372
argument. More fundamental considerations should be addressed first. These patients are not dead. The physician’s responsibility should be primarily to his patient. Obtaining "decisions" and "permission" from relatives about patients who are not dead seems wrong: surely the relatives should be asked to indicate what they believe the patient might have wanted. The use of artificial ventilation in intensive care units has allowed two rare conditions to become recognised: one, the more common, is brain death, which can result in the fortunate "byproduct" of organs for donation; and the other is a persistent vegetative state, which is far less fortunate. When a seriously ill patient is about to be admitted to intensive care for artificial ventilation should relatives be told that a vegetative state is a possible outcome? The way that information is given can greatly influence decisions. In the data presented, one out of eleven patients did not die as predicted. Do Feest et al propose that relatives are told that there is this level of probability that death is not inevitable, but that the patient may survive in a severely damaged state? Three patients were intubated and ventilated in the emergency department before discussion with a consultant or the relatives. This seems to be presenting the relatives with a fait accompli, and may well force a decision that might not otherwise have been reached. The proposed protocol moves away from the well-defined condition of "brain death" to one of "expected death on a ventilator", which is far less well characterised, and is possibly a dangerous concept. Department of Anaesthetics, Royal Infirmary, Edinburgh EH3 9YW, 1 Schneiderman
G. B. DRUMMOND
UK
LJ, Spragg RG Ethical decisions J Med 1988; 318: 984-88
m
discontinuing
mechanical
ventilation. N Engl
Clinical significance of antenatal calyceal dilatation detected by ultrasound SiR,—The role of ultrasound in the prenatal detection of renal tract anomalies is well established’ but the significance of the findings and the most rational use of confirmatory investigations remain unclear. At the Birmingham Maternity Hospital we have been evaluating antenatal ultrasound screening since 1986. Once an abnormality is detected ultrasound scans are repeated during pregnancy, in the immediate postnatal period, and at six weeks after birth. Investigations and management decisions are planned at monthly meetings of the perinatal renal group at the Birmingham Children’s Hospital. Over the two years January, 1988, to December, 1989, 110 infants were live-born after detection of fetal urological abnormalities; significant lesions were demonstrated in 42%:
All scans showed pelvic dilatation; they were according to the worst appearance antenatally, after delivery, or at 6 weeks. The presence of additional calyceal dilatation was closely associated with the subsequent demonstration of renal lesion, whereas no abnormalities were found in any of the 13 infants with isolated dilatation of the renal pelvis:
cystourethrogram. then classified
These data raise two important issues. The true incidence of VUR in childhood is unknown. However, on the assumption that all VUR is congenital, and given that urinary-tract infection develops in 1% of infants and that one-quarter of these have significant reflux, a conservative estimate of the incidence of VUR would be 2per 1000. If some infants with reflux do not proceed to infection then the incidence would be greater. Since we studied well over 10 000 pregnancies it seems likely that antenatal ultrasound screening does not always identify VUR. Our results also suggest that calyceal dilatation found at any time is an indication for action, including cystourethrogram-but is this investigation justifiable in infants with dilatation confined to the renal pelvis? Our limited data suggest that VUR occurs in less than 25 % of these infants. To explore the issue further we calculated how large a study would be necessary to provide a 90% chance of allowing us to infer, with 95% confidence, the frequency of VUR m this group. If the frequency in infants with isolated pelvic dilatation is assumed to be 10-20%, a further 20 infants are required; if it is 5-10%, 100 infants are needed. An assumed frequency of less than 5% would require a large number of infants and would not justify the widespread use of cystourethrography in clinical practice. These important questions can be answered only by collaborative studies and we would welcome the comments of others involved in programmes to detect fetal renal anomalies. We thank Dr R. L.
Holder, department of
statistics,
University of
Birmingham. Departments of Paediatrics and Radiology, Birmingham Maternity Hospital, Birmingham B15 2TG, UK
S. J. NEWELL M. E. I. MORGAN J. M. MCHUGO R. H. R. WHITE C. M. TAYLOR S. CHAPMAN
Departments of Nephrology. Radiology, and Surgery, Children’s Hospital, Birmingham 1
K. J. SHAH P. GORNALL J. J. CORKERY
Tudor J, Whitaker RH Detection and management of the dilated fetal urinary tract
Clin Radiol 1989, 40: 229-30. 2 White RHR Fetal uropathy Br Med J 1989; 298: 1408-09. 3 Reuter KL, Lebowitz RL. Massive vesico-ureteric reflux
mimicking posterior
urethral valves in a fetus. J Clin Ultrasound 1985; 13: 584-87 4. Rolleston GL, Shannon FT, Utley WLE Relationship of infantile reflux to renal damage Br Med J 1970, i 460-64.
vesicoureteric
Bicarbonate and cytoplasmic pH
One aim of screening is to detect asymptomatic vesicoureteric reflux (VUR), in an attempt to prevent reflux nephropathy.2 We have analysed our data with special reference to the detection of VUR by cystourethrography in infants with antenatal renal tract dilatation .3 In 41 infants investigations included a cystourethrogram because of pelvicalyceal dilatation. 9 infants had grade4 III VUR and 2 had grade II. A further infant showed bladder outlet obstruction and another a ureterocoele. In 10 the cystourethrogram was normal but other investigations showed renal tract lesions. In 18 infants no significant lesions was found on cystourethrography or other investigations. A cystourethrogram is an invasive investigation so we looked at the ultrasound scans to see if they were predictive of an abnormal
SIR,-We would like to respond to points raised by correspondents (June 30, p 1586-88) about our paper on effects of bicarbonate on cytoplasmic pH (May 26, p 1243). We argued that the finding that an abrupt increase in bicarbonate concentration can cause cytoplasmic acidification undermines the logic of using bicarbonate in severe acidaemia, especially in cardiopulmonary arrest, and adds to the evidence that such treatment may be dangerous. Dr Buckler and Dr Gasser argue that the mechanism of acidification may not be exclusively permeation of CO, followed by hydration and dissociation, and we agree that we cannot exclude additional mechanisms. Dr Unwin makes the point that platelets may not be a good general model of cell acid-base regulation. However, platelets are prepared fresh from blood, whereas many cells require culture and/or dispersal methods that may influence acid-base responses, especially if these depend on carbonic anhydrase on the outer
373
communication). The cited rate of inhibitor formation in previously unexposed patients (7 of 39 or 18%) is not extraordinary or unexpected.2 The Malmö procedure for induction of immune tolerance has been successfully applied to many patients, but the actual response rate of patients is unknown.3 Of the 11 patients treated by Nilsson et al, 2 (18%) did not respond. There is no way to know the significance of "the failure of the immune tolerance induction" in Kessler and Sachse’s patient. The speculation that the manufacture of monoclonal-antibody-derived factor VIII:C could be associated with subtle denaturation of a native factor VIII:C protein, and creation of a neoantigen, is without factual support. The article referenced, with respect to alterations in the factor VIII:C gene, actually provides data that argue against this mechanism. Extensive animal testing during the preclinical laboratory evaluation of ’Monoclate’ failed to reveal any evidence of neoantigen formation.4 It is premature to suggest that use of ultra-pure factor VIII:C products is related to an increased rate of inhibitor development, or to speculate on their failure to suppress inhibitors.
Effect of sodium bicarbonate (12 mmol/I. at arrow) on cytoplasmic pH (pH) in bovine aortic endothelial cells (AG4762, National Instituteof Aging cellrepository, passage 11) loaded with BCECF, trypsinised and suspended in stirred HEPES buffered Tyrode’s solution at room temperature.
surface of the cell membrane. Despite this, we have observed qualitatively similar responses in suspensions of bovine aortic endothelial cells (figure), so the effect is not peculiar to platelets. Dr Iles and Dr Beech have very interesting data demonstrating increased pH in liver caused by a 15 min intravenous infusion of molar sodium bicarbonate in the ketoacidotic rat. It will be particularly interesting when the resolving power of this method is increased to permit measurement of the effect of bicarbonate on pH in cardiac muscle in animal models of cardiac arrest. Professor Cohen and Professor Woods in their 1987 chapter’ on treatment of acid-base disturbances presented a balanced account of the arguments, and we are sorry that they feel that our statement that they advocated the use of bicarbonate misrepresented them. They did write that hypertonic bicarbonate is standard therapy for the acidosis of cardiac arrest, but also pointed out that such therapy is not of unequivocal benefit. There are several precedents of treatments once used routinely in severely ill patients on grounds that they appeared reasonable but which have subsequently been proved to be positively harmful. Although our studies showing reduction in pH, in response to bicarbonate in vitro are not conclusive evidence that such a deleterious effect occurs during bicarbonate therapy, we feel that the findings shift the burden of proof towards those that favour such treatment: perhaps the time is right to plan controlled trials of bicarbonate therapy despite the formidable difficulties that Cohen and Woods rightly identify. ’
J. M. RITTER Department of Clinical Pharmacology, Guy’s Hospital, London Bridge SE1 9RT
MICHAEL B. RODELL GARRETT E. BERGMAN
Armour Pharmaceutical Company, Blue Bell, Pennsylvania 19422, USA
H. S. DOKTOR N. BENJAMIN
RD, Woods HF. Disturbances of acid-base homeostasis. In. Weatherall DJ, Ledingham JGG, Warrell DA, eds. Oxford textbook of medicine, 2nd ed Oxford: Oxford University Press, 1987: 164-75.
1 Cohen
Factor VIII:C inhibitor associated with monoclonal-antibody-purified FVIII
CW, Shapiro SS, et al The natural history of factor VIII:C inhibitors in patients with hemophilia A: a national cooperative study, II: observations on the initial development of factor VIII:C inhibitors Blood 1988; 71: 344-48. 2. Lusher JM, Salzman PM, et al Viral safety and inhibitor development associated with factor VIII:C ultra-purified from plasma in hemophiliacs previously unexposed to factor VIII.C concentrates. Seminar Hematol 1990; 27 (suppl 2): 1-8 3. Nilsson IM, Bemtorp E, Zettervall O. Induction of immune tolerance in patients with hemophilia and antibodies to factor VIII by combined treatment with intravenous IgG, cyclophosphamide, and factor VIII N Engl JMed 1988; 318: 947-50 4. Hrinda ME, Feldman F, Schreiber AB. Preclinical characterization of a new pasteurized monoclonal antibody purified factor VIII.C. Seminar Hematol 1990; 1 McMillan
27
(suppl 2):
19-25.
Prevention of spontaneous diabetes in BB rats with FK 506 SIR,-Cyclosporin
can
mitigate
or
prevent autoimmune BB
rat
diabetes,’-3 and can induce remissions of human type I diabetes if instituted early after diagnosis and in adequate doses.4-6 The potent new immunosuppressive agent, FK 506, which has had extensive clinical trials in transplantation7 suppresses T-cell activation in the same general way as cyclosporin does, by inhibiting the transcription of mRNA for early phase lymphokines such as interleukin-2, interleukin-3, and y-interferon.8 We have examined the influence of daily enteral tap water (group 1, n = 20), 1 mg/kg FK 506 (group 2, n = 19), or 2 mg/kg FK 506 (group 3, n = 20) on the incidence of spontaneous diabetes in our BB rat colony. The 59 animals were obtained from seven consecutive litters, randomly deployed to the three sex-matched groups, raised under controlled lighting, and allowed tap water and regular rat food (Purina ’Chow 5001’) ad libitum. Treatment was between 30 and 120 days of age. The FK 506 (Fujisawa Pharmaceutical, Osaka) was dispersed with hydroxypropylmethylcellulose, diluted in tap water by sonication, and administered by gastric intubation. Onset of diabetes was diagnosed by glycosuria (tested thrice weekly) and confirmed by non-fasted blood sugar above 200 mg/dl from tail vein samples for 3 consecutive days.
concentrate
SIR,-Several comments need to be made to put in perspective the letter by Dr Kessler and Dr Sachse (June 9, p 1403) concerning factor VIII:C inhibitor formation. The actual rate of inhibitor formation in patients with severe haemophilia A, followed from childhood to adulthood, is unknown. The estimated rates of 5-15% are derived from cross-sectional rather than longitudinal studies and include patients with all degrees of severity. In one large "natural history" study, 29% of patients had inhibitor after 30 years of age, and at a time when only crude preparations of factor VIII:C were available.’ Most inhibitors developed in children. It has been estimated that if severely affected (factor VIII:C < 1 %) children were followed for 20 or 30 years then an inhibitor might develop in 30% or more (L. Aledort, personal
Onset and incidence of diabetes during FK 506 treatment.
argument. More fundamental considerations should be addressed first. These patients are not dead. The physician’s responsibility should be primarily to his patient. Obtaining "decisions" and "permission" from relatives about patients who are not dead seems wrong: surely the relatives should be asked to indicate what they believe the patient might have wanted. The use of artificial ventilation in intensive care units has allowed two rare conditions to become recognised: one, the more common, is brain death, which can result in the fortunate "byproduct" of organs for donation; and the other is a persistent vegetative state, which is far less fortunate. When a seriously ill patient is about to be admitted to intensive care for artificial ventilation should relatives be told that a vegetative state is a possible outcome? The way that information is given can greatly influence decisions. In the data presented, one out of eleven patients did not die as predicted. Do Feest et al propose that relatives are told that there is this level of probability that death is not inevitable, but that the patient may survive in a severely damaged state? Three patients were intubated and ventilated in the emergency department before discussion with a consultant or the relatives. This seems to be presenting the relatives with a fait accompli, and may well force a decision that might not otherwise have been reached. The proposed protocol moves away from the well-defined condition of "brain death" to one of "expected death on a ventilator", which is far less well characterised, and is possibly a dangerous concept. Department of Anaesthetics, Royal Infirmary, Edinburgh EH3 9YW, 1 Schneiderman
G. B. DRUMMOND
UK
LJ, Spragg RG Ethical decisions J Med 1988; 318: 984-88
m
discontinuing
mechanical
ventilation. N Engl
Clinical significance of antenatal calyceal dilatation detected by ultrasound SiR,—The role of ultrasound in the prenatal detection of renal tract anomalies is well established’ but the significance of the findings and the most rational use of confirmatory investigations remain unclear. At the Birmingham Maternity Hospital we have been evaluating antenatal ultrasound screening since 1986. Once an abnormality is detected ultrasound scans are repeated during pregnancy, in the immediate postnatal period, and at six weeks after birth. Investigations and management decisions are planned at monthly meetings of the perinatal renal group at the Birmingham Children’s Hospital. Over the two years January, 1988, to December, 1989, 110 infants were live-born after detection of fetal urological abnormalities; significant lesions were demonstrated in 42%:
All scans showed pelvic dilatation; they were according to the worst appearance antenatally, after delivery, or at 6 weeks. The presence of additional calyceal dilatation was closely associated with the subsequent demonstration of renal lesion, whereas no abnormalities were found in any of the 13 infants with isolated dilatation of the renal pelvis:
cystourethrogram. then classified
These data raise two important issues. The true incidence of VUR in childhood is unknown. However, on the assumption that all VUR is congenital, and given that urinary-tract infection develops in 1% of infants and that one-quarter of these have significant reflux, a conservative estimate of the incidence of VUR would be 2per 1000. If some infants with reflux do not proceed to infection then the incidence would be greater. Since we studied well over 10 000 pregnancies it seems likely that antenatal ultrasound screening does not always identify VUR. Our results also suggest that calyceal dilatation found at any time is an indication for action, including cystourethrogram-but is this investigation justifiable in infants with dilatation confined to the renal pelvis? Our limited data suggest that VUR occurs in less than 25 % of these infants. To explore the issue further we calculated how large a study would be necessary to provide a 90% chance of allowing us to infer, with 95% confidence, the frequency of VUR m this group. If the frequency in infants with isolated pelvic dilatation is assumed to be 10-20%, a further 20 infants are required; if it is 5-10%, 100 infants are needed. An assumed frequency of less than 5% would require a large number of infants and would not justify the widespread use of cystourethrography in clinical practice. These important questions can be answered only by collaborative studies and we would welcome the comments of others involved in programmes to detect fetal renal anomalies. We thank Dr R. L.
Holder, department of
statistics,
University of
Birmingham. Departments of Paediatrics and Radiology, Birmingham Maternity Hospital, Birmingham B15 2TG, UK
S. J. NEWELL M. E. I. MORGAN J. M. MCHUGO R. H. R. WHITE C. M. TAYLOR S. CHAPMAN
Departments of Nephrology. Radiology, and Surgery, Children’s Hospital, Birmingham 1
K. J. SHAH P. GORNALL J. J. CORKERY
Tudor J, Whitaker RH Detection and management of the dilated fetal urinary tract
Clin Radiol 1989, 40: 229-30. 2 White RHR Fetal uropathy Br Med J 1989; 298: 1408-09. 3 Reuter KL, Lebowitz RL. Massive vesico-ureteric reflux
mimicking posterior
urethral valves in a fetus. J Clin Ultrasound 1985; 13: 584-87 4. Rolleston GL, Shannon FT, Utley WLE Relationship of infantile reflux to renal damage Br Med J 1970, i 460-64.
vesicoureteric
Bicarbonate and cytoplasmic pH
One aim of screening is to detect asymptomatic vesicoureteric reflux (VUR), in an attempt to prevent reflux nephropathy.2 We have analysed our data with special reference to the detection of VUR by cystourethrography in infants with antenatal renal tract dilatation .3 In 41 infants investigations included a cystourethrogram because of pelvicalyceal dilatation. 9 infants had grade4 III VUR and 2 had grade II. A further infant showed bladder outlet obstruction and another a ureterocoele. In 10 the cystourethrogram was normal but other investigations showed renal tract lesions. In 18 infants no significant lesions was found on cystourethrography or other investigations. A cystourethrogram is an invasive investigation so we looked at the ultrasound scans to see if they were predictive of an abnormal
SIR,-We would like to respond to points raised by correspondents (June 30, p 1586-88) about our paper on effects of bicarbonate on cytoplasmic pH (May 26, p 1243). We argued that the finding that an abrupt increase in bicarbonate concentration can cause cytoplasmic acidification undermines the logic of using bicarbonate in severe acidaemia, especially in cardiopulmonary arrest, and adds to the evidence that such treatment may be dangerous. Dr Buckler and Dr Gasser argue that the mechanism of acidification may not be exclusively permeation of CO, followed by hydration and dissociation, and we agree that we cannot exclude additional mechanisms. Dr Unwin makes the point that platelets may not be a good general model of cell acid-base regulation. However, platelets are prepared fresh from blood, whereas many cells require culture and/or dispersal methods that may influence acid-base responses, especially if these depend on carbonic anhydrase on the outer
373
communication). The cited rate of inhibitor formation in previously unexposed patients (7 of 39 or 18%) is not extraordinary or unexpected.2 The Malmö procedure for induction of immune tolerance has been successfully applied to many patients, but the actual response rate of patients is unknown.3 Of the 11 patients treated by Nilsson et al, 2 (18%) did not respond. There is no way to know the significance of "the failure of the immune tolerance induction" in Kessler and Sachse’s patient. The speculation that the manufacture of monoclonal-antibody-derived factor VIII:C could be associated with subtle denaturation of a native factor VIII:C protein, and creation of a neoantigen, is without factual support. The article referenced, with respect to alterations in the factor VIII:C gene, actually provides data that argue against this mechanism. Extensive animal testing during the preclinical laboratory evaluation of ’Monoclate’ failed to reveal any evidence of neoantigen formation.4 It is premature to suggest that use of ultra-pure factor VIII:C products is related to an increased rate of inhibitor development, or to speculate on their failure to suppress inhibitors.
Effect of sodium bicarbonate (12 mmol/I. at arrow) on cytoplasmic pH (pH) in bovine aortic endothelial cells (AG4762, National Instituteof Aging cellrepository, passage 11) loaded with BCECF, trypsinised and suspended in stirred HEPES buffered Tyrode’s solution at room temperature.
surface of the cell membrane. Despite this, we have observed qualitatively similar responses in suspensions of bovine aortic endothelial cells (figure), so the effect is not peculiar to platelets. Dr Iles and Dr Beech have very interesting data demonstrating increased pH in liver caused by a 15 min intravenous infusion of molar sodium bicarbonate in the ketoacidotic rat. It will be particularly interesting when the resolving power of this method is increased to permit measurement of the effect of bicarbonate on pH in cardiac muscle in animal models of cardiac arrest. Professor Cohen and Professor Woods in their 1987 chapter’ on treatment of acid-base disturbances presented a balanced account of the arguments, and we are sorry that they feel that our statement that they advocated the use of bicarbonate misrepresented them. They did write that hypertonic bicarbonate is standard therapy for the acidosis of cardiac arrest, but also pointed out that such therapy is not of unequivocal benefit. There are several precedents of treatments once used routinely in severely ill patients on grounds that they appeared reasonable but which have subsequently been proved to be positively harmful. Although our studies showing reduction in pH, in response to bicarbonate in vitro are not conclusive evidence that such a deleterious effect occurs during bicarbonate therapy, we feel that the findings shift the burden of proof towards those that favour such treatment: perhaps the time is right to plan controlled trials of bicarbonate therapy despite the formidable difficulties that Cohen and Woods rightly identify. ’
J. M. RITTER Department of Clinical Pharmacology, Guy’s Hospital, London Bridge SE1 9RT
MICHAEL B. RODELL GARRETT E. BERGMAN
Armour Pharmaceutical Company, Blue Bell, Pennsylvania 19422, USA
H. S. DOKTOR N. BENJAMIN
RD, Woods HF. Disturbances of acid-base homeostasis. In. Weatherall DJ, Ledingham JGG, Warrell DA, eds. Oxford textbook of medicine, 2nd ed Oxford: Oxford University Press, 1987: 164-75.
1 Cohen
Factor VIII:C inhibitor associated with monoclonal-antibody-purified FVIII
CW, Shapiro SS, et al The natural history of factor VIII:C inhibitors in patients with hemophilia A: a national cooperative study, II: observations on the initial development of factor VIII:C inhibitors Blood 1988; 71: 344-48. 2. Lusher JM, Salzman PM, et al Viral safety and inhibitor development associated with factor VIII:C ultra-purified from plasma in hemophiliacs previously unexposed to factor VIII.C concentrates. Seminar Hematol 1990; 27 (suppl 2): 1-8 3. Nilsson IM, Bemtorp E, Zettervall O. Induction of immune tolerance in patients with hemophilia and antibodies to factor VIII by combined treatment with intravenous IgG, cyclophosphamide, and factor VIII N Engl JMed 1988; 318: 947-50 4. Hrinda ME, Feldman F, Schreiber AB. Preclinical characterization of a new pasteurized monoclonal antibody purified factor VIII.C. Seminar Hematol 1990; 1 McMillan
27
(suppl 2):
19-25.
Prevention of spontaneous diabetes in BB rats with FK 506 SIR,-Cyclosporin
can
mitigate
or
prevent autoimmune BB
rat
diabetes,’-3 and can induce remissions of human type I diabetes if instituted early after diagnosis and in adequate doses.4-6 The potent new immunosuppressive agent, FK 506, which has had extensive clinical trials in transplantation7 suppresses T-cell activation in the same general way as cyclosporin does, by inhibiting the transcription of mRNA for early phase lymphokines such as interleukin-2, interleukin-3, and y-interferon.8 We have examined the influence of daily enteral tap water (group 1, n = 20), 1 mg/kg FK 506 (group 2, n = 19), or 2 mg/kg FK 506 (group 3, n = 20) on the incidence of spontaneous diabetes in our BB rat colony. The 59 animals were obtained from seven consecutive litters, randomly deployed to the three sex-matched groups, raised under controlled lighting, and allowed tap water and regular rat food (Purina ’Chow 5001’) ad libitum. Treatment was between 30 and 120 days of age. The FK 506 (Fujisawa Pharmaceutical, Osaka) was dispersed with hydroxypropylmethylcellulose, diluted in tap water by sonication, and administered by gastric intubation. Onset of diabetes was diagnosed by glycosuria (tested thrice weekly) and confirmed by non-fasted blood sugar above 200 mg/dl from tail vein samples for 3 consecutive days.
concentrate
SIR,-Several comments need to be made to put in perspective the letter by Dr Kessler and Dr Sachse (June 9, p 1403) concerning factor VIII:C inhibitor formation. The actual rate of inhibitor formation in patients with severe haemophilia A, followed from childhood to adulthood, is unknown. The estimated rates of 5-15% are derived from cross-sectional rather than longitudinal studies and include patients with all degrees of severity. In one large "natural history" study, 29% of patients had inhibitor after 30 years of age, and at a time when only crude preparations of factor VIII:C were available.’ Most inhibitors developed in children. It has been estimated that if severely affected (factor VIII:C < 1 %) children were followed for 20 or 30 years then an inhibitor might develop in 30% or more (L. Aledort, personal
Onset and incidence of diabetes during FK 506 treatment.
