal deformity was found. A kidney with a smooth renal outline, uniform parenchymal reduction and normal calyces was called a small and smooth kidney, and was considered to be damaged. Classification of renal damage: In order to classify the extent of renal damage, each kidney was subdivided into three zones of approximately similar size: Cranial, middle and caudal zones. The criteria above for renal damage were applied for each renal zone. An absence of renal damage in all zones was recorded as 0. The classification of the extent of renal damage ranged from A to F according to the following criteria: Renal damage limited to one zone was subdivided into damage in a single part of the zone (A) or multiple parts of the zone (B). Renal damage in two zones was classified as C. Renal damage in all three zones, but with areas of normal parenchymal thickness, was classified as D. Renal damage in all three zones with reduced parenchymal thickness in all parts of the kidney and with deformation of at least 90% of the calyces (or all except 1) was classified as E. A small and smooth kidney was rated F. Grade of reflux: All patients were examined at least once by MUCG. The highest reflux grade obtained was chosen for grading of reflux if more than one examination had been performed. A three-grade scale was used to classify the degree of reflux: Grade I, reflux into the ureter; grade II, reflux into the renal pelvis without dilatation of the pelvis and calyces; and grade III, reflux into the renal pelvis with dilatation of the pelvis and calyces. Criteria of possible aetiology of renal damage: In the analysis of possible aetiology of renal damage (acquired or developmental) the following criteria, which support or might support developmental renal damage, were applied. 1) Pathoanatomical: histological evidence of renal dysplasia (35). 2) Urographic: less than seven calyces, and/or subnormal-sized collecting system, and/or smooth renal outline (5, 9). 3) Associated congenital urogenital abnormalities (35): renal aplasia, apparent renal malformation (maldeveloped, small non-functional pelvic-calyceal system without demonstrable 357

5 J. KÖHLER ET AL. Table 2 Renal damage (patients) at inclusion urography in 115 patients and the radiological renal development during follow-up in the 87 patients who had at least two urographies (first and last) performed Renal damage Inclusion First urography Last urography urography, (pre-inclusion (inclusion or 115 patients or inclusion), follow-up), 87 patients 87 patients None Unilateral a Bilateral a,b In single kidney a No renal damage in the residual part of the kidney after heminephrectomy in 1 additional patient. b Renal damage in single kidney after contralateral nephrectomy in 4 additional patients. Table 3 Renal damage (kidneys) at inclusion urography in 115 patients and the radiological renal development in the 87 patients who had at least two urographies (first and last) performed Extent of Inclusion First urography Last urography renal damage urography, (pre-inclusion (inclusion or 225 kidneys or inclusion), follow-up), 169 kidneys 165 kidneys A B a C a,b,c D E d F Ωno renal damage. AªFΩrenal damage, see text. a No renal damage in the residual part of the kidney after heminephrectomy in 1 additional kidney. b Progressive deformity of 1 calyx in the cranial zone in 1 kidney without change of renal damage classification. c Change of renal damage classification from C to D in 3 additional kidneys. d Four additional kidneys were removed. renal parenchyma), various duplication abnormalities of the collecting system or aplasia or malformation of genital organs. 4) Family history: relatives with reflux, RN, urological malformation or death from ESRD. Acute pyelonephritis: APN was considered to occur when the patient had bacteriuria (Ø10 5 bacteria/ml) and a body temperature of more than 38æC. Findings of loin pain and/or a serum C-reactive protein concentration of more than 30 mg/l gave additional support to the diagnosis. In the analysis of progression of radiological renal damage and APN frequency, the 2-year interval before the first urography was included, as it may take up to 2 years for renal damage to become evident at urography after APN (8). Statistical methods: The Spearman rank correlation test was performed to evaluate the correlation between the extent of renal damage and the grade of reflux, and to evaluate a possible relationship between the number of APN in patients with progressive renal damage and patients without change of the urographic morphology. The chisquared test was used to relate the grade of reflux to the proportion of damaged kidneys. The nonparametric Wilcoxon signed rank test was used to analyse the change in renal length and zonal renal parenchymal thickness between the first and the last urographies. A p-value less than 0.05% was considered significant. Results Inter-observer agreement in classification at urography: Inter-observer agreement (L.F. and M.H.) was analysed in 179 (80%) of 225 kidneys. There was complete agreement in all cases on whether there was renal damage (134 kidneys) or not (45 kidneys). There was disagreement on the extent of renal damage in 11 (9%) of 134 assessed kidneys, on the regularity of renal outline in 3 (2%) and on the size of renal collecting system in 5 (3%) of the kidneys. Renal findings at inclusion urography: Table 2 shows the radiological findings at the patient level. Renal damage was found in 87% of the patients. The reasons for single kidneys in 5 patients were renal aplasia (2 women, 1 man) and nephrectomy before inclusion (2 women). Eight additional patients (6 women) were subjected to nephrectomy and 4 women to heminephrectomy after entry to the study. Renal damage was seen in 166 (75%) of the 225 kidneys (Table 3). It involved one renal zone in 29 kidneys (18%), two renal zones in 35 kidneys (21%) and all three zones in 102 kidneys (61%). Of those with involvement of three zones, 2 (in 1 woman and 1 man) were classified as small and smooth, and 2 (in 2 women) constituted small refluxing non-functional pelvic-calyceal systems without demonstrable renal parenchyma. Renal damage was found in 405 zones: 151 cranial zones (37%), 128 middle zones (32%) and 126 caudal zones (31%). The proportion of renal damage at corresponding zones in the right and left kidneys was similar. Renal damage and grade of reflux: The proportion of kidneys with renal damage was significantly higher (p ) the higher the grade of reflux (Table 4). The extent of renal damage showed a significant, but weak, positive correlation 358

6 RADIOLOGICAL FINDINGS IN ADULTS WITH VESICOURETERAL REFLUX (p , r s Ω0.39) with the grade of reflux (Table 5). Renal radiological development (87 patients; 165 kidneys): Table 2 (patients) and Table 3 (kidneys) show the radiological development during followup in the 87 patients who had at least two urographies performed. No renal damage occurred in the 45 previously undamaged kidneys. Progressive renal damage was seen in 4 (3%) of the 120 previously damaged kidneys (in 4 patients). There was no significant change of renal length or parenchymal thickness, except in the caudal zone which showed a significant decrease (p 0.01) (Table 6). Characteristics of the 4 patients with progressive renal damage (4 kidneys): One woman developed progressive deformity of a left-sided calyx during a radiological follow-up of 19 years. She had ipsilateral reflux of grade I at diagnosis at the age of 21 years (no follow-up MUCG). She had 1 recognized APN shortly before the first urography. Another woman developed progression of a left-sided calyx deformity and scarring in the caudal zone during a radiological follow-up of 23 years. Ipsilateral reflux of grade I was documented at the time of the last urography at the age of 59 years. She had had nephro-lithotomy in the cranial part of the kidney performed 8 years after the first urography. She had no recognized APN during the study period. One man developed 5 new calyx deformities in three zones, 2 new scars in two zones and progressive scarring in one zone in the left kidney during a radiological follow-up of 3 years. He had ipsilateral reflux of grade II at the time of the last urography at the age of 36 years. Eight APNs were documented during the follow-up period. Another man (with a single kidney) developed progression of a left-sided scar in the caudal zone during a radiological follow-up of 5 years. Ipsilateral reflux of grade II was documented at inclusion at the age of 28 years, but no reflux was demonstrated 5 years later. He had no recognized APN in the follow-up period. None of these 4 patients had anti-reflux surgery performed. Change of urographic renal morphology in relation to APN between repeated urographies: APN during follow-up occurred in 47 (62%) of the 76 patients with renal damage and in 6 (55%) of the 11 patients without renal damage at inclusion urography. The number of APN in the 4 patients with progressive renal damage (median, 1 APN; range, 0 8 APN) was similar to that of the 83 patients without change of the urographic renal morphology (median, 1 APN; range, 0 31 APN) (not significant). Reflux status before and during follow-up: Reflux status was investigated before and during follow- Table 4 Grade of reflux and frequency of renal damage in 225 kidneys (115 patients) at MUCG and urography at inclusion Grade of Kidneys, nω225 Kidneys with renal damage, nω166 reflux. n n % I II III The proportion of kidneys with renal damage was significantly higher (p ) the higher the grade of reflux (Chi-squared test). Table 5 Extent of renal damage and frequency distribution of the grade of reflux in 214 kidneys (114 patients) at MUCG and urography at inclusion Extent of renal Grade of vesicoureteral reflux.. damage and 0 I II III no. of kidneys. 0, nω A, nω B, nω C, nω D, nω E, nω F, nω2 2 Total A significant, but weak, positive correlation (rsω0.39, p ) between the extent of renal damage and grade of reflux was found.. 0Ωno renal damage. AªFΩrenal damage, see text... Kidneys with complete duplication of the collecting system and ureter (incl. 1 patient with complete bilateral duplication) (nω11) were excluded. up in 165 renal units (kidney and ureter) in the 87 patients who had at least two urographies performed. Anti-reflux surgery was performed in 47 patients (77 renal units). At least one follow-up MUCG (irrespective of anti-reflux surgery) was performed in 66 patients, demonstrating reflux in 33 patients (48 renal units) and absence of reflux in 46 patients (77 renal units). All the patients with anti-reflux surgery had at least one follow-up MUCG. If the 22 patients without anti-reflux surgery and without follow-up MUCG (40 renal units) were assumed to have unchanged reflux status during follow-up, a total of 52 patients (75 renal units, 46%) had documented or assumed reflux during at least some part of the follow-up period. Pathoanatomical examinations of kidney specimens after nephrectomy or heminephrectomy or renal biopsy: All of the 10 nephrectomized kidneys had renal damage rated E, in 8 cases according to inclusion urography and in the 2 patients who were nephrectomized before inclusion according to pre- 359

7 J. KÖHLER ET AL. Table 6 Renal length and renal parenchymal thickness in the 87 patients who had at least two urographies (first and last) performed Kidneys Renal measurements, median Median difference between first a and last b p-value n First urography a Last urography b urography mm range mm range difference range Renal length ª ª166 0 ª34 to π23 ns Cranial zone thickness ª ª46 0 ª18 to π7 ns Middle zone thickness ª ª40 0 ª9 toπ9 ns Caudal zone thickness ª ª48 0 ª12 to π a Pre-inclusion or inclusion. b Inclusion or follow-up. operative urographic reports (end-stage shrunken kidneys). There were duplex collecting systems in all the 4 kidneys subjected to heminephrectomy, and the removed parts of those kidneys were extensively damaged at inclusion urography. Twelve of the 14 kidneys subjected to surgery had demonstrable reflux, 3 cases of grade I, 6 of grade II and 3 of grade III, while no information on reflux status was available in the 2 women subjected to nephrectomy before entry to the study. Pathoanatomical reports were available for all 14 removed kidney specimens. Of those, 13 were confirmative of chronic pyelonephritis and 1 was compatible with chronic pyelonephritis. One (woman) of the 13 specimens confirmative of chronic pyelonephritis also had features of renal dysplasia (primitive tubuli and cortical cysts) and 1 of the 2 ureters ended blindly. In the specimen (woman) compatible with chronic pyelonephritis, additional features suggesting Ask-Upmark kidney (14) were observed. No evidence of dysplasia (primitive ducts, metaplastic cartilage, primitive glomeruli or tubuli, fibrous walled cysts) (27) was revealed in the other specimens. In all the specimens confirmative or compatible with chronic pyelonephritis, macroscopic findings of segmental scars and microscopic findings of tubular atrophy and loss, interstitial fibrosis, periglomerular sclerosis, sclerotic glomeruli either alone or in combination with varying degrees of interstitial inflammation were found (27). Some specimens contained areas of normal parenchyma or areas of enlarged nephrons (indicating compensatory hypertrophy) interspersed with damaged parenchyma. Twelve renal biopsies (open wedge biopsy in 5 cases, open needle biopsy in 3 cases and blind - needle biopsy in 4 cases) were performed in 12 non-nephrectomized patients (8 women). Ten of the 12 biopsied kidneys had renal damage at inclusion urography classified as E (nω4),d(nω4), and C (nω2). Of the 10 biopsy specimens from damaged kidneys, 9 were compatible with chronic pyelonephritis and 1 had insufficient biopsy material for diagnosis. In 2 of the 9 cases compatible with chronic pyelonephritis, additional features of chronic glomerulonephritis (1 confirmative and 1 suspected) were found. Of the 2 biopsy specimens from kidneys without renal damage at inclusion urography, 1 was normal and 1 was compatible with chronic pyelonephritis. Evidence of renal dysplasia was not found in any renal biopsy specimen. Radiological features in damaged kidneys (nω 166) suggesting acquired or developmental renal damage: Renal outline could be assessed in 163 damaged kidneys, of which 144 (88%) (96 patients) had an irregular renal outline and 19 (12%) (17 patients) had a smooth renal outline. The size of the renal collecting system could be assessed in 160 kidneys. It was considered subnormal in 14 kidneys in 14 patients (13 women). The number of calyces could be counted in 133 damaged kidneys (median, 9 calyces; range, 5 19 calyces), of which 1 had 5 calyces and 2 had 6 calyces. The median number of calyces in the 45 non-damaged kidneys was 9 (range, 7 15 calyces). At least one of the radiological features above (smooth renal outline, subnormal-sized renal collecting system, or subnormal number of calyces), possibly in support of renal maldevelopment, was seen in 35 kidneys in 32 patients (3 men) with renal damage. Of these 35 kidneys, 19 had a smooth renal outline and 17 of these 19 kidneys had calyceal damage and a normal-sized collecting system, i.e. findings suggestive of acquired damage. The 2 remaining kidneys with a smooth renal outline had no calyceal damage (1 man) and 1 of them had a subnormal-sized collecting system (woman). These 2 kidneys were classified as small and smooth (grade F). Of the remaining 13 kidneys with a subnormal-sized collecting system, all had calyceal damage and 12 had an irregular outline, suggesting acquired damage. One of these 13 kidneys was 360

8 RADIOLOGICAL FINDINGS IN ADULTS WITH VESICOURETERAL REFLUX small, non-functioning and without demonstrable renal parenchyma. Three kidneys had 6 or less calyces, but did not have any of the other radiological features of renal maldevelopment. Associated congenital urogenital abnormalities: Among the 100 patients with renal damage, 2 (1 man) had contralateral renal aplasia, 2 women had a contralateral non-functioning kidney without detectable parenchyma, 14 women had complete or incomplete duplication of the collecting system (in 2 with a blind ending ureter), 2 had maldeveloped uterus, and 1 had aplasia of uterus, vagina and ovary (also co-existing renal aplasia). Five of the 15 patients without renal damage had unilateral complete or incomplete duplication of the collecting system. Family history: Seventeen patients (16 with renal damage, 15 women) of the 115 patients had a positive family history. It involved 23 relatives: 5 siblings, 6 parents, 4 children, 5 parent siblings, 1 grandparent, 1 grandchild and 1 sibling grandchild. Among these relatives, 5 had documented RN (2 of whom had died in ESRD), 2 had suspected RN (1 of whom had died in ESRD), 4 had reflux, 3 had a small kidney, 2 had congenital single kidney, 1 had congenital abnormality of the urinary tract, and 1 had reflux and pelvoureteral junction obstruction. The remaining 5 relatives had died from ESRD due to chronic interstitial nephritis (nω1), suspected chronic glomerulonephritis (nω1) and unknown renal disease (nω 3). History of acute pyelonephritis before inclusion urography: A history of APN could be documented in 73 (73%) of the 100 patients with renal damage and in 10 (67%) of the 15 patients without renal damage at inclusion. Fifty of the 73 patients with a history of APN and renal damage had their first known APN after four years of age. The corresponding figures among those without renal damage were 7 out of 10. Discussion The present study provided a unique opportunity to analyse the consequences of reflux and APN on the development of kidney morphology in adult patients. Previous studies on RN in adults comprise mixed populations (children and adults) for radiological follow-up (18, 22, 31) or focus on follow-up from childhood to early adult age alone (11, 16), but little information is available on the development of renal morphology in adult patients with documented reflux. All the children in the study of GOONASEKERA et al. (11) had reflux at di- agnosis, but not at follow-up because of early successful anti-reflux surgery. In other studies reflux was documented at diagnosis only in some of the patients (13, 16, 23, 25). The radiological follow-up time in the present study (mean, 14.5 years; range, 3 29 years in 87 patients) was longer than in previous studies (13, 22, 23, 25) with the exception of the studies by JONSSON et al. (18) (mean of 25 years in 15 patients) and GOONASEKERA et al. (11) (median of 15 years in 26 patients). All previous studies in adults have involved only patients with renal damage, while a group of adult patients with documented reflux but without renal damage was included in our study. The present study population represents a highly selected group of patients that presented therapeutic and/or diagnostic problems, as half of the patients were referred from outside the primary uptake area. The frequency of renal damage (87% of the patients) was higher and the extent of renal damage more advanced than in children with UTI and reflux in Sweden today (33). Possible explanations are late diagnosis of reflux (all patients were selected by reflux detected in adulthood) and the low detection rate of childhood UTI before the 1960s (91% of the patients were born before 1960). After 1960 the reported incidence of UTI in children has increased for each decade (33), accompanied by earlier detection of reflux. Simultaneously, the median age at detection of first UTI has decreased (33). The severity of reflux has been related to the risk of renal damage. The positive relationship between the occurrence or extent of renal damage and the severity of reflux has been clearly demonstrated in the present study, as well as in previous studies (1, 15, 30). However, in a recent review on primary reflux, the association between reflux and renal scarring has been questioned (10) and new insights for identifying the factors of importance that contribute to the development of scarring were asked for. It is not clear how often new renal scars in previously undamaged kidneys or progression of renal damage in adults do occur, but previous studies have indicated that they are rare (13, 18, 23, 25). On the other hand, it has been demonstrated in children (16, 18, 21, 29, 30), although it is not very common (26). During the median follow-up period of 14.3 years (range, 3 29 years) in the present study, there was no emergence of renal scars in previously undamaged kidneys, and radiological progression of renal damage was seen in only 4 kidneys (3%) in 4 patients (5%). Thus, our findings support previous studies suggesting that new renal 361

9 J. KÖHLER ET AL. scars in previously undamaged kidneys or progression of renal damage in adults are rare in the absence of obstruction (16, 18, 21, 29, 30). Whether continuing reflux causes renal damage in adults is an intriguing issue. Half of the renal units had documented or assumed persistent reflux during at least some part of the follow-up period, in no case leading to damage in previously undamaged kidneys. Interestingly, the 4 kidneys with progressive renal damage were considered to have persistent reflux, but only 2 of them had APN during follow-up. In children, the number of APN, among other factors, has been shown to be associated with new or progressive renal damage (17). However, the role of APN in the development of renal damage (16) or in the decline of renal function (18, 25) in adults has rarely been elucidated in long-term follow-up studies. In the present study, over half of the patients had APN during the observation period. There was no significant difference in the number of APN between the patients with progressive renal damage and the patients without change of the urographic morphology. A similar observation has been made by JACOBSON et al. (16). The results indicate that adults have a low risk for new or progressive renal damage, despite the presence of risk factors such as reflux and APN. When renal damage is discovered in adults, it is difficult to assess the role of earlier APN for the subsequent renal scar formation. It has been well documented that renal damage associated with UTI and reflux develops in the first years of life in the vast majority of cases (17, 26). In the present study, 27% of the patients with renal damage had no known history of APN and half of the patients with renal damage had no recognized APN before 4 years of age. This could be explained by the wellknown fact that UTI may be overlooked in early childhood or that APN in childhood was not recalled by the patient. It also raises the question of renal maldevelopment unassociated with APN (3). Pathoanatomical diagnosis in populations with RN has rarely been reported (16, 23, 27). Therefore, the results of histological renal tissue specimens from renal surgery or biopsy, available in 23 patients, were analysed. Features confirmative or compatible with chronic pyelonephritis, i.e. acquired renal damage, were found in all cases and additional features of renal maldevelopment were found in only 1 specimen and glomerulonephritis in 2 specimens. Thus, this limited histological analysis demonstrated little evidence of renal maldevelopment. We also tried to distinguish between acquired and developmental renal damage by means of uro- graphic features. Urographic diagnosis of renal hypoplasia requires a subnormal number of calyces in a small kidney (5, 9, 35). We identified only 1 kidney with 5 calyces and 2 kidneys with 6 calyces possibly representing renal hypoplasia, although this seems unlikely as they were not particularly small in size. Another feature sometimes considered in differentiating between acquired and developmental renal damage is the appearance of the renal outline. In our study, 88% of the damaged kidneys had irregular outlines suggestive of acquired disease. Of the damaged kidneys with smooth outlines, all but 2 had calyceal changes suggestive of acquired disease. Furthermore, scarring may occur without involving the renal outlines or may be confined to the anterior or posterior parts of the kidney that are difficult to visualize. Thus, a smooth renal outline is not a reliable feature for distinguishing developmental from acquired renal damage. Nevertheless, the finding of a small, smoothly outlined kidney without calyceal deformity in 2 cases indicated the possibility of renal maldevelopment or growth retardation, although other aetiologies, e.g. renovascular disease, may give a similar appearance (5, 9). While many kidneys were small in overall size, 14 kidneys had collecting systems that were subjectively assessed as subnormal-sized, possibly indicating developmental abnormality. However, this may be an unspecific finding, and all but 1 of these kidneys had calyceal deformities suggesting acquired renal damage. Although co-existing acquired extensive renal damage might have obscured some evidence of renal maldevelopment (26), neither the histological analysis nor the detailed urographic analysis revealed evidence of renal maldevelopment as a major cause of renal damage in the present study. We also analysed available data on congenital urogenital abnormalities, that may indicate the possibility of associated renal maldevelopment. Three patients with renal damage had contralateral renal aplasia and 2 had a contralateral maldeveloped, non-functioning kidney. Duplication of the renal collecting system is associated with an increased risk of congenital abnormalities in the urinary tract (35). Duplication, in 2 cases with blind ending ureters, occurred in 17% of the patients, confirming its association with reflux. Congenital malformations of the internal or external genital organs were seen in 3 women, 1 of whom had co-existing renal aplasia. A positive family history of reflux, RN, associated congenital urological abnormalities or death from ESRD was revealed in 15% of the patients. These figures are minimum figures, as the data 362

10 RADIOLOGICAL FINDINGS IN ADULTS WITH VESICOURETERAL REFLUX were collected from the hospitals files and/or interviews of the study patients, and additional cases might well have been missed in the absence of more active search for such information. It is well known that reflux is overrepresented in siblings of refluxing individuals (24), hence a hereditary component is of importance. The high frequencies of associated urogenital developmental abnormalities and a positive family history in our study suggest that congenital and/or hereditary factors may play a significant role as background factors for the development of renal damage (3, 26, 27, 35). In conclusion, no renal damage was found to develop in previously undamaged kidneys and urographic progression of renal damage was rare during follow-up, despite episodes of APN and persisting reflux. Therefore, repeated urography to investigate progression of renal damage in adult patients with non-obstructive RN is rarely justified. The high frequency and the severe extent of renal damage in this study are probably explained by late diagnosis of reflux and low detection rate of childhood UTI in the 1960s or earlier. If acquired renal damage is to be prevented or limited, APN must be suspected, diagnosed and treated rapidly in infancy and childhood (29). The high frequency of associated congenital abnormalities and of a positive family history in our study suggest that further interest should be focussed on these relationships. ACKNOWLEDGEMENTS The authors thank Professor Emeritus Ulla Bengtsson, Ph.D., who initiated this study and gave constructive criticism. We also thank Professor Elisabeth Svensson for her critical evaluation of the statistical analyses. REFERENCES 1. BISSET G. S. III, STRIFE J.L.&SCOTT DUNBAR J.: Urography and voiding cystourethrography. Findings in girls with urinary tract infection. AJR 148 (1987), CLAËSSON I., JACOBSSON B., OLSSON T. & RINGERTZ H.: Assessment of renal parenchymal thickness in normal children. Acta Radiol. 22 (1981), CRAIG J. C., IRWIG L. M., KNIGHT J. F. & PAUL ROY L.: Does treatment of vesicoureteric reflux in childhood prevent end-stage renal disease attributable to reflux nephropathy? Pediatrics 105 (2000), CREMIN B. J.: Observations on vesico-ureteric reflux and intrarenal reflux. A review and survey of material. Clin. Radiol. 30 (1979), DAVIDSON A. J., HARTMAN D. S., CHOYKE P. L. & WAGNER B. J.: Renal parenchymal disease. In: Davidson s radiology of the kidney and genitourinary tract, 3rd edn. p Edited by A. J. Davidson. W. B. Saunders Co., Philadelphia ELDER J. S.: Commentary. Importance of antenatal diagnosis of vesicoureteral reflux. J. Urol. 148 (1992), FENTON S., DESMEULES M., COPLESTON P. JR. et al.: Renal replacement therapy in Canada. A report from the Canadian Organ Replacement Register. Am. J. Kidney Dis. 25 (1995), FILLY R., FRIEDLAND G. W., GOVAN D. E. & FAIR W. R.: Development and progression of clubbing and scarring in children with recurrent urinary tract infections. Pediatr. Radiol. 113 (1974), FRIEDLAND G. W., DEVRIES P. A., NINO-MURCIA M., COHEN R. & RIFKIN M. D.: Congenital anomalies of the urinary tract. In: Clinical urography. An atlas and textbook of urological imaging, p Edited by H. M. Pollack. W. B. Saunders Co., Philadelphia GARIN E. H., CAMPOS A. & HOMSY Y.: Primary vesicoureteral reflux. Review of current concepts. Pediatr. Nephrol. 12 (1998), GOONASEKERA C. D. A., GORDON I. & DILLON M. J.: 15- year follow-up of reflux nephropathy by imaging. Clin. Nephrol. 50 (1998), GORDON I.: Vesico-ureteric reflux, urinary-tract infection, and renal damage in children. Lancet 346 (1995), GOWER P. E.: A prospective study of patients with radiological pyelonephritis, papillary necrosis and obstructive atrophy. Q. J. Med. 45 (1976), HEPTINSTALL R. H. (ALSO EDITOR): Pyelonephritis. Pathologic features. In: Pathology of the kidney, p Little, Brown & Co., Boston HODSON C. J., MALING T. M. J., MCMANAMON P. J. & LEWIS M. G.: The pathogenesis of reflux nephropathy (chronic atrophic pyelonephritis). Br. J. Radiol. 48 (1975), Suppl. 13, p JACOBSON S. H., EKLⁿF O., LINS L.-E., WIKSTAD I. & WIN- BERG J.: Long-term prognosis of post- infectious renal scarring in relation to radiological findings in childhood a 27-year follow-up. Pediatr. Nephrol. 6 (1992), JODAL U.: The natural history of bacteriuria in childhood. Infect. Dis. Clin. North Am. 1 (1987), JONSSON K.&STERNER G.: Long-term follow-up in chronic non-obstructive pyelonephritis. Scand. J. Urol. Nephrol. 19 (1985), KⁿHLER J., TENCER J., THYSELL H. & FORSBERG L.: Vesicoureteral reflux diagnosed in adulthood. Incidence of urinary tract infections, hypertension, proteinuria, back pain and renal calculi. Nephrol. Dial. Transplant. 12 (1997), KⁿHLER J., THYSELL H., TENCER J., FORSBERG L. & HELLSTRⁿM M.: Conservative treatment and antireflux surgery in adults with vesicoureteral reflux. Effect on urinary tract infections, renal function and loin pain in a longterm follow-up study. Nephrol. Dial. Transplant. 16 (2001), LENAGHAN D., WHITAKER J. G., JENSEN F. & STEPHENS F. D.: The natural history of reflux and long-term effects of reflux on the kidney. J. Urol. 115 (1976), MIHINDUKULASURIYA J. C. L., MASKELL R. & POLAK A.: A study of fifty-eight patients with renal scarring associated with urinary tract infection. Q. J. Med. 49 (1980), MOREAU J.-F., GRENIER P., GR NFELD J.-P. & BRABANT J.: Renal clubbing and scarring in adults. A retrospective study of 110 cases. Urol. Radiol. 1 (1980), NOE H. N., WYATT R. J., PEEDEN J. N. JR &RIVAS M. L.: The transmission of vesicoureteral reflux from parent to child. J. Urol. 148 (1992), OWEN J. P., RAMOS J. M., KEIR M. J. et al.: Urographic findings in adults with chronic pyelonephritis. Clin. Radiol. 36 (1985), RISDON R. A.: The smalled scarred kidney of childhood. A congenital or an acquired lesion? Pediatr. Nephrol. 1 (1987),

11 J. KÖHLER ET AL. 27. RISDON R. A., YEUNG C. K. & RANSLEY P. G.: Reflux nephropathy in children submitted to unilateral nephrectomy. A clinicopathological study. Clin. Nephrol. 40 (1993), ROBERTS J. A.: Mechanisms of renal damage in chronic pyelonephritis (reflux nephropathy). In: Current topics in pathology, p Edited by S. M. Dodd. Springer-Verlag, Berlin SMELLIE J. M., PRESCOD N. P., SHAW P. J., RISDON R. A. & BRYANT T. N.: Childhood reflux and urinary infection. A follow-up of years in 226 adults. Pediatr. Nephrol. 12 (1998), SMELLIE J. M., RANSLEY P. G., NORMAND I. C. S., PRESCOD N. & EDWARDS D.: Development of new renal scars. A collaborative study. Br. Med. J. 290 (1985), STERNER G., CHRISTENSSON A. & JONSSON K.: Clinical and radiological follow-up of chronic non-obstructive pyelonephritis. Scand. J. Urol. Nephrol. 22 (1988), STOKLAND E., HELLSTRⁿM M., JACOBSSON B., JODAL U. & SIXT R.: Evaluation of DMSA scintigraphy and urography in assessing both acute and permanent renal damage in children. Acta Radiol. 39 (1998), WENNERSTRⁿM M.: Children with symptomatic urinary tract infection in Göteborg followed for two decades. Dissertation, Göteborg University, Sweden WING A. J.: Causes of end-stage renal failure. In: Oxford textbook of clinical nephrology, p Edited by S. Cameron et al. Oxford University Press, Oxford, UK WOOLF A. S.: Clinical impact and biological basis of renal malformations. Semin. Nephrol. 15 (1995),