NEONATAL SEIZURES DUE TO MATERNAL PRIMARY HYPERPARATHYROIDISM

J. Paediatr. Child Health (2004) 40, 326–333

Letters to the Editor 2 February 2004 Dear Editor, DIGOXIN DOSAGE ERROR and PAEDIATRIC CARDIOLOGY HANDBOOK, 3RD EDITION I would like to bring notice the significant error concerning digoxin dosage that occurs in the above book on page 205 Table 7-4: ‘Oral Digoxin Dosage for Congestive Heart Failure’. The total daily dose and maintenance dose have been written in milligrams (mg) not micrograms (µg). I would urge all those who have access to this handbook to change the table. The publishers of this book have notified those that they can identify through sales records. I would concur with Dr Weintraub’s recent review of the book in this journal1 that there are some outdated management strategies such as the routine use of digoxin for heart failure and ‘digoxin is no longer preferred therapy for chronic management of supraventricular arrhythmias.’ The history of use of digoxin for management of congenital heart defects in infancy and childhood has been characterized by too many fatal overdoses already.

in the first place. Perhaps then the HREC’s decision was based on a rationale that if the parents/guardians were not aware of their child’s involvement in the study, then surely no-one else should be able to identify their child either – hence the de-identification. It seems to me that there is a simple way to avoid such dilemmas in the future. The parents/guardians of participants should be informed of such studies prior to commencement of the study. At this time they should be given three options. (i) To decline for their child’s blood spot to be used. (ii) To agree to use of the blood spot but not to de-identification, meaning that they will be informed of any results. (iii) To agree to use of the blood spot and to de-identification, meaning that results will not be reported back to them. Such a protocol is the only way to properly respect the autonomy of parents/guardians. In regard to the study by Barnes et al. I can’t imagine such a protocol would have rendered the time-frame ineffective. In regard to future research, not only would we respect people’s autonomy, but we would reap two additional benefits – being able to report back results to those who desire them and being able to help researchers avoid unethical positions. REFERENCES

REFERENCE 1

1

Weintraub R. Book Review: Paediatric Cardiology Handbook. J Paediatr. Child Health 2003; 39: 722–5.

Dr Nigel Wilson Paediatric Cardiologist Paediatric Cardiology Departments Starship and Green Lane Hospitals Auckland New Zealand

2

Williamson R. The balance between privacy, safety and community health. J. Paediatr. Child Health 2003; 39: 507–8. Barnes G, Srivastava A, Carlin J, Francis I. Delta-F508 cystic fibrosis mutations is not linked to intussesception: Implications for rotavirus vaccine. J. Paediatr. Child Health 2003; 39: 516–17.

Ms Rony Duncan PhD Candidate Ethics Unit Murdoch Childrens Research Institute University of Melbourne Departments of Paediatrics & Public Health Royal Children’s Hospital Melbourne, Victoria Australia

12 November 2003 Dear Editor, RESPECTING THE AUTONOMY OF PARTICIPANTS

28th November 2003 I write in response to Williamson’s editorial concerning the need for a balance between privacy, safety and community health.1 Williamson refers to the study by Barnes et al. published in the September 2003 issue of the Journal.2 According to Williamson, a Human Research Ethics Committee (HREC) put Barnes et al. in the ‘unethical position’ of not being able to inform people of their child’s CF carrier test results. This is because the HREC granted approval on condition of complete de-identification of samples. What Williamson fails to note is that the participant’s parents/guardians were not aware of their child’s involvement in the study (or the involvement of their child’s blood). Samples were obtained from participant’s Guthrie cards without their parent’s/guardian’s knowledge. The justification for this was that prospective recruitment would only have yielded a maximum rate of 20–25 participants per year, leading to a time frame that was then of no use to decisions needing to be made quickly. Thus, if the parents/guardians of participants had been informed of their child’s CF result, this would have entailed an additional explanation of why their child was tested

Dear Editor, PRIVACY, SAFETY and COMMUNITY HEALTH In the editorial ‘The balance between privacy, safety and community health’,1 Professor Williamson questions a decision of the Human Research Ethics Committee regarding a study which involved access to Guthrie cards.2 The project was designed to determine whether the delta-F508 cystic fibrosis (CF) gene might predispose to intussusception – a question of interest in relation to rotavirus vaccine development and implementation. He states that ‘the samples were anonymized to ensure privacy, a strategy encouraged by the Human Research Ethics Committee’, and later pointed out that we now cannot warn parents of CF gene carriers, of an increased risk of having a CF child next pregnancy ‘thanks to the decision of the Ethics Committee’. He refers to an obsession with privacy taking precedence over a duty to warn of a real risk to health. As authors of the article, we must indicate that the HREC only considered the proposal that we put forward. In our

Letters to the Editor

application, we proposed that all specimens should be deidentified, and asked the HREC to agree or disagree to that proposal. We did not ask the HREC to suggest an alternative protocol that would be acceptable, and the committee did not encourage any particular strategy. Our proposal was written after discussion with colleagues about various alternatives before it was submitted for ethical consideration. The options that we considered included: 1 Requesting permission from parents of children who had been admitted in the past with intussusception, and from parents of control children, to access their Guthrie cards and test for the gene abnormality. 2 Setting up a prospective study, and asking parents for permission, as children with intussusception were admitted, and identifying and approaching controls in the same way. 3 Accessing Guthrie cards without requesting parental permission, and ensuring that specimens were totally deidentified before testing. We did not have the capacity to pursue the first option, and if that course had been deemed to be necessary, the study would not have been performed. The second option would have taken several years to complete, based on the admission rate for intussusception in our hospital, and a multicentre study was not an option for the same reason as for option 1. The potential information anticipated from the study needed to be available quickly, if it was to be of value in development of new vaccines that were already reaching phase III clinical trial stage. So in our view, option 3 was the only practical approach. We felt it was an ethically sound proposal, since there did not appear to be potential for harm to any of the children or parents whose Guthrie cards were used, while the study provided the potential to answer a question of importance to public health. Hence, option 3 was the only one put forward in the application, and thus the only one that the HREC was in a position to consider. Professor Williamson does raise issues that deserve open debate. We wonder whether he is suggesting that it would be ethical to conduct the study as we did, without seeking consent, and then to inform parents of the results of a genetic test that they had not requested, and of which they had no knowledge. There were differing opinions, and different shades of opinion amongst those we consulted before presenting our application to the HREC, but no-one proposed that the study should perform tests on identified Guthrie cards without consent. They believed that if consent was not to be obtained, then the specimens should not be identifiable. The whole question of access to Guthrie cards for research purposes is contentious. We hope that the editorial will generate a lively and informative debate within the correspondence columns of The Journal. REFERENCES 1 2

Williamson R. The balance between privacy, safety and community health. J. Paediatr. Child Health 2003; 39: 507–8. Barnes G, Srivastava A, Carlin J, Francis I. Delta F508 cystic fibrosis mutation is not linked to intussusception: Implications for rotavirus vaccine. J. Paediatr. Child Health 39: 516–17.

Professor Graeme Barnes Dr Anshu Srivastava Professor John Carlin Mr Ivan Francis Murdoch Childrens Research Institute Department of Paediatrics, University of Melbourne Genetic Health Services Victoria Royal Children’s Hospital Melbourne, Victoria, Australia

327

18 December 2003 Dear Editor, DEFENDING DE-IDENTIFICATION OF RESEARCH SAMPLES ON THE GROUNDS OF PUBLIC HEALTH BENEFIT There is another way in which we might achieve a balance between privacy, safety and community health than that argued for by B. Williamson in the September editorial.1 Commenting on a study reported by Barnes et al. in the same issue of the journal,2 Williamson argues that an ‘obsession with “privacy” has taken precedence over a duty to warn of a real risk to health’.1 The context is a clinical trial in which researchers were given ethics committee approval to access Guthrie cards to test whether a mutation in the gene for cystic fibrosis predisposes to intussusception. A guarantee that the samples were deidentified (anonymous and not re-identifiable) was required by the ethics committee. The study was important and potentially valuable. If the results supported the predisposition hypothesis, rotavirus vaccine could be delivered to those children who would benefit (the majority) and withheld from those predisposed to intussusception.3 The vaccine was withdrawn in 1999 because of a rare but serious association with intussusception. Unfortunately the study was not able to identify those with the predisposition to intussusception. Williamson’s concern is that in the course of the study five infants were found to be CF carriers. He argues that ‘concerns over theoretical or philosophical risks to privacy’ mean that we don’t know who these babies are, and we cannot warn their parents who have a 16-fold increased risk of having a child with CF next pregnancy. It is regrettable that parents who want this information will not be warned of their increased risk but perhaps we should be advocating antenatal screening for CF carrier status in the parents rather than relying on screening in newborns or chance findings in a research project. I will argue that we can defend the de-identification of samples in the study by Barnes et al. on the grounds of public health benefit. As noted by Williamson in his editorial and elsewhere, blood taken from the heel of newborn infants to test for PKU and other genetic conditions including CF is ‘an essential public health screening program’4 and the Guthrie card collection has great value as a research tool allowing targeted research – possible because the cards contain identifying information.3 The cards also assist quality control. In their deliberations about the de-identification of samples in the study, it is appropriate that researchers and ethics committees factor in what could be a threat to a significant public health benefit. Both the screening programme and valuable research are threatened by public fears about third parties gaining access to private information, including what has been referred to as the ‘untapped genetic secrets of the babies’ close relatives’.3 Protection of privacy (i.e. the de-identification of research samples) has a role in that it can allay any public concerns about the newborn screening programme and the use of Guthrie cards in research. Whether the risk to privacy is real or theoretical is not clear but concerns about privacy can be real in the sense that they impact on practice. Police access to samples in Western Australia has prompted some parents to request the return of their children’s cards and is being blamed for a decrease in participation in the screening programme.5 Western Australia now has a requirement that the cards be destroyed after 2 years.3 In South Australia, there has been a call for the samples to be destroyed so they won’t be used for criminal or

328

Letters to the Editor

insurance investigations.5 An Inquiry by the Australian Law Reform Commission has heard reports of parents refusing newborn screening tests because of concerns about the storage of samples.3 If de-identification of samples in research helps promote public confidence in research and the newborn screening programme, surely that is the way to go. ACKNOWLEDGEMENT Thank you to Martin Delatycki for comments on this letter. REFERENCES 1 2 3

4 5

Williamson B. The balance between privacy, safety and community health. J. Paediatr. Child Health 2003; 39: 507–8. Barnes G, Srivastava A, Carlin J, Francis I. Delta-F508 cystic fibrosis mutation is not linked to intussusception: Implications for rotavirus vaccine. J. Paediatr. Child Health 2003; 39: 516–17. Australian Law Reform Commission (ALRC 96). Essentially Yours: The Protection of Human Genetic Information in Australia. 2003; 19.25 [Referring to B. Williamson, Consultation, 5 September 2001]. Ryan K, Kelly J, Hodgson S. Alarm over DNA secrets. The Herald Sun 2003; 27 Jun (1, 4). Call to destroy state’s 400 000 blood samples. Sunday Mail 2003; 29 Jun.

Dr Merle Spriggs Post doctoral researcher Ethics Unit, Murdoch Children’s Research Institute Centre for the Study of Health and Society University of Melbourne Centre for Human Bioethics Monash University Melbourne, Victoria Australia

A newborn screening programme is offered to all babies born with parental consent. The clear benefits to the early diagnosis of treatable and potentially harmful conditions have already been established. The blood spots retained on the so-called Guthrie cards were originally kept for quality assurance. There are enormous benefits from potential additional testing under very specific and restricted circumstances. Retrospective diagnosis in a deceased child or with the availability of newer tests may be performed, with consent, for the benefit of the individual or their family. Research may have potential benefits for the greater good, but it is our responsibility as guardians of this resource to keep the cards secure. Only a few research projects have been approved. Identified samples have only been tested with consent. The onus of research study design is on the researcher. The hospital ethics committee has the obligation of ensuring that NHMRC Guidelines relating to research on human subjects are upheld. If specific consent is not obtained, samples are deidentified as a matter of respect for personal autonomy rather than privacy. In such circumstances there is always the potential of discovering information that cannot be passed on to the individual. If such information were to constitute potential serious harm – and CF carrier status does not fall in that category – then the study design is unethical and should not be carried out without consent. All people in the community are unknowingly carriers of some recessive genetic conditions. One in 25 of the normal population is a carrier of a mutation in CFTR, the gene for cystic fibrosis. Whilst we may and in fact should offer screening for common mutations it is contrary to accepted practice to test without consent.

REFERENCES 1 2

Williamson B. The balance between privacy, safety and community health. J. Paediatr. Child Health 2003; 39: 507–8. Australian Law Reform Commission (ALRC 96). Essentially Yours: The Protection of Human Genetic Information in Australia, 2003.

13 February 2004

Associate Professor Agnes Bankier Director Genetic Health Services Victoria Royal Children’s Hospital Melbourne, Victoria Australia

Dear Editor, LETTER OF RESPONSE TO EDITORIAL BY R WILLIAMSON (SEPTEMBER 2003) In response to R Williamson’s editorial ‘The balance between privacy, safety and community health1’ the clinical perspective on provision of genetic health services raises two critical issues. These relate to respect for autonomy rather than privacy and to study design rather than limitations placed by hospital ethics committees. Medical investigations and specifically genetic studies are only performed in clinical practice with informed consent. This imperative applies even when people have a known high risk of a genetic disorder based on their family history. Genetic advice regarding risk is only provided on request. The ALRC in their report ‘Essentially Yours’2 have recommended that information about risk can only be divulged without consent by health professionals when there is serious and imminent harm to an individual. It recommends furthermore that genetic testing performed without specific consent should be a new criminal offence (other than that performed by court order). Genetic Health Services Victoria provides genetic diagnosis, counselling and support to individuals who seek such services.

Dear Editor, REPLY I am pleased that my editorial has provoked several people into sending letters with other views. This is an important area for discussion where all views should be heard and taken into account. Professor Bob Williamson Director The Murdoch Children’s Research Institute Royal Children’s Hospital Melbourne, Victoria Australia

Letters to the Editor

329

12 September 2003 Dear Editor, NEONATAL SEIZURES DUE TO MATERNAL PRIMARY HYPERPARATHYROIDISM We would like to report a rare case of asymptomatic maternal hyperparathyroidism due to a parathyroid adenoma presenting with recurrent neonatal hypocalcaemic seizures in her offspring. Following an uneventful pregnancy, a 36-year-old gravida 3 para 2 Chinese shop assistant delivered a 3000-g male infant at term gestation via elective lower segment Caesarian section. He was discharged on the fourth day of life and remained well until the tenth day of life when he developed multiple brief episodes of jerky movements of his right upper and lower limbs, and twitching of his eyelids. In between these episodes he was well. On the fifteenth day of life, his mother brought him to the hospital as the episodic seizures were becoming more frequent at about 20 times a day. There was no associated fever, diarrhoea or vomiting. He was the third child from a non-consanguineous marriage. There was no family history of seizures. Physical examination revealed a well infant weighing 3880 g, and no abnormalities were detected. His vital signs were normal. There was no dysmorphic features or neurocutaneous stigma. Culture of his blood, urine and cerebrospinal fluid specimens were negative. His serum electrolytes and random blood sugar were all within normal limits. His corrected serum total and ionized calcium levels were 1.12 and 0.4 mmol/L, respectively. His serum magnesium level was 0.56 mmol/L, and serum parathyroid hormone level was 0.916 pmol/L (reference range: 1.3–7.6). He was commenced on intravenous and later oral calcium supplement therapy. His seizures ceased when his serum calcium level rose above 1.4 mmol/L. His serum calcium level became normal when he was treated with alpha-calcidol at 0.3 mg per day and calcium lactate at 300 mg four times a day. Supplemental therapy with calcium and vitamin D was reduced from the age of 2 months and stopped altogether when he was 3 months old. His serum calcium level remained normal subsequently. Laboratory investigation revealed that his mother had high serum calcium (3.11 mmol/L), low serum phosphate (0.59 mmol/L) and high serum parathyroid hormone (239.5 pmol/L). Ultrasonography showed an enlarged parathyroid gland behind the middle and lower pole of the right lobe of her thyroid gland. Her abdominal ultrasonogram revealed the presence of gallbladder calculi and bilateral medullary nephrocalcinosis. At surgery, she was found to have an enlarged right inferior parathyroid gland measuring 25 × 12 mm and weighing 1.5 g, which was excised. Histopathological examination of this specimen confirmed the presence of a parathyroid adenoma. Post-operatively she developed symptomatic hypocalcaemia for which she was treated with oral calcium supplement. Review of medical literature showed that maternal primary hyperparathyroidism during pregnancy is rare with an estimated incidence of eight per 100 000 population per year.1–6 A majority of these women are asymptomatic or have subtle symptoms due to physiological changes of calcium metabolism during pregnancy. Morbidity in both mother and her infant, as shown in our patients, was high. In mothers who were not treated, the incidence of foetal complications was reported to be as high as 80%; these included intrauterine growth restriction, low birthweight, preterm delivery, intrauterine death and

neonatal hypocalcaemic tetany and seizures.2 The neonatal hypocalcaemia results from intrauterine suppression of foetal parathyroid function by the maternal hypercalcaemia, sudden deprivation of a rich source of calcium from its mother after birth, and the neonate’s incapability to mobilize calcium from its bones due to its low serum level of parathyroid hormone and high serum level of calcitonin. Most infants became symptomatic at the end of first week of life with seizure, as was found in our patient,3,4 requiring short-term calcium supplemental therapy of 3 to 5 month duration until the hypoparathyroidism resolves. Our case illustrates that when newborns present with hypocalcaemic seizures towards the end of the first week of life or later, appropriate investigations should be carried out to exclude maternal hyperparathyroidism. REFERENCES 1 2 3 4 5 6

Heath H, Hodgson S, Kennedy M. Primary hyperparathyroidism: incidence of morbidity and potential economic impact on a community. N. Engl. J. Med. 1980; 302: 189–93. Delmonico FL, Neer RM, Cosimi AB. Hyperparathyroidism during pregnancy. Am. J. Surg. 1976; 131: 328–37. Behrman RE, Kleigman RM, Jenson HB. Nelson Textbook of Pediatrics, 16th edn. Philadelphia: Saunders, 2000. Beattie GC, Ravi NR, Lewis M et al. Rare presentation of maternal primary hypeparathyroidism. BMJ 2000; 312: 223–4. Brisse F, Breton D, Gagey V, Cheron G. Convulsions and neonatal hypoparathyroidism revealing maternal hyperparathyroidism. Arch. Paediatr. 1994; 3: 225–9. Brock JB, Terslev E, Lund B, Helmer SO. Neonatal hypocalcaemia associated with maternal hyperparathyroidism. Arch. Dis. Child 1978; 53: 308–11.

Dr Rohana Jaafar Professor Nem Yun Boo Associate Professor Rahmah Rasat Dr Hasniah Abdul Latiff Department of Paediatrics Faculty of Medicine Universiti Kebangsaan Malaysia Kuala Lumpur Malaysia

12 November 2003 Dear Editor, SAFETY OF RAPID INTRAVENOUS REHYDRATION It was with some concern that I read the recent article by Phin et al. published in your journal describing a practice of using rapid intravenous rehydration with hypotonic saline for children with mild or moderate dehydration caused by gastroenteritis.1 For children with mild dehydration (as defined) there was clearly little benefit in instituting the proposed clinical pathway as there was no statistically significant reduction in the admission rate for these children, and there was a statistically significant increase in the number of procedures that these children underwent. My major concern however was in relation to the management of children with moderate dehydration. For these children who fail a trial of oral rehydration the authors advocate the use of either a trial of fluids via a nasogastric tube or the administration of 20 mL/kg per hour of N/2 saline with 2.5% dextrose

330 for 2 h. There is a discrepancy in the total number of children who had such intervention between Table 5 (107 children) and Fig. 1(106 children) in the article and it is not stated clearly how many of the children with moderate dehydration received the intravenous rapid rehydration regimen. It is the management of this group that causes most concern. A total of 83/145 children (with both mild and moderate dehydration) are stated to have received the intravenous fluid regimen. Blood was not taken for electrolytes on all of these children and when taken the authors did not wait until results were available prior to commencing the rapid intravenous rehydration regimen. In the group as a whole (those with oral as well as nasogastric and intravenous rehydration) electrolyte results were subsequently available for 78/145 children. Hyponatraemia was present in 22% of these samples and two patients with moderate dehydration had sodium levels less than 130 mmol/L. The authors do comment on their concerns over the safety of using a rapid infusion of a total of 40 mL/kg of hypotonic saline in children who may be hyponatraemic but stated that in the two children with significant hyponatraemia no adverse effects were noted. They also comment that as a safety measure children were reassessed after the rapid rehydration regimen by a doctor and if the child appeared unwell at that point, electrolytes were measured if this had not already been done. The authors quote one previous study as support for their rapid rehydration regimen with hypotonic saline.2 This was a study of only 17 patients and specifically excluded children with abnormal sodium levels. The safety of such a regimen is therefore untested. I would suggest that the use of such volumes of hypotonic saline in children with moderate dehydration in whom sodium derangements are common is potentially dangerous. I have personal experience of a 7-year-old child who had a fatal outcome following the administration of 20 mL/kg of hypotonic saline. This child, who was unwell for less than 48 h with suspected gastroenteritis, presented with features of moderate dehydration and failed a trial of oral rehydration. An infusion of 20 mL/kg of N/5 saline with 4% dextrose was administered over a period of 2 h and after further review the intravenous infusion was continued at a rate calculated to correct for 5% dehydration over a period of 24 h. Initial sodium level prior to commencing intravenous rehydration was 130 mmol/L. Forty-five minutes after the rapid infusion regimen had finished the child became increasingly agitated and confused and 1.5 h after the rapid infusion regimen had ended she rapidly developed clinical features of brainstem herniation, suffering a sudden respiratory arrest with fixed dilated pupils. Electrolytes at this time revealed a sodium level of 120 mmol/L. She was resuscitated and transferred to the Paediatric Intensive Care unit but pronounced brain dead. Autopsy revealed a grossly oedematous brain. Results of metabolic and microbiological studies revealed no other abnormality and cause of death was determined to be due to acute cerebral oedema and brain stem herniation as a consequence of rapid fall in sodium levels related to rapid rehydration with hypotonic saline. There have been many cases of fatal iatrogenic hyponatraemia reported in recent literature both in children with gastroenteritis and in children treated with intravenous fluids whilst undergoing minor surgery.3,4 Death or permanent neurological damage as a result of the rapid administration of hypotonic fluids has been recognized to occur over a matter of hours and may occur in previously healthy children at volumes less than 40 mL/kg and even at ‘maintenance’ rates of fluid administration. In ill children there are many factors that lead to disruption in the usual mechanisms for control of body water and

Letters to the Editor sodium levels. Many authors now recommend banning the use of N/5 saline completely and avoiding the use of other hypotonic saline solutions when sodium levels are less than 138 mmol/L.3,5,6 Certainly the use of any saline solution other than an isotonic solution for the purpose of rapid infusion or bolus makes no physiological sense and would be difficult to defend in the event of causing symptomatic hyponatraemia. Could the authors please reconsider their approach to the rapid rehydration of children with moderate dehydration to include the measurement of electrolytes prior to commencing a rapid intravenous infusion. In addition, further study is required to determine the safest saline solution and the safest infusion rate to be used in treating children with moderate dehydration caused by gastroenteritis. REFERENCES 1 2 3 4 5

6

Phin SJ, McCaskill ME, Browne GJ, Lam LT. Clinical pathway using rapid rehydration for children with gastroenteritis. J. Paediatr. Child Health 2003; 39: 343–8. Moineau G, Newman J. Rapid intravenous rehydration in the pediatric emergency department. Pediatr. Emerg. Care 1990; 6: 186–8. Playfor S. Fatal iatrogenic hyponatraemia. Arch. Dis. Child 2003; 88: 646–7. Arieff AI, Ayus JC, Fraser CL. Hyponatraemia and death or permanent brain damage in healthy children. BMJ 1992; 304: 1218–22. Halberthal M, Halperin ML, Bohn D. Lesson of the week: acute hyponatraemia in children admitted to hospital: retrospective analysis of the factors contributing to its development and resolution. BMJ 2001; 322: 780–2. Bohn D. Problems associated with intravenous fluid administration in children: do we have the right solutions? Curr. Opin. Pediatr. 2000; 12: 217–21.

Dr Catherine Riney Clinical Research Fellow Neurosciences and Biophysics Unit Institute of Child Health (UCL) and Great Ormond Street Hospital for Children NHS Trust London, United Kingdom

9 December 2003 Dear Editor, REPLY We appreciate the comments of CJ Riney and the opportunity to reply. In our article we stressed that this pathway was effective for children who were moderately dehydrated from their gastroenteritis, resulting in a 42% reduction in admission rates and a 12-fold increase in the number who were discharged from the emergency department in 8 h or less. We reported the increase in procedure rate in the children who were only mildly dehydrated and suggested that the pathway needed to be modified and subsequently evaluated for these children. We have since instituted a number of measures in our Emergency Department in an attempt to improve the success rate of oral rehydration therapy in these children and are currently evaluating their effect on admission rates and procedure rates, hoping to see a fall in both.

Letters to the Editor

331

The discrepancy mentioned in the numbers of patients receiving rapid rehydration between Fig. 1 and Table 5 is a typing error in Fig. 1 – 84 (not 83) children received rapid IV and 23 rapid NG rehydration making the total 107. Subsequent to the study, much discussion has been generated about the type of fluid and the rate of administration when intravenously rehydrating children with gastroenteritis. The choice of fluid regime is governed by the underlying aetiology and pathophysiology. We note that the fluid chosen in CJ Riney’s unfortunate case was N/5 + 4% dextrose, which contains 2.5 times less sodium and hence more free water than the N/2 saline + 2.5% dextrose, which we routinely use to rehydrate our children with gastroenteritis in Australia. Nevertheless, opinion seems to be changing as to the ideal amount of sodium in intravenous rehydration fluids, and a higher than expected incidence of mild hyponatraemia at presentation has been noted in our study population. As a result, we are currently studying the safety and efficacy of Hartmann’s solution in rapidly rehydrating children with gastroenteritis. As recommended in our article, close monitoring and timely reassessment of these children are, of course, an integral part of their management. Dr Susan J Phin Paediatric Emergency Physician Emergency Department The Children’s Hospital at Westmead Westmead, New South Wales Australia

(T score = 67), whereas patient A reported moderately elevated social anxiety (T score = 64). Although parent ratings of patient A did not indicate any clinical problems, patient B’s mother reported evidence of significantly elevated externalizing and internalizing problems (T scores = 69 and 67). Regarding the discordance of the ratings of patient A and his mother, it is relevant to note that it is not uncommon for parents to be unaware of internalizing symptoms in their children.9 Our data are noteworthy in that they suggest that some boys with gynecomastia may experience elevated levels of psychopathology. As such, it is imperative that clinicians working with these children should be trained regarding identifying psychological distress. On this end, conducting a rapid assessment of adjustment in clinical settings seems warranted. One brief and cost-effective manner of accomplishing this would be to include measures of psychological functioning as a standard component of the clinical visit. Child and parent report questionnaires, such as those used in this report are sound examples of possible assessment measures. In addition, although some recommend supportive reassurance to assist the youth in coping with their condition,1 these cases provide preliminary evidence that more focused clinical intervention may be necessary for some. However, further study in larger samples is needed to document if rates of psychosocial maladjustment in this population differ from normative samples. Author Note: This study was supported by grants awarded to Dr Storch from the Human Growth Foundation and University of Florida Center for Pediatric and Family Psychology.

REFERENCES 11 December 2003 Dear Editor,

1 2

PSYCHOSOCIAL ADJUSTMENT OF TWO BOYS WITH GYNECOMASTIA Gynecomastia is a benign condition in males, characterized by proliferation of glandular elements resulting in concentric enlargement of one or both breasts. Although this condition typically resolves by age 18 years when adult androgen/estrogen ratios are achieved, it has been documented that boys experience significant psychosocial distress related to their perceptions about their appearance and internalization of others’ negative reactions.1,2 Indeed, the extant literature examining the psychosocial adjustment of boys with gynecomastia is quite sparse, consisting of single case reports that do not incorporate psychometrically sound measures of adjustment.3,4 As such, we report quantitative data on the psychosocial adjustment of two boys seen consecutively in our Pediatric Endocrinology service. Patient A was 15-years-old, of Caucasian ethnicity, and resided with both parents. Patient B was 17-years-old, of Caucasian ethnicity, and resided with both parents and a younger brother. During their regularly scheduled clinic appointment, both boys completed the following well-validated self-report questionnaires in their clinic room: Children’s Depression Inventory,5 Social Anxiety Scale for Children – Revised,6 and Asher Loneliness Scale. Their mothers completed the Child Behavior Checklist,8 a parent-rated index that provides ratings of child externalizing and internalizing symptoms. As compared to same age peers, both boys reported significantly elevated levels of depressive symptomatology (T scores = 88 and 85). Patient B reported elevated feelings of loneliness

3 4 5 6 7 8 9

Lazala C, Saenger P. Pubertal gynecomastia. J. Pediatr. Encocrinol. Metab. 2002; 15: 553–60. Money J, Clopper RR. Psychosocial and psychosexual aspects of errors of pubertal onset and development. Hum. Bio. 1974; 46: 173–81. Fisher M, Fornari V. Gynecomastia as a precipitant of eating disorders in adolescent males. Int. J. Eat Dis. 1990; 9: 115–19. Mandoki MW, Sumner GS. Klinefelter syndrome: the need for early identification and treatment. Clin. Pediatr. 1991; 30: 161–4. Kovacs M. The Children’s Depression Inventory. Toronto, Canada: Multi-Health Systems, Inc. 1992. LaGreca AM, Stone WL. Social Anxiety Scale for Children– Revised: Factor structure and concurrent validity. J. Clin. Child Psychol. 1993; 22: 17–27. Asher SR, Wheeler VA. Children’s loneliness: a comparison of rejected and neglected peer status. J. Consult. Clin. Psychol. 1985; 53: 500–5. Achenbach TM. Manual for the Child Behavior Checklist and 1991 Profile. Burlington, VT: University of Vermont, Department of Psychiatry, 1991. Engel NA, Rodrigue JR, Geffken GR. Parent child agreement on ratings of anxiety in children. Psychol. Report 1994; 75: 1251–60.

Dr Eric A Storch1,2 Mr Adam B Lewin1 Dr Gary R Geffken1 Dr Amanda D Heidgerken1 Ms Melissa S Strawser1 Ms Audrey Baumeister1 Dr Janet H Silverstein2 1Department of Psychiatry and 2Division of Pediatric Endocrinology University of Florida Gainesville, Florida USA

332

Letters to the Editor 14 November 2003

30 January 2004 Dear Editor, THE IMPACT OF INLUENZA A ON CHILDREN WITH DISABILITIES The influenza A (H3N2) outbreak during 2003 had a significant impact on children with disabilities. In the week ending 25 August 2003, peak influenza-like illness (ILI) consultation rates in Victorian sentinel general practices reached 28.6 ILI per 1000 consultations, and over 80 ILI per 100 Melbourne Medical Locum Service locum consultations.1 During the month of August 2003 alone, children with influenza detected on nasopharyngeal aspirate accounted for seven of the 26 (26%) admissions under the Department of Child Development and Rehabilitation at the Royal Children’s Hospital, Melbourne. Their average age was 15 years, and all but one were nonambulant. The median length of stay was 12 days, all but one of these patients required ventilatory support in the Intensive Care Unit and one child died. The cumulative length of stay in ICU was 20 days, and total hospitalization 186 days. Only three of seven children had received influenza vaccine in the preceding Autumn months. By contrast, influenza A accounted for 13 of 68 admissions (19%) under a General Medical Unit during the same time period. The profile was quite different; the children had an average age of 4 years, their median length of stay was 3 days; two only required ventilatory support (15%) and one died as a result of hypoxia secondary to respiratory arrest. Influenza A is a serious disease, with significant morbidity and mortality. The care of these patients demands substantial input from nursing, physiotherapy and medical staff. This small analysis of the patients admitted in August 2003, however, would confirm that the impact for children with disabilities is far greater. Inactivated influenza vaccine is well tolerated in children and efficacious in preventing influenza A.2 Children over 6 months of age with chronic medical conditions requiring regular follow-up or hospitalization are a high risk-group recommended to receive annual influenza vaccination.3 Parents should be encouraged to seek influenza immunization for children with disabilities in an effort to reduce this increased burden for children, their families and health care resources. Paediatricians caring for this vulnerable group of children should also be aware of their high risk of serious illness and recommend immunization.

Dear Editor, RE: PARENTERAL NUTRITION SOLUTION RETRIEVED BY LUMBAR PUNCTURE FOLLOWING LEFT SAPHENOUS VEIN CATHETERIZATION ‘It happens on the right also, and it need not be lifethreatening’ Clarke et al.1 describe two instances of life-threatening complication of placement of a peripherally inserted central catheter (PICC) with its tip in the left ascending lumbar vein. Of 2186 silicone PICCs inserted at the Royal Brisbane and Women’s Hospital (RBWH) between 1 January 1984 and 31 December 2002, 160 were inserted via a lower limb vein. The catheter tip was in the ascending lumbar vein on two occasions only, once on the right and once on the left, and I submit that it need not lead to any complication. The simple solution is to inject all catheters with radio-opaque contrast medium while taking a radiograph to determine catheter tip position. As the Fig. 1 below illustrates, catheter placement in the ascending lumbar vein (the right side in this case) is then easily identified. It can then be pulled back to the common or external iliac vein. The pattern of dye in the paravertebral veins is distinctive, as shown. In this case the catheter is also curled on itself in the ascending lumbar vein on the right. At the Royal Brisbane and Women’s Hospital it is, and has always been, our practice to contrast-inject all PICCs, continuing injection during radiographic exposure. This readily reveals malpositioning of catheters such as those in the anterior abdominal wall,2 peritoneal cavity,3 aorta, vasa vasorum, butting up against the wall of the right atrium or ventricle, and many others, and allows the catheter to be repositioned before lifethreatening complications occur. Our practice for insertion and management of PICCs is to follow these principles:4 1 Insert by experienced staff (consultant or senior registrar) 2 Never leave a catheter where it does not easily and repeatedly withdraw blood. Aim to insert too far and pull back 3 ALWAYS inject with radio-opaque contrast during X-ray (If you don’t inject it, you don’t know where the tip is) 4 Be actively injecting during X-ray to see contrast coming from the end of the catheter (many positions will not be identified if the catheter is simply injected and later X-rayed as a static X-ray).

REFERENCES 1 2

3

Turner J, Kelly H. The Melbourne Medical Locum Service and surveillance for influenza-like illness. Victorian Infectious Dis. Bull. 2003; 6: 54–6. Neuzil KM, Dupont WD, Wright PF, Edwards KM. Efficacy of inactivated and cold-adapted vaccines against influenza A infection, 1985–90: the pediatric experience. Pediatr. Infect. Dis. J. 2001; 20: 733–40. NH, MRC. The Australian Immunization Handbook, 8th edn. Canberra: NH & MRC, 2003.

Dr Zornitza L Stark1 Dr James P Buttery2,3,4 Dr Giuliana C Antolovich1,3 Associate Professor Dinah S Reddihough1,3,4 1 Department of Child Development and Rehabilitation 2Department of Paediatrics Royal Children’s Hospital 3Murdoch Children’s Research Institute 4University of Melbourne Melbourne, Victoria Australia

Fig. 1 Radio-opaque contrast injection of a peripherally inserted central catheter (PICC) with its tip in the right ascending lumbar vein.

Letters to the Editor

333

There are ample examples of serious complications from PICCs resulting from not really knowing where the tip is, and descriptions of the superiority of contrast injection for determining tip position.5 The message is clear, we cannot continue to ignore it. Inject the catheter. REFERENCES 1

2 3 4 5

Clarke P, Wadhawan R, Smyth J, Emmerson AJ. Parenteral nutrition solution retrieved by lumbar puncture following left saphenous vein catheterization. J. Paediatr. Child Health 2003; 39: 386–9. Baker J, Imong SA. Rare complication of neonatal central venous access. (Case Report). Arch. Dis. Child Fetal Neonatal Ed. 2002; 86: F61. Mehta NM, Nicholl RM. Positioning of Long Lines. Arch. Dis. Child Fetal Neonatal Ed. 2002; 86: F68–9. Cartwright DW. Placement of neonatal central venous catheter tips: is the right atrium so dangerous? Arch. Dis. Child Fetal Neonatal Ed. 2002; 87: F155. Reece A, Ubhi T, Craig AR et al. Positioning of long lines: contrast versus plain radiography. Arch. Dis. Child Fetal Neonatal Ed. 2001; 84: F129–30.

Dr David W Cartwright Director of Neonatology Royal Brisbane and Women’s Hospital Brisbane, Queensland Australia

12 December 2003 Dear Editor,

and appearance of symptoms was 9 days. It would appear that catheters malpositioned in the ALV are only likely to lead to serious complications if left in situ for a sufficient duration of time. Clearly, the complication will not arise when: (i) catheter malposition is identified at time of insertion and promptly remedied; (ii) a PICC unrecognized to be malpositioned is fortuitously removed due to redundancy, within the lag phase that precedes symptoms. Routine injection with radio-opaque contrast medium following PICC placement would likely avoid the complication we reported. Simply knowing where a catheter tip lies at the time of insertion is in itself insufficient. There are many reports of life-threatening complications due to migration of catheters.3 Some instances of supposed PICC migration were probably due to unrecognized malposition from the outset.4 In other cases a PICC tip has wandered from a satisfactory initial position to a dangerous one. Limb movements and posture changes have been shown to cause significant displacement of PICCs.3,5 A catheter pulled back from the ALV to the common or external iliac vein thus has the potential to migrate back into the ALV, especially if there is retrograde flow (e.g. in association with inferior vena caval thrombosis). The safest solution is not just to routinely inject the PICC at insertion, but also to retain a high index of suspicion for as long as any PICC is indwelling. One of our units now performs routine X-rays twice weekly in catheterized infants specifically to recheck PICC position. Ultrasound of the vessels for catheter position may be a less invasive option. The only truly safe position for a PICC is ex vivo.

REFERENCES 1

REPLY We welcome the comments of Dr Cartwright following our article. We were interested to read of his unit’s considerable experience with peripherally inserted central catheters (PICCs), albeit surprised that only 7% of his unit’s catheters had been inserted via a lower limb vein. The rate of lower limb PICC lodgement in an ascending lumbar vein (ALV) that he reports (2/160 = 1.3%) is not significantly different from that found by Zenker et al.1 (3/44 = 6.8%), P = 0.07. A pooled estimate using these figures suggests that one in every 40 lower limb-inserted PICC may end up in an aberrant paraspinal position in the ALV. We thank Dr Cartwright for the report of a further case of right ALV malposition, with its instructive illustration. Malposition in the ALV may indeed occur on the right side,1 although with a much lower frequency: we discussed this in our report.2 To date however, extravasation of parenteral nutrition solution into the epidural space has only ever been reported in association with PICC lodgement in the left ALV. We agree that paraspinal malposition need not be lifethreatening. In the cases we reported and reviewed with the complication, the median time interval between PICC insertion

2

3 4 5

Zenker M, Rupprecht T, Hofbeck M, Schmiedl N, Vetter V, Ries M. Paravertebral and intraspinal malposition of transfemoral central venous catheters in newborns. J. Pediatr. 2000; 136: 837–40. Clarke P, Wadhawan R, Smyth J, Emmerson AJ. Parenteral nutrition solution retrieved by lumbar puncture following left saphenous vein catheterization. J. Paediatr. Child Health 2003; 39: 386–9. Nadroo AM, Glass RB, Lin J, Green RS, Holzman IR. Changes in upper extremity position cause migration of peripherally inserted central catheters in neonates. Pediatrics 2002; 110: 131–6. Odaibo F, Fajardo CA, Cronin C. Recovery of intralipid from lumbar puncture after migration of saphenous vein catheter. Arch. Dis. Child 1992; 67: 1201–3. Ohki Y, Nako Y, Morikawa A, Maruyama K, Koizumi T. Percutaneous central venous catheterization via the great saphenous vein in neonates. Acta Paediatr. Jpn. 1997; 39: 312–16.

Dr Paul Clarke1 Dr Rajan Wadhawan2 Dr John Smyth3 Dr Anthony J Emmerson4 1 Neonatal Unit, Hope Hospital, Salford, UK 2Women & Infants’ Hospital, Providence, RI, USA 3Neonatal Unit, The Royal Sussex County Hospital Brighton, UK 4Neonatal Medical Unit, St Mary’s Hospital, Manchester, UK