2
The case for surgery Barry O'Donnell
Controversy about the mangement (of primary uncomplicated vesicoureteric reflux seeims to be a straight choice between long term tre atment and surgery; "supervision" alone is not enougb1.' Classification of vesicoureteric reflux Classification of vesicoureteric reflux irs one of the unresolved issues. Three grades (1, 2, and 3) have been used in the Birmingham study and some c)ther studies in the United Kingdom. The classifi(cation most widely used in Europe and North America -that in the international reflux study in children3-hassfive grades, of which III-V correspond to grade 3 of the] Birmingham classification (figure 1). The difference affezcts management decisions on the vital issues of kidne y damage or scarring. Whereas perhaps less than halff of patients with grade III reflux have mild scarring aat diagnosis,
Royal College of Surgeons in Ireland Barry O'Donnell, FRCSI, professor of paediatnc surgery.
Correspondence to: Children's Research Centre, Our Lady's Hospital for Sick Children, Crumlin, Dublin 12, Republic of Ireland. Br Ac'd 7 1990;300:1393-4
International
II Grade I Birmingham
In
0
A I
n
.i
k
Grade 1
IV
k
2
--v-
3
FIG 1 -International and Birmingham classification systems for vesicoureteric reflux, illustrating wide variation within grade 3 of the Birmingham classification which are represented by grades III, IV, and V in the international classification to reflect prognostic differences
FIG 2-Left: cystogram showing bilateral grade V vesicoureteric reflux with severe scarring in a boy aged 4 years; right: micturating cystogram one year after endoscopic correction showing abolition of
reflux
BMJ
VOLUME 300
26 MAY 1990
85% or more of those with grade V reflux either have or will develop severe scarring. The more precise classification makes it easier to identify those systems in which reflux will cease with time. In many patients reflux stops over the years; grade III reflux eventually ceases in perhaps 60-70% of patients, grade IV in 2030%, and grade V in less than 10%. These divisions are important for the individual patient as well as having implications for the management policy for the condition. Will stopping reflux help to prevent progression of scarring or provide any added safety factor? We do not know the answer to this for certain, but doctors who prescribe long term antibiotic treatment for their patients do so for the duration of the reflux, presumably on the basis that the reflux of infected urine is the real threat. In the higher grades of reflux the bladder, ureters, pelves, and kidney substance are an open system, which is subject to voiding pressures of up to 100 cm H20 and to the resting pressure of a full bladder (40 cm H20) for some parts of the day. The question of whether sterile reflux can cause scarring over the years at these pressures remains unproved,4 but we do know that intrauterine reflux with presumably sterile urine can cause diffusely scarred, shrunken kidneys. We have found an increasing incidence of resistance to trimethoprim for long term treatment. It is now effective in only half of patients with the commonly infecting organisms, and spot checks show how seldom sterile urine is maintained over six months.5 Informed consent? Initially, parental views vary from "no surgery at any price" to "let's get it over with," with a wide range of views between these extremes. We have a duty to prompt parents to ask good, precise questions, to which they may get careful but imprecise answers. We tell parents of patients with grades IV and V vesicoureteric reflux that their children will probably have kidney scarring and loss of substance, and we carry out a `smTc dimercaptocuccinic acid (DMSA) renal scan as intravenous urography can underestimate damage.4 Scarring is the issue, and it takes nuclear scanning to measure it. Nephrectomy is then an option in cases of unilateral scarring with poor function (less than 12% of total).6 In patients with bilateral grades IV and V reflux and proved damage the condition may well lead to hypertension (a 10% possibility) or even to renal failure (a 3-5% possibility). With these higher grades of reflux the prospect of reflux persisting after five years of treatment is much more than 50%. White referred to reflux during filling being more likely when the "ureter is grossly dilated and the vesicoureteric orifice is permanently open."' We agree that there are such patients in whom this occurs, and the appearance of the ureteric orifice at cystoscopy is often dramatic and compelling.
Surgical management The open operation (of one and a half to two and a half hours' duration) necessary to reimplant the ureters in the bladder in such a way as to stop the reflux has, realistically, a 90% success rate in grade IV vesicoureteric reflux and an 80% success rate in grade V reflux. In patients with these grades of reflux it may fail to stop the reflux or, in perhaps 5% of all patients, it may cause obstruction necessitating reoperation. The 1393
successful operation enables patients to stop treatment within weeks. Surveillance continues; patients with scarred kidneys are under surveillance for life, but the nature of the surveillance changes. There are fewer specimens of urine and blood sent to the laboratory, and the patients make fewer visits to the diagnostic imaging department. The anxiety generated by the obsessional follow up necessary in managing antibiotic treatment is reduced when parents realise that their children's kidneys are unlikely to get worse if the reflux has been corrected. Surgery is of most obvious benefit to patients who have had frequent attacks of symptomatic infection either during treatment (breakthrough infection) or thereafter. Surgery often improves the distressing symptoms of wetting and frequency or abolishes them, and in young infants the capacity of the bladder increases.
Endoscopic management Vesicoureteric reflux can be corrected endoscopically; after experimental work' and six years' experience the method has favourable reports."-" The procedure is carried out as a day case; under general anaesthesia a cystoscope is passed and about 0 5 ml of polytetrafluoroethylene paste is injected under the affected orifice. This "sting"-subureteric Teflon injection-is painless. The "reported" total number of patients treated in this way world wide is over 4000. Are many patients offered this option? It has been suggested that endoscopic management be part of another controlled trial, but it can be set against the results of open surgery, which has been on trial for over 30 years. The endoscopic method works without the trauma of abdominal incisions, delicate surgery within the bladder, uncomfortable drainage tubes, and a week or more in hospital. New, smaller, instruments allow it to be applied even to full term newborn infants. It changes the balance in favour of intervention. Stop severe reflux today! 1 O'Donncll B, AMalonev MA, Lynch V. Vesico-urcteric reflux in intants and children: results of "supervision," chemothcrapy anid surgery. Br] U.rol
1969;41:6-13. 2 Birmingham Study Reflux (.roup.
Prospectivc trial of operative
versus
nlon-
opierative treatimncrt ol' sev-ere csicoirctcric reflux in chillrci: fihc ycars observation. Ir Mld 7 1987;295:237-41. 3 Intcrilationial Rcfllix Study Committee. Medical versus suirgical treatmelnt of primary vesicourcteral refltx: a prospective inlternationial reflux study in childrcn. 7 Urol 1981;125:774-8. 4 Risdon RA. Histological changcs in rcfl ix nephropathy. In: Hoey Hf(iC, Plotrt 1P, eds. lPiidiatrtc nephirologv: progress in researcli atnd practice. Chicliester: Weilv, 1990:14-7. 5 Deasv PF. Antibiotic treatinieiti of urinary tract itlfection. In: Hoev HMCV, Purl P, cds. I'aediatnctnephrolitgv: priogriss int risearih and practtice. Chichester: Wilex, 1990:34-9. 6 Quinlan 1), O'Dolincll B. Tnilateral tiretcric rcimplantation for primary vesicoureteric reflux in childrcn. Br 7 'rol 198S;57:4016-9. 7 White RHR. Vcsicoureteric reflux atid retial scarriig. Ireh I)is Child 1989;64:407- 12. 8 Ptirt P, O'Donnell B. Cosrrcction of experimenetally produced vesicourctcric reflux in the piglet by intravesical injection of tclhon. BrMcdj 1984;289:5-7. 9 Purl P, O'Donnell B. Endoscoupic correction of' grades IV and V primary vesicoureteric rcflux: six to 30 mnonth follow-up itt 42 ureters. 7 Pcditatr Surg 1987;22: 1087-91. 10 Kaplan WE, Dalton DP, Firlit CF. The endoscopic correction of reflux by polytetrafluoroethylene intjection. J U rol 1987;138:953-5. 11 Schulman CC. Simon J, Parmor D, et al. Endoscopic treatment otu veCsicsurcteral reflux in children. Urol 1987;138:950-1. 12 King PA, (iollow I. The endoscopic correctioia ou' vciscourctcric renux. Aist N Z,7 Surg 1988;58:569-7 1.
Correction General practitioner obstetrics in Bradford Two authors' errors occurred in table VI of this paper by Dr F C Brvce and others (17 March, p 725). In women transferred antenatallv the figures for the stillbirth rate and perinatal mortalitv were 8-0 per 1000 and 10 6 per 1000 respectively and not 27 0 per 1000 and 38-6 per 1000 as published.
Can the community care? Two printer's errors occurred in this article by Dr Trish Groves (5 May, p 1186). The beginning of the second paragraph on p 1188 should read "We also need better information about the failures of community care. It is widely assumed that there is a direct relation between closures of mental hospitals and the increase in homelessness in Britain." Also, the seventeenth line of the second column on p 1 188 should state that the insight that the case gives is depressing and not depressive as published.
Vitamin B-12 and folate deficiency presenting as leukaemia A printer's error occurred in this lesson of the week by Dr I S Dokal and others (12 May, p 1263). The legends accompanying figures 3 and 4 were inadvertently transposed after the page proof stage.
ANY QUESTIONS What contraindications are there for anaesthesia in patients receiving 13 blocking agents? There are no contraindications to anaesthetising patients taking 13 blocking agents, but there are certain implications for the conduct of anaesthesia and postoperative care. The commoner indications for 13 blockade in patients presenting for anaesthesia are hypertension, angina pectoris, tachyarrhythmia, and thyrotoxicosis. Withdrawal of treatment in such cases is dangerous, and an appropriate level of 3 blockade should be continued perioperatively in such patients, by intravenous administration if necessary. The anaesthetist must assess, however, whether the level of 3 blockade found on examination is appropriate and mav decide that the dosage should be adjusted before operation. Patients whose symptoms are inadequately controlled should receive additional treatment before surgery. The medication of patients receiving 3 blockers who are also suffering from bronchospasm, heart failure, or syncopal attacks requires particularly careful review before surgery. The difficult question is whether otherwise asymptomatic patients with a profound bradycardia pose a particular risk under anaesthesia. A heart rate of :40 beats/minute is potentially dangerous because any further fall in cardiac output under anaesthesia might be critical. In these circumstances and in the absence of an appropriate heart rate response to mild exercise the dose of I3 blocking agent should be reduced before surgery. Modern anaesthetic agents are largely free of vagolytic side effects, and
1394
the reduction in the routine use of anticholinergic agents in premedication has led to reports of profound bradycardia during anaesthesia. Patients receiving 13 blockers may be at extra risk,' but atropine or glycopyrrolate can be used to block increases in vagal tone. Continuous electrocardiographic monitoring is essential, and during surgery it is particularly important to prevent tachycardia in response to laryngoscopy or operative stimulation.' After major surgery circulating catecholamine concentrations rise and tissue oxygen demand increases. Patients who fail to achieve adequate increases in cardiac output and oxygen delivery to the tissues have a higher risk of dying postoperatively.) The conflicting goals of protecting the myocardium from excessive work demands while simultaneously providing a supranormal cardiac, output require close monitoring of the cardiorespiratory state after the operation with prompt pharmacological intervention when necessary. Treatment in an intensive care unit may be appropriate for such patients. Esmolol, a short acting 1 blocking agent suitable for continuous infusion during anaesthesia and intensive care, should soon be available for clinical use. -T WOODCOCK, consultant in anaesthesia and intensive care, Southampton Eldor J, Hoffman Bn Davidson JT. Pr6longed bradycardia and hypotension after neostigmine administration in a patient receiving atenolol. Anaesthesia 1987;42:1294-7. 2 Slogoff S, Keats AS. Does perioperative mvocardial ischemia lead to postoperative mvocardial infarction? Anesthesiology 19855;62:107-14. 3 Shoemaker WC. Circulatory mechanisms of shock and their mediators. Crit Care Med 1987:15:787-96.
BMJ VOLUME 300
26 MAY 1990
The case for surgery Barry O'Donnell
Controversy about the mangement (of primary uncomplicated vesicoureteric reflux seeims to be a straight choice between long term tre atment and surgery; "supervision" alone is not enougb1.' Classification of vesicoureteric reflux Classification of vesicoureteric reflux irs one of the unresolved issues. Three grades (1, 2, and 3) have been used in the Birmingham study and some c)ther studies in the United Kingdom. The classifi(cation most widely used in Europe and North America -that in the international reflux study in children3-hassfive grades, of which III-V correspond to grade 3 of the] Birmingham classification (figure 1). The difference affezcts management decisions on the vital issues of kidne y damage or scarring. Whereas perhaps less than halff of patients with grade III reflux have mild scarring aat diagnosis,
Royal College of Surgeons in Ireland Barry O'Donnell, FRCSI, professor of paediatnc surgery.
Correspondence to: Children's Research Centre, Our Lady's Hospital for Sick Children, Crumlin, Dublin 12, Republic of Ireland. Br Ac'd 7 1990;300:1393-4
International
II Grade I Birmingham
In
0
A I
n
.i
k
Grade 1
IV
k
2
--v-
3
FIG 1 -International and Birmingham classification systems for vesicoureteric reflux, illustrating wide variation within grade 3 of the Birmingham classification which are represented by grades III, IV, and V in the international classification to reflect prognostic differences
FIG 2-Left: cystogram showing bilateral grade V vesicoureteric reflux with severe scarring in a boy aged 4 years; right: micturating cystogram one year after endoscopic correction showing abolition of
reflux
BMJ
VOLUME 300
26 MAY 1990
85% or more of those with grade V reflux either have or will develop severe scarring. The more precise classification makes it easier to identify those systems in which reflux will cease with time. In many patients reflux stops over the years; grade III reflux eventually ceases in perhaps 60-70% of patients, grade IV in 2030%, and grade V in less than 10%. These divisions are important for the individual patient as well as having implications for the management policy for the condition. Will stopping reflux help to prevent progression of scarring or provide any added safety factor? We do not know the answer to this for certain, but doctors who prescribe long term antibiotic treatment for their patients do so for the duration of the reflux, presumably on the basis that the reflux of infected urine is the real threat. In the higher grades of reflux the bladder, ureters, pelves, and kidney substance are an open system, which is subject to voiding pressures of up to 100 cm H20 and to the resting pressure of a full bladder (40 cm H20) for some parts of the day. The question of whether sterile reflux can cause scarring over the years at these pressures remains unproved,4 but we do know that intrauterine reflux with presumably sterile urine can cause diffusely scarred, shrunken kidneys. We have found an increasing incidence of resistance to trimethoprim for long term treatment. It is now effective in only half of patients with the commonly infecting organisms, and spot checks show how seldom sterile urine is maintained over six months.5 Informed consent? Initially, parental views vary from "no surgery at any price" to "let's get it over with," with a wide range of views between these extremes. We have a duty to prompt parents to ask good, precise questions, to which they may get careful but imprecise answers. We tell parents of patients with grades IV and V vesicoureteric reflux that their children will probably have kidney scarring and loss of substance, and we carry out a `smTc dimercaptocuccinic acid (DMSA) renal scan as intravenous urography can underestimate damage.4 Scarring is the issue, and it takes nuclear scanning to measure it. Nephrectomy is then an option in cases of unilateral scarring with poor function (less than 12% of total).6 In patients with bilateral grades IV and V reflux and proved damage the condition may well lead to hypertension (a 10% possibility) or even to renal failure (a 3-5% possibility). With these higher grades of reflux the prospect of reflux persisting after five years of treatment is much more than 50%. White referred to reflux during filling being more likely when the "ureter is grossly dilated and the vesicoureteric orifice is permanently open."' We agree that there are such patients in whom this occurs, and the appearance of the ureteric orifice at cystoscopy is often dramatic and compelling.
Surgical management The open operation (of one and a half to two and a half hours' duration) necessary to reimplant the ureters in the bladder in such a way as to stop the reflux has, realistically, a 90% success rate in grade IV vesicoureteric reflux and an 80% success rate in grade V reflux. In patients with these grades of reflux it may fail to stop the reflux or, in perhaps 5% of all patients, it may cause obstruction necessitating reoperation. The 1393
successful operation enables patients to stop treatment within weeks. Surveillance continues; patients with scarred kidneys are under surveillance for life, but the nature of the surveillance changes. There are fewer specimens of urine and blood sent to the laboratory, and the patients make fewer visits to the diagnostic imaging department. The anxiety generated by the obsessional follow up necessary in managing antibiotic treatment is reduced when parents realise that their children's kidneys are unlikely to get worse if the reflux has been corrected. Surgery is of most obvious benefit to patients who have had frequent attacks of symptomatic infection either during treatment (breakthrough infection) or thereafter. Surgery often improves the distressing symptoms of wetting and frequency or abolishes them, and in young infants the capacity of the bladder increases.
Endoscopic management Vesicoureteric reflux can be corrected endoscopically; after experimental work' and six years' experience the method has favourable reports."-" The procedure is carried out as a day case; under general anaesthesia a cystoscope is passed and about 0 5 ml of polytetrafluoroethylene paste is injected under the affected orifice. This "sting"-subureteric Teflon injection-is painless. The "reported" total number of patients treated in this way world wide is over 4000. Are many patients offered this option? It has been suggested that endoscopic management be part of another controlled trial, but it can be set against the results of open surgery, which has been on trial for over 30 years. The endoscopic method works without the trauma of abdominal incisions, delicate surgery within the bladder, uncomfortable drainage tubes, and a week or more in hospital. New, smaller, instruments allow it to be applied even to full term newborn infants. It changes the balance in favour of intervention. Stop severe reflux today! 1 O'Donncll B, AMalonev MA, Lynch V. Vesico-urcteric reflux in intants and children: results of "supervision," chemothcrapy anid surgery. Br] U.rol
1969;41:6-13. 2 Birmingham Study Reflux (.roup.
Prospectivc trial of operative
versus
nlon-
opierative treatimncrt ol' sev-ere csicoirctcric reflux in chillrci: fihc ycars observation. Ir Mld 7 1987;295:237-41. 3 Intcrilationial Rcfllix Study Committee. Medical versus suirgical treatmelnt of primary vesicourcteral refltx: a prospective inlternationial reflux study in childrcn. 7 Urol 1981;125:774-8. 4 Risdon RA. Histological changcs in rcfl ix nephropathy. In: Hoey Hf(iC, Plotrt 1P, eds. lPiidiatrtc nephirologv: progress in researcli atnd practice. Chicliester: Weilv, 1990:14-7. 5 Deasv PF. Antibiotic treatinieiti of urinary tract itlfection. In: Hoev HMCV, Purl P, cds. I'aediatnctnephrolitgv: priogriss int risearih and practtice. Chichester: Wilex, 1990:34-9. 6 Quinlan 1), O'Dolincll B. Tnilateral tiretcric rcimplantation for primary vesicoureteric reflux in childrcn. Br 7 'rol 198S;57:4016-9. 7 White RHR. Vcsicoureteric reflux atid retial scarriig. Ireh I)is Child 1989;64:407- 12. 8 Ptirt P, O'Donnell B. Cosrrcction of experimenetally produced vesicourctcric reflux in the piglet by intravesical injection of tclhon. BrMcdj 1984;289:5-7. 9 Purl P, O'Donnell B. Endoscoupic correction of' grades IV and V primary vesicoureteric rcflux: six to 30 mnonth follow-up itt 42 ureters. 7 Pcditatr Surg 1987;22: 1087-91. 10 Kaplan WE, Dalton DP, Firlit CF. The endoscopic correction of reflux by polytetrafluoroethylene intjection. J U rol 1987;138:953-5. 11 Schulman CC. Simon J, Parmor D, et al. Endoscopic treatment otu veCsicsurcteral reflux in children. Urol 1987;138:950-1. 12 King PA, (iollow I. The endoscopic correctioia ou' vciscourctcric renux. Aist N Z,7 Surg 1988;58:569-7 1.
Correction General practitioner obstetrics in Bradford Two authors' errors occurred in table VI of this paper by Dr F C Brvce and others (17 March, p 725). In women transferred antenatallv the figures for the stillbirth rate and perinatal mortalitv were 8-0 per 1000 and 10 6 per 1000 respectively and not 27 0 per 1000 and 38-6 per 1000 as published.
Can the community care? Two printer's errors occurred in this article by Dr Trish Groves (5 May, p 1186). The beginning of the second paragraph on p 1188 should read "We also need better information about the failures of community care. It is widely assumed that there is a direct relation between closures of mental hospitals and the increase in homelessness in Britain." Also, the seventeenth line of the second column on p 1 188 should state that the insight that the case gives is depressing and not depressive as published.
Vitamin B-12 and folate deficiency presenting as leukaemia A printer's error occurred in this lesson of the week by Dr I S Dokal and others (12 May, p 1263). The legends accompanying figures 3 and 4 were inadvertently transposed after the page proof stage.
ANY QUESTIONS What contraindications are there for anaesthesia in patients receiving 13 blocking agents? There are no contraindications to anaesthetising patients taking 13 blocking agents, but there are certain implications for the conduct of anaesthesia and postoperative care. The commoner indications for 13 blockade in patients presenting for anaesthesia are hypertension, angina pectoris, tachyarrhythmia, and thyrotoxicosis. Withdrawal of treatment in such cases is dangerous, and an appropriate level of 3 blockade should be continued perioperatively in such patients, by intravenous administration if necessary. The anaesthetist must assess, however, whether the level of 3 blockade found on examination is appropriate and mav decide that the dosage should be adjusted before operation. Patients whose symptoms are inadequately controlled should receive additional treatment before surgery. The medication of patients receiving 3 blockers who are also suffering from bronchospasm, heart failure, or syncopal attacks requires particularly careful review before surgery. The difficult question is whether otherwise asymptomatic patients with a profound bradycardia pose a particular risk under anaesthesia. A heart rate of :40 beats/minute is potentially dangerous because any further fall in cardiac output under anaesthesia might be critical. In these circumstances and in the absence of an appropriate heart rate response to mild exercise the dose of I3 blocking agent should be reduced before surgery. Modern anaesthetic agents are largely free of vagolytic side effects, and
1394
the reduction in the routine use of anticholinergic agents in premedication has led to reports of profound bradycardia during anaesthesia. Patients receiving 13 blockers may be at extra risk,' but atropine or glycopyrrolate can be used to block increases in vagal tone. Continuous electrocardiographic monitoring is essential, and during surgery it is particularly important to prevent tachycardia in response to laryngoscopy or operative stimulation.' After major surgery circulating catecholamine concentrations rise and tissue oxygen demand increases. Patients who fail to achieve adequate increases in cardiac output and oxygen delivery to the tissues have a higher risk of dying postoperatively.) The conflicting goals of protecting the myocardium from excessive work demands while simultaneously providing a supranormal cardiac, output require close monitoring of the cardiorespiratory state after the operation with prompt pharmacological intervention when necessary. Treatment in an intensive care unit may be appropriate for such patients. Esmolol, a short acting 1 blocking agent suitable for continuous infusion during anaesthesia and intensive care, should soon be available for clinical use. -T WOODCOCK, consultant in anaesthesia and intensive care, Southampton Eldor J, Hoffman Bn Davidson JT. Pr6longed bradycardia and hypotension after neostigmine administration in a patient receiving atenolol. Anaesthesia 1987;42:1294-7. 2 Slogoff S, Keats AS. Does perioperative mvocardial ischemia lead to postoperative mvocardial infarction? Anesthesiology 19855;62:107-14. 3 Shoemaker WC. Circulatory mechanisms of shock and their mediators. Crit Care Med 1987:15:787-96.
BMJ VOLUME 300
26 MAY 1990
