From 1960 to the end of the 1980s, there was a determined attempt to investigate and treat what was thought to be the single leading cause of permanent kidney injury in the young – Reflux Nephropathy (RN). That term was introduced in the 1970s[1] having been previously described as chronic atrophic pyelonephritis (CPN), a pathologist’s terminology, or more loosely as coarse renal scarring, a radiologist’s description.
A large medical literature developed around this concept and from time to time it been reviewed comprehensively[2]. The scientific methods and clinical observations speak for themselves. However, the deductions and priorities that shaped clinical practice are contextual. Clinical practice moves on as technologies improve. Also, the language that we use to define clinical conditions evolves continuously. Without the wider historical context, it would be easy to dismiss past practices as short sighted, and a modern reader might share my nagging hubris. This article also reminds us that UK investigators played important roles in clarifying and progressing this field. Important early UK protagonists included John Hodson (uro-radiologist at UCH until he emigrated to North America in 1970); Philip Ransley, urologist, and Tony Risdon, pathologist, (both at GOS); Jean Smellie, paediatrician (UCH), and Richard White, paediatric nephrologist (Birmingham). The Birmingham Reflux study group (nurtured by White) which I joined as a young research fellow in 1978, also made a key contribution.
Before adding the context, I will re-visit the disease model that dominated this period, to see what gaps and criticisms were considered at the time.
Contents
VUR + IRR + UTI = CPN (Chronic Pyelonephritis or Reflux Nephropathy) [3]
Vesicoureteric reflux (VUR), the regurgitation of urine from bladder into the upper urinary tract, is not normal in humans. The lower ureter usually has a short section that lies under the mucosa before opening internally within the bladder. This effectively provides the ureter with a flap valve so that during voiding the ureter is squeezed shut between the contracting detrusor muscle of the bladder and the pressure of the bladder urine. Additionally, ureteric peristalsis, initiated high in the calyces of the kidney, ensure a one-way route for urine from kidney to bladder. For this to work the peristaltic wave needs to seal down on the ureter to propel a bolus of urine. Severe dilatation makes this ineffective and so gross VUR implies a failure both of the valvular structure of the trans-vesical lower ureter and of normal peristalsis. It is generally observed that VUR in childhood is a congenital rather than acquired abnormality.
Radiologists grade VUR according to the amount of distension in the upper urinary tract seen on a micturating cystourethrogram (MCU).
From: Tullus K. Lancet 2015, 385: 371-9)
Trivial VUR, where a brief whiff of contrast appears in the lower ureter without any dilation, is only exceptionally associated with kidney damage, and in a young child is very likely to disappear on follow up imaging.
Severe grades of VUR in which upper tract dilation is prominent tends to persist, and has three associations: intrarenal reflux, urinary infection and structural damage in the kidney. This of course is not the order in which the associations were discovered, but I will address intrarenal reflux first to explain the model.
Occasionally during cystourethrography, contrast in the upper tract of a refluxing ureter is observed to enter the tubules of the kidney and generate a nephrogram[4],[5]. This phenomenon is termed intrarenal reflux (IRR).
Cystogram in an infant showing intrarenal reflux (nephrogram seen best in left lower pole)
Rolleston, Maling and Hodson[6] (the former two based in Christchurch New Zealand) building on a previous observation by Hodson, studied the cystograms of 386 patients who had been investigated for urinary tract infections. IRR was found in 20 kidneys in 16 patients, but only in children up to 4 years of age. The sites at which IRR was seen correlated with established areas of CPN. When followed up, radiological progression, that is a further loss of kidney parenchyma, or more rarely “new” scarring, also occurred at these sites.
Development of renal scars. From: Rolleston GL, Maling TM, Hodson CJ. Arch Dis Childh 1974 49: 531-9
This suggested that IRR participated in the pathogenesis of scarring, and to advance this hypothesis, it was necessary to understand why IRR occurs in some individuals and not others, and why only some segments of the kidney are affected.
The answer lay in the anatomy of the renal papilla[7]. Human kidneys have a lobar structure[8] most obvious in the foetus and in infancy. As the kidney grows the outer shape of the kidney loses the lobulated appearance. At the poles of the kidney the lobes may fuse together during embryogenesis, but this is not seen in the central parts. An individual lobe (also described as a renunculus) typically has a single conical renal papilla that projects into its own calyx. All the collecting ducts from the lobe emerge together onto a cribriform area near the tip of the papilla. These openings are slit-like, often tangential to the surface of the papilla. By contrast where lobes fuse during development the papillae are compound. The cribriform area may then open perpendicularly to the surface. The former papillae are resistant to the retrograde flow of urine perhaps because the tangential emergence of the final collecting ducts affords a valve like protection or because the papilla is squeezed shut. The emerging collecting ducts on a compound papilla are not so protected.[9]
Other mammals such as the ox typically have entirely separate lobes all with conical papillae. Most rodents develop the whole kidney as a single lobe creating a deep conical medulla that enables astonishing urine concentration. Pigs have kidneys that are structurally and functionally very like humans with a mix of conical and compound papillae. It was therefore to this animal that two investigators turned.
Hodson found it necessary to partly obstruct the urethra to sustain VUR in piglets but then could generate extensive IRR. Ransley and Risdon took steps to avoid the accusation that the pig model might simply be one of obstruction; in their model scarring did not occur with the reflux of sterile urine. However, adding urinary infection into the model generated focal scarring very quickly, predominately at the poles of the kidney and in segments where there were compound papillae and IRR. The time available for preventing scarring by arresting infection with antibiotics was a matter of days.
While these models did not determine whether IRR alone (that is reflux of sterile urine) could cause scarring over longer periods of time , the rapid onset of segmental inflammation showed the sinister partnership of IRR with infection. Hodson thought that IRR alone could cause scarring. In his model large deposits of Tamm Horsfall protein formed in the renal interstitium which he thought was pro-inflammatory.
(In 1979 with Richard White I visited John Hodson, who can be regarded as the lead investigator in the field of children’s uro-radiology, at his home in New Haven in 1979. He had worked at UCH for many years until he emigrated to North America in 1970). Hodson’s zest for life, love of the American seaboard and enthusiasm for research were tangible. He showed us various radiographs of his mini-pig model and then some histology. The renal interstitium was stuffed with eosinophilic whorls of Tamm-Horsfall protein (THP), uromodulin. How did it get there, and in such quantity? There was already some interest in THP as a possible autoantigen in kidney disease; this before its physiological role in resisting epithelial water transport was discovered. White however was very focussed on glomerular disorders and if this encounter had been an invitation to collaborate it did not register.)
A review of the possible role of THP in reflux nephropathy including Hodson’s pig model was provided by Andriole in 1985[10].
Scandinavian researchers (led by Anders Fasth) had considered the involvement of THP in urinary infection throughout the 1970s. Other than Bill Asscher in Cardiff, the UK did not engage. Many of these players were participants in the “Conference on Reflux Nephropathy” held at the Kroc Foundation California 1982
Participants at the Kroc Foundation Conference on Reflux Nephropathy 1982
Back row from L: Philip Ransley (UK), Walter Garey USA), ?, Tony Risdon (UK), Jan Van Gool (Netherlands), Ray Lauenstein (USA), Tom Parker (USA), ?, ?
Middle row from L . Ray Kroc, Anders Fasth (Sweden), Donald Wheldon (UK), ?, Peter Avis (UK), Dick White (UK), John Duckett (USA), Mike Dillon (UK), Mark Taylor (UK), Naomi Goldraich (USA), Stephen Koff (USA), Ross Bailey (New Zealand), Bill Asscher (UK)
Front row from L: Stefan Svenson (Sweden), Bob Weiss (USA), Priscilla Kincaid-Smith (Australia), Alice Kroc, John Hodson (UDS, formerly UK), Jan Winberg (Sweden), Kate Verrier-Jones (UK), Nancy Boucot Cummings (USA)
This widely held model proposing the critical combined roles of infection and reflux in scarring therefore encouraged clinicians to identify urinary infection and treat it with urgency. It also suggested that young children with UTI should undergo full urinary tract imaging to identify damage done and consider future risks.
Paediatricians broadly adopted E H Kass’s diagnostic criteria for urinary infection in adults and devised ingenious ways of collecting urine samples in the young. Suprapubic needle aspiration of urine was also used in infants, and in these any pure growth of a pathogen was regarded as significant. While in adults the concept of “asymptomatic bacteriuria” could be based on reliable reporting, this was not so in infants. Could silent or unrecognised bacteriuria cause renal damage, and if so in whom? It was not until the late 1980s that it was appreciated that truly asymptomatic bacteriuria in older children would often clear spontaneously if left alone[11]. Screening school children for asymptomatic bacteriuria would therefore have no place. This contrasted with the risks of symptomatic infection.
Jean Smellie led a specialist clinic at UCH for children with UTI and VUR from 1955 for nearly forty years. A skilled and meticulous clinician, she influenced a generation of paediatricians to pay particular attention to bladder symptoms – delay in achieving day time continence or return to wetting being typical of urinary infection.
Randolf sequentially screened infant girls for UTI in primary care in the USA[12]. This added weight to the idea that first urinary infections belonged to infancy, an age at which signs and symptoms are often non-specific and the diagnosis is too easily delayed or missed[13]. Randolf found that first infections occurred by 9 months of age after which (until school age) all infections were recurrences. This observation was prophetic. These children had been treated with antibiotic for each episode and in this period antibiotic use in children for various infections was widespread. The adverse impact of antibiotics, especially amoxycillin, on the protective urogenital microbiome was just coming into view[14], raising the concern that some urinary infection might be iatrogenic. Jean Smellie and others quickly turned to prophylactic use of low doses of agents such as trimethoprim or nitrofurantoin to break the reinfection cycle[15].
The concept that VUR was simply a fast-track route for bacteria dispersed in urine to ascend to the vulnerable kidney was widely held. So too was the untested mathematics that bacteria in a small volume of refluxed urine would be sufficient to sustain infection against the cycles of bladder emptying that washed bacteria away. These assumptions required revision when it was shown that certain uropathogenic E coli could adhere to the uroepithelium through their P-fimbria. Bacteria could swarm their way to the upper tract with or without reflux. Notwithstanding this, the association between UTI and VUR remained firm. Clinicians in primary care and paediatrics were very strongly encouraged to identify UTI in small children, and once confirmed imaging was de rigeur.
The animal models of interstitial renal inflammation developed by Ransley and Hodson were acute not chronic. In these and in some clinical cases there is swelling of the affected segments of the kidney: “acute lobar nephronia,” quipped Hodson. Affected segments show decreased cellular uptake of DMSA acutely. This contrasts with the chronic atrophic pyelonephritis found in postmortems in adults of both sexes. The term atrophic refers to the fibrosis and shrinkage seen in the condition.
Kass had commented on the high prevalence of postmortem CPN in adults in the 1950s. It would be tempting but premature to propose that this all had its origin in childhood, and worth noting that those adults in the 1950s would have lived much of their youth before the widespread availability of antimicrobials. In children found to have advanced kidney failure in the 1970s perhaps a third had irregularly shaped shrunken kidneys. These kidneys were seldom examined pathologically.
Doubt over the pathogenesis was expressed in this period. Ask-Upmark had described “focal renal hypoplasia” back in 1929, the condition typically coming to clinical attention through hypertension and for which nephrectomy was sometimes lifesaving. The pathology shown by Habib (in Paris) in 27 cases had no evidence of chronic inflammation[16]. Had there been previous acute inflammation that had resolved to fibrosis? Renée Habib thought not.
(Mme Habib was arguably the most widely experienced, go to paediatric renal pathologist of that era. She had enthusiasm beyond measure and a need to communicate it. This made her the dread of symposium organisers and session chairpersons who knew that a 10 minute presentation would push half an hour. You could not possibly intervene nor mind the disruption. Driving her car in torrential Parisian traffic she needed to explain some fine point of pathology to her friend and admirer Richard (Dick) White in the passenger seat. Her explanation required the car to stop abruptly releasing both hands for better expression. The Parisian traffic responded as expected. The Englishman, we believe, remained calm. Nobody was hurt.)
Notwithstanding these doubts, most British and American paediatricians assumed that a small irregular kidney was CPN by default. Reluctant to apply a pathological term to a radiological finding, and given the frequency of associated VUR they were content to diagnose “reflux nephropathy” and considered the lesion to have been acquired.
It was rare to have clear observation of the development of a scar in a previously normal kidney. In Jean Smellie’s UTI clinic she, like others, noted that scarring was most often evident when the child was first investigated. This fitted the rapidity of damage seen in the experimental models but could equally represent a congenital kidney malformation. A most careful and through investigator she scoured all the paediatric nephrology centres of the UK for cases of new scar formation[18]. This Herculean task found 87 new scars in 74 children, mostly girls, all of whom had presented with UTI. Sixty seven of the 74 children had been shown to have reflux. Given the referral pattern of British children with kidney disease at the time, the denominator for this finding was vast. Reflux nephropathy may or may not be rare but catching it in the act certainly was.
(Jean Smellie was married to Colin Normand, professor of paediatrics in Southampton, who was also her research collaborator. I was Normand’s senior registrar, and encouraging me into nephrology, he promised contact with Jean Smellie who knew of a research post that had just got MRC funding. When attending a late summer party in their garden, I was apparently the only person in the world not to know that Normand and Smellie were devotedly husband and wife, and this event was indeed the promised contact. Embarrassment and hilarity were transient. Providence and Jean’s direction proved durable).
Before the emergence of high-definition ultrasound scans during the 1980s and 1990’s the main stay of urinary tract imaging was intravenous contrast urography and cystourethrography. These invasive procedures, detested as much by parents as their children, were technically challenging. They were justified beyond the identification of renal scarring or reflux. For example, identifying posterior urethral valves in infant boys or deciding whether upper urinary tract dilatation was obstructive or not had clear surgical implications. One benefit of these techniques was the permanent record of the images that could be interpreted by different clinicians with reasonable agreement. For example, the grading of VUR had a very good concordance, so too the measurement of kidney length and cortical thickness on sequential IVUs in defining scars.
Clinical decisions in the 1960s and 1970s were made on scanty information compared with today. Doctors were obliged to navigate the unknown between the scattered islands of facts, and permitted to do so as long as the conclusions they drew were intended to be solely in the interest of patients. Because of the paucity of firm knowledge, it was accepted that every clinical detail should be mined for the maximum information.
This context underlies the broad agreement at the time that extensive radiological investigations should follow a confirmed diagnosis of UTI in a child, irrespective of their yield. A position later rescinded under NICE guidelines.
Human values and judgements are non-linear and asymmetrical. Even when facts concerning risks and benefits are known, the way decisions are taken is influenced by the experiences that surround them. Losses are valued differently from benefits[19].
In the period under review, paediatric nephrology was a new and emerging specialty. Those recruited to its cause were all too familiar with the image of a child with chronic kidney failure. This included stunting, acidaemia, rickets, muscle wasting, anaemia, pain and deformity, and the psychomotor delay that seemed universal in the preschool child. These manifestations, now rarely seen in the West, were barely treatable before the 1980s, and dialysis and transplantation in small children was experimental at best. Every nephron was worth fighting for.
Where better to take the crusade than the ramparts of reflux nephropathy! The lesion was understood to be acquired or to progress because of infection. This was the antibiotic era. The congenital malformation of VUR was a risk factor. Surgery could correct it. It seemed that we had the tools for the job. The risks of treatment appeared far less than the anticipated benefit. Furthermore, all this could be applied to what was thought to be the major cause of advanced chronic kidney disease both in the young and in adult survivors.
The model suggested that both infection and severe reflux were needed to cause scarring. Therefore, it was logical to consider intervention to modify either one or both. A number of underpowered trials were undertaken with the end point being the prevention of new scars. The underpowering seems like a criticism today, but these trials were set up without knowing the rate of new scar development. It was widely deemed unethical to have a cohort of children acting as controls, and no historic prevalence or follow up surveys could come to the rescue to inform a power calculation Prophylactic antibiotic use had been shown to reduce the rate of symptomatic infection over periods of 2 years or more. However, adherence to medication was not widely measured, and close medical follow up in an enthusiastic clinic probably had an unseen benefit. Surgical correction of reflux, reimplanting the lower ureter through the bladder with a good submucosal tunnel, sometimes with tapering of a very dilated ureter to better the chances of peristaltic function, became a successful routine in many centres. Obstructive complications were rare.
Before multidisciplinary clinical teams became de rigeur, Richard White (nephrologist), J J (Sean) Corkery (paediatric surgeon) and Roy Astley (radiologist) met weekly in Birmingham to review urinary radiographs and discuss clinical approaches. The group expanded over time to include Kishore Shah (radiologist), Peter Gornall (surgeon), and later myself as research fellow.
The Birmingham Children’s Hospital Reflux Study Group in the 1980s
Back row from left: JJ (Sean) Corkery (surgeon), Kishore Shah (radiologist), Richard White (nephrologist), Mark Taylor (nephrologist)
Front from: left Michael Winterborn (nephrologist), Peter Gronall (surgeon) Roy Astley (radiologist), Helen Alton (radiologist)
This group developed a protocol for a study to investigate the benefits of anti-reflux surgery in children with UTI and severe VUR receiving antibiotics.
This is a landmark study, one of the first well designed randomised trials in paediatric nephrology, and a study which changed practice. The justification for the trial was that in cases of reflux there was genuine equipoise in the opinions of the clinical team. It was not that surgeons promoted surgery and physicians non-surgical approaches. That equipoise was crucial in convincing the hospital Medical Ethics Committee that patients could be allocated randomly to either operative correction with antibiotic cover, or antibiotic alone without parental consent. Parental preference might have skewed allocation and invalidate the trial. This way of thinking would become anathema before the end of the last century.
(One mother had a child doing well in the non-operative group and then its sibling was found to have reflux too. The second sibling drew operative management and the mother wanted to know why this time it was different. The discussion with her was honest, and notably she expressed acceptance of the trial and the randomisation. The impact of medical paternalism and covert authority can only be guessed at. Past acceptance cannot mitigate against a later sense of injustice that could arise at any time).
The trial began in 1978 the findings in 96 children were published in 1983[20], and the final report on 161 children in 1987[21]. Children with severe (at least grade 3) vesicoureteric reflux were allocated randomly to either operative or non-operative treatment and followed up. Altogether 161 children were observed for two years, of whom 104 were followed up for five years. Reflux was abolished in 98% of ureters reimplanted (with no obstructive complications), but more than half of the patients treated non-operatively continued to show severe reflux at five years. There were no significant differences between treatment groups in the incidence of breakthrough urinary infection, renal function and concentrating ability, renal growth, progression of existing renal scars, or new scar formation. So surgery did not improve outcomes.
Inevitably the findings were not uniformly welcomed among paediatirc urologists whose practice was heavily committed to anti-reflux surgery.
(When the study was approaching publication, the suave and handsome Philip Ransley (the Sean Connery of paediatric urologists) arranged for me to present the findings at an “international” (Italian actually) paediatric urology conference, which seemed to focus on the prestige of a local professor. A first sign of that was his extensive entourage of beautiful and unpaid juniors; the second sign was affluence. On the second day with Ransley in attendance, the great man operated, and the live stream images appeared on a huge screen with simultaneous translation in the auditorium. The beautiful made extensive and unpaid notes. Later, presenting the Birmingham Reflux Study I had to tell them we could not detect any superiority of surgery over antibiotics. There was a dreadful silence and gaze was avoided: apostacy. Philip near the front caught my eye and gave the briefest smile: absolution.)
Subsequent trials reprised these findings, and could only show equivalence between surgery and antibiotic prophylaxis. It did not look as though VUR independently promoted UTI as some for many years had thought.
I was also investigating other aspects of the pathophysiology of VUR, work which advanced further my exposure to international medical conferencing. I had an MRC grant to determine the role of bladder pressure and dysfunction in VUR (urodynamic studies in children at that time were highly experimental). I met Stephen Koff later to become a distinguished urologist, who it turned out was similarly engaged on the other side of the Atlantic. We had common findings that there were frequent detrusor contractions capable of generating very high bladder pressures, which in severe VUR would likely be transmitted back to the kidneys. This impacts on the two possible interpretations of the pig model, where one is more akin to obstruction, and one is not. Koff and I shared the uncomfortable thought that our common findings of very high pressures in the urinary tracts of children with reflux would be unpublishable as there could never be data from controls. In the event both were successful, Koff more so. (Taylor CM Corkery JJ and White RHR. Micturition symptoms and unstable bladder activity in girls with primary vesicoureteric reflux. Br J Urology 1982 54: 449-8.
(We were both invited to a conference on all things reflux at the Kroc Foundation at its conference centre and ranch outside Santa Barbara, California Figure 7. (Ray Kroc had used his share of the family fortune, made from McDonald’s hamburger chain, to invest heavily in support of biomedical research). Chauffeured in a stretch limo, the drive up into the Californian countryside was impressive. Then the automatic gates opened on the private road that crested a rise to reveal the ranch, the state-of-the-art conference centre and library, massive satellite dish and tennis courts. The reception hall in the ranch was panelled, a two-inch-thick white carpet graced the floor, the far wall was a given over to a bar with drinks of every complexion and an ice maker. Above all this presided the stuffed head and horns of a huge elk. Was this the true god of the Kroc empire whose everyday wealth (McDonald’s) came from lesser bovines?)
The porcine model of reflux nephropathy was a major advance in its day. However, like all good observations it leaves more questions than it answers. Does retrograde flow of sterile urine into the distal nephron signal changes in the tubular epithelium that affect function, polarity, or prime the locality for fibrosis or inflammation? Is infection a primary cause with bacteria or their products confined to the affected segment? What components of the immune system are at play in the inflamed renunculus and what switches that to fibrosis? Realistically none of these questions will be answered. Permission and facilities for large animal research have almost vanished, costs would be high and porcine reagents would need to be developed .
Perhaps because of these blocks to further investigation interest in the model has waned. The 21st century has seen terminology changed again. Reflux nephropathy has given way to Congenital Abnormality of Kidney and Urinary Tract (CAKUT) a catch all term that at least has the honesty of not pretending to describe a disease mechanism. The genetics revolution identified new causes of structural congenital disease such as HNF1beta mutations, but VUR even in family studies and outside known syndromic forms remains multifactorial. Although CAKUT encompasses about a third of advanced kidney disease in the young, it is far from clear how much of that is attributable to the acquired disorder described in the reflux nephropathy model above. From time to time a spirited defence of it is made, such as that by Malcolm Coulthard from Newcastle in 2009[22]. He also reported similar lesions acquired in transplanted kidneys exposed to infection[23]. The model is tantalising incomplete but not valueless.
References:
[1] Reflux Nephropathy. Hodson J. Med Clin N Amer 1978 62:1201-21
[2] Vesicoureteric reflux in children. Tullus K. Lancet 2015 385: 371-9
[3] Lancet Leader Lancet 2:301.1978
[4] Amar AD. Calicotubular backflow with vesicoureteral reflux. JAMA 1970 213: 293
[5] A potential hazard of Barium cystography. Brodeur AE, Goyer RA, Melick W, Radiology 1965 85: 1080
[6] Intrarenal reflux and the scarred kidney. Rolleston GL, Maling TM, Hodson CJ. Arch Dis Childh 1974 49: 531-9
[7] Tamminen TE Kaprio EA The relationship of the shape of the renal papilla and of the collecting duct openings to intrarenal reflux. Br J Urol 1977 49 345
[8] Hodson CJ. The lobar structure of the kidney. Br J Urol. 1972 44: 246-261
[9] Ransley PG and Risdon RA. Renal papillary morphology and intra renal reflux in infants and young children. Urol Res. 3: 111-113, 1975
[10] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2589891/?page=1 .
[11] Jodal U. The natural history of bacteriuria in childhood. Infect Dis Clin North Am 1987 1 713-29
[12] Randolph M Morris KE Gould EB The first urinary tract infection in the female infant. Prevalence, recurrence, and prognosis: a 10-year study in private practice. J Pediatr 1975 86 342-8 and personal communication.
[13] Smellie J et al. Clinical and radiological features of urinary infection in children Br Med J 1964 2 1222
[14] Hertelius-Elman M et al Infection 1988 16 263-6
[15] Smellie JM Normand IC. Urinary infection in children 1985. Postgrad Med 1985 61 895-905
[16] Habib M and Broyer M. Highlights segmental hyposplasia with hypertension(Ask-Upmark kidney) in Strauss J Editor Pediatric Nephrology 1981 pp113-4. Plenum New York.
[18] Smellie JM, Ransley P, Norman IC, Prescod N, Edwards D. Development of new renal scars: a collaborative study. Br Med J 1985 290 1957-60.
[19] Kahneman D. Thinking Fast and Slow. Penguin London 2011
[20] Birmingham Reflux Study Group. Prospective trial of operative versus non-operative treatment of severe vesicoureteric reflux: two years’ observation in 96 children. BrMed J 1983;287:171-4.
[21] Birmingham Reflux Study Group Prospective trial of operative versus non–operative treatment of severe vesicoureteric reflux in children: five years’ observation. Brit Med J 1987;295:237–241.
[22] Coulthard MG. Vesicoureteric reflux is not a benign condition. Pediatr Nephrology 2009 24 227-32
[23] Coulthard MG and Keir MJ. Reflux nephropathy in kidney transplants, demonstrated by dimercaptosuccinic acid scanning Transplantation 2006 82 205-10
Author: C Mark Taylor
Last Updated on October 28, 2024 by John Feehally