characteristic external features as well as life- ... short life expectancy, and all long-term survivors ... each had at least one life-threatening malformation (three.
913
CLINICAL PRACTICE Avoidance of emergency surgery in newborn infants with trisomy 18
Trisomy 18 (Edwards’ syndrome) presents with characteristic external features
well as lifethreatening abnormalities; many of these abnormalities require surgical correction during the neonatal period. Children with trisomy 18 have a very short life expectancy, and all long-term survivors have severe mental retardation. Difficult medical and ethical issues arise over whether or not to institute treatment when a newborn infant with suspected trisomy 18 has a life-threatening anomaly. We studied the policy of treatment in seven patients with clinical Edwards’ syndrome. For three, the period of uncertainty was shortened because trisomy 18 was rapidly diagnosed by karyotyping of a bone-marrow aspirate. Four of the patients underwent surgery before the diagnosis of trisomy 18 was confirmed by routine karyotyping in lymphocytes; karyotyping in bone marrow might have allowed invasive treatment to be avoided in three of these. Rapid confirmation of clinically suspected Edwards’ syndrome is very important because surgery may then be withheld. A newborn infant with trisomy 18 should be considered as a patient with a hopeless outlook who ought not to be subjected to invasive procedures. The decision to withdraw or withhold treatment should be discussed frankly with the parents. The period of uncertainty can be reduced to a minimum by the use of karyotyping in bone marrow. as
Introduction Advances in intensive care and paediatric surgery have greatly improved the outlook for newborn infants with congenital anomalies. However, the availability of lifesaving treatments complicates the decision-making process for malformations that suggest a chromosomal abnormality, such as trisomy 18. This is the second commonest trisomy, with a frequency of about 1 in 3000 births. Life expectancy is limited; 30% of patients die within the first month of life, and only 10% survive beyond a year.1 There are long-term survivors, but all have profound mental retardation and restricted growth.2 The clinical syndrome includes characteristic external features as well as life-threatening malformations such as congenital diaphragmatic hernia, oesophageal atresia, or cardiac defects.3 What should be done when trisomy 18 is suspected in a newborn infant with a life-threatening anomaly?
Karyotyping in lymphocytes takes 3 days. We report our experience with seven trisomy-18 patients and the use of rapid karyotyping in bone marrow to prevent unwarranted surgery.
Patients Our tertiary referral hospital has a 13-bed paediatric surgical intensive-care unit. From 1986 to 1990, 408 newborn infants with structural congenital anomalies were
admitted to the unit within the first month of life. All underwent routine evaluation by a consultant geneticist. Chromosomal analysis, on peripheral blood lymphocytes, was done for 84 infants with several congenital anomalies. Seven patients had the karyotype 47, XX/XY, + 18; they each had at least one life-threatening malformation (three had oesophageal atresia, three congenital diaphragmatic hernia, and six cardiac defects). In patients 5-7 bonemarrow aspiration, which allows karyotyping within 6 h, was done in an attempt to avoid surgery on an affected child. Until the result of the karyotype was available, all patients received full support including artificial ventilation (six patients) and emergency surgery if indicated (four patients). The median time in the intensive-care unit was 4 days (range 3-14) and the median survival time 11days (3-131).
Case-reports Patient 1, a girl, was born at 42 weeks’ gestation, weighing 2010 g. examination showed Edwards’ syndrome (micrognathia, low-set ears, short palpebral fissures, hypoplastic nails, and clenched fists) and oesophageal atresia. A blood sample was taken for karyotyping. 36 h after birth, the stomach became distended and respiratory distress necessitating artificial ventilation developed. Although the result of chromosomal analysis was not available, we decided to carry out only a gastrostomy. The lymphocyte culture failed and conclusive karyotype analysis could not be done. This delay led to continuation of treatment, including artificial ventilation. Karyotyping of cells from a second blood sample confirmed the clinical diagnosis of Edwards’ syndrome on day 8. Support was gradually withdrawn under a regimen of sedation and analgesia and the baby died aged 11 days. Patient 2, a girl, was delivered by caesarean section at 35 weeks’ gestation (birthweight 1510 g) because of fetal distress. Persisting respiratory distress necessitated endotracheal intubation and artificial ventilation. On admission to intensive care, Edwards’ syndrome was suspected and blood was taken for karyotyping. The baby had congenital diaphragmatic hernia and oesophageal atresia, and echocardiography showed a large ventricular defect and atrial
Physical
ADDRESS Department of Paediatric Surgery, (A. P. Bos, MD, C. J M. Broers, MD, F W J. Hazebroek, MD, D. Tibboel, MD, Prof J. C. Molenaar, MD). and Department of Clinical Genetics (J. O van Hemel, MD, E Wesby-van Swaay, MD), Sophia Children’s Hospital, Erasmus University Medical School, PO Box 70029, NL-3000 LL, Rotterdam, the Netherlands. Correspondence to Dr A P Bos
914
septal defect. Progressive distension of the stomach and bowel loops, which resulted in further deterioration of oxygenation, meant that an emergency operation was required after 36 h. The diaphragmatic defect was repaired, the tracheo-oesophageal fistula was closed, and a gastrostomy was done. The baby died 2 days after the operation. 3 days later karyotype analysis in cultured lymphocytes showed trisomy 18. Patient 3, a girl, was born at 40 weeks’ gestation, weighing 2505 g. Karyotyping was done because of microphthalmia, blepharophimosis, "whistling face", hypoplastic nails, and a small omphalocele. She had a large ventricular septal defect and a pulmonary valve stenosis. After confirmation of trisomy 18, the parents agreed to treatment aimed at comfort rather than active support of vital functions. The baby was referred to the local hospital, where she had septic episodes, failure to thrive, and congestive heart failure and died aged 4 months. Patient 4-This boy was bom at 39 weeks’ gestation to a 47-year-old mother with diabetes mellitus. Prenatal ultrasonography showed hydronephrosis, hydroureters, and megacystis. Amniocentesis was refused by the parents. Postnatal physical examination showed microcephaly with a small mouth, low-set posteriorly rotated ears, micrognathia, short sternum, hypoplastic nails, and a prune belly. There was a large ventricular septal defect. Urological examination showed a urachus fistula and confirmed prenatal ultrasound findings. On day 2 of life the baby became ventilator-dependent and a bilateral ureterocutaneostomy was done. The result of chromosomal analysis was consistent with trisomy 18. The doctors and parents agreed to withdraw artificial ventilation and to withhold invasive diagnostic procedures and corrective surgery. The postnatal period was complicated by septic episodes and cardiac problems. 4 weeks later the baby died after an increasing number of apnoeic attacks. Patient 5-A girl bom at 38 weeks’ gestation, weighing 1730 g, required endotracheal intubation and artificial ventilation soon after birth. Rocker-bottom feet and clenched hands were noted; a double-outlet right ventricle and oesophageal atresia were diagnosed. The next morning a blood sample and a bone-marrow aspirate were taken for karyotyping. Further deterioration and progressive ventilatory problems due to gastric distension necessitated surgical intervention, only closure of the tracheooesophageal fistula was achieved. The same day karyotype analysis of the bone-marrow sample confirmed our clinical suspicion of trisomy 18. At first, the parents wanted full medical and surgical support for their child, but after extensive consultation with the doctors they agreed to allow the disorder to take its natural course. A cervical fistula and a gastrostomy were made and the baby was weaned from the ventilator. She was discharged to the local hospital, where she died 9 days later due to an increasing number of central apnoeic episodes in combination with congestive heart failure. Patient 6-This boy was born at 41 weeks’ gestation, weighing 2170 g, with a right-sided diaphragmatic hernia and a ventricular septal defect. He showed several anomalies characteristic of Edwards’ syndrome. Karyotyping of bone-marrow aspirate on the same day showed trisomy 18 in all cells. We discussed withdrawal of treatment with the parents and after brief hesitation they agreed. The boy died on day 9 of life after an increasing number of convulsions and central apnoeic attacks. 2 months later the parents expressed their satisfaction with the way he was managed. Patient 7 was a boy born after an uneventful pregnancy of 41 weeks. Prenatal ultrasonography showed no fetal anomalies. After the birth, he was intubated because of respiratory distress due to a left-sided congenital diaphragmatic hernia. Examination showed microcephaly with a small mouth, a narrow palate, low-set malformed ears, clenched hands, and rocker-bottom feet. There was a large ventricular septal defect and a unilateral hydronephrotic left kidney. Karyotyping in bone-marrow aspirate showed trisomy 18. No corrective surgery was done and the infant died 3 days later. Discussion The clinical spectrum of trisomy 18 ranges from characteristic external features, such as clenched hands with low-arch dermal ridge patterning on the fingertips, to
life-threatening malformations, such as congenital diaphragmatic hernia, cardiac anomalies, or oesophageal atresia. The birth of a child with congenital malformations that suggest Edwards’ syndrome presents doctors with difficult medical and ethical decisions in management of life-threatening malformations. Smith1 recommended limitation of all medical intervention once the diagnosis of trisomy 18 is established and Carter concluded that there is no point in corrective surgery, even if it were indicated on other grounds. What should be done when a newborn infant with suspected Edwards’ syndrome requires prompt medical or surgical intervention before the suspected diagnosis has been confirmed by karyotyping? Many decisions to intervene have to be made under emergency conditions, without much opportunity for discussion or reflection. The availability of neonatal intensive care and skilled paediatric surgery complicates decision-making further. A 1988 article on clinical recognition of trisomy 18 did not discuss this
subject.4 Four of our seven patients, already on ventilators, underwent surgery before the diagnosis was confirmed by routine karyotyping of peripheral lymphocytes. Rapid karyotyping of bone marrow might have allowed invasive treatment to be avoided in patients 1, 2, and 4. In patient 5, even bone-marrow karyotyping was not rapid enough. Patients 6 and 7, both with life-threatening anomalies, did not undergo life-saving surgery after rapid karyotyping in bone-marrow aspirate had confirmed the diagnosis. In all cases we discussed the clinical suspicion of Edwards’ syndrome with the parents and expressed our concern about the short life expectancy and the hopeless option of corrective surgery. The discussion is aimed at the parents’ accepting that their child’s prognosis is hopeless and that there is no point in doing more invasive procedures. It is our experience that parents share our point of view, because they all want the best treatment for their children, which does not necessarily mean full intensive care.5 We opted for the most conservative treatment while awaiting results of karyotyping in peripheral lymphocytes. We believe that immediately karyotyping confirms the clinical diagnosis, the infant undergoes a transformation from a child with a potentially treatable malformation to a patient whose death is imminent, or whose outlook is hopeless when he or she survives the neonatal period. Corrective surgery and life-saving medical support should then be witheld. This view contrasts with that expressed by Van Dyke and Allen;6 they described how medical treatment was continued in long-term survivors, and how in two of six patients, invasive therapy, including open-heart surgery, was used. We believe that, because of their limited life expectancy or profound mental retardation, such patients ought to be transferred from the intensive-care unit and be allowed to die, as does Truog7 for children in a persistent vegetative state. But letting such children die may be insufficient; perhaps we ought to help them to die, and change the aim of treatment from cure into comforting care in their terminal state, as Safar and Winter advocate8 in other patients without hope. The issue is not active euthanasia, which is incompatible with medical ethics, but withholding or withdrawal of treatment in newborn infants whose physical condition is incompatible with life. A strong consensus is growing that the best interest of the infant should be the main criterion for
decision-making.9
915
JH, Harnden DG, Cameron AH, Crosse VM, Wolff OH. A trisomic syndrome. Lancet 1960; i: 787-90. Marion RW, Chitayat D, Hutcheon RG, et al. Trisomy 18 score: a rapid, reliable diagnostic test for trisomy 18. J Pediatr 1988; 113: 45-48. Miraie ED, Mahowald MB. Withholding nutrition from seriously ill newborn infants: a parent’s perspective. J Pediatr 1988: 113: 262-65. van Dijke DC, Allen M. Clinical management considerations in long-term survivors with trisomy 18. Pediatrics 1990; 85: 753-59. Truog RD. Allowing to die. Crit Care Med 1990: 18: 790-91. Safar P, Winter P. Helping to die. Crit Care Med 1990; 18: 788-89. Walters JW. Approaches to ethical decision making in the neonatal intensive care unit. Am J Dis Child 1988; 142: 825-30. Waldstein G, McGavran L. Four-year experience with rapid bone marrow chromosome analyses in newborns. Arch Pathol Lab Med 1987; 111: 703-07.
3. Edwards
It is essential to reduce the period of uncertainty to a minimum, thus avoiding redundant surgery. Karyotyping in bone-marrow aspirate is a useful tool, which should be used whenever Edwards’ syndrome is suspected. 10
new
4. 5.
6. 7. 8. 9.
REFERENCES Trisomy 18 syndrome. In: recognizable patterns of human malformation, 4th ed. Philadelphia: Saunders, 1988: 16-19. 2. Carter PE, Pearn JH, Bell J, Martin N, Anderson NG. Survival in trisomy 18: life tables for use in genetic counselling and clinical paediatrics. Clin Genet 1985; 27: 59-61. 1. Smith DW.
10.
VIEWPOINT Time to abandon TNM
staging of breast cancer?
L. C. BARR In 1968, the International Union Against Cancer (UICC) published its first livre de poche, a booklet on the TNM system of classification of malignant tumours at 23 sites, including breast cancer. The booklet was translated into eleven languages, and the classification has been amended in each of the three subsequent editions. The revision in the current (fourth) edition was such as to make the classification correspond exactly with that of the American Joint Committee for Cancer (AJCC) Manual for Staging of Cancer (1987), to enable oncologists across the world to use a "common language".1 The general rules of the TNM system are well known, with the T-stage reflecting the size of the primary tumour, the N-stage the extent of regional lymph-node metastases, and the M-stage the presence or absence of distant metastases. TNM staging can be clinical (cTNM) or pathological (pTNM). For breast cancer, as with many other tumours, the clinical TNM stage is the more important of the two, since many patients do not undergo primary breast surgery or axillary dissection. The UICC believe that the TNM system can fulfil the following five objectives: (i) aid in the planning of treatment; (ii) give an indication of prognosis; (iii) assist in evaluation of the results of treatment; (iv) facilitate the exchange of information between centres; and (v) contribute to the continuing investigation of human
M. BAUM
larger cancers are more likely to require mastectomy or, if locally advanced, no surgery at all. However, decisions about the relative indications for breast-conserving surgery
With the benefit of over 20 years of hindsight, we would suggest that, within the context of breast cancer, the TNM system of tumour classification has been singularly unsuccessful in the attainment of these objectives, and may even have given false guidance.
mastectomy also need to take into account the proximity of the tumour to the nipple, the size of the tumour relative to that of the breast, multifocality, the presence of adverse pathological features such as an extensive intraduct component, and patient preference. Decisions on whether a tumour can be dealt with by mastectomy are similarly a matter of surgical judgment requiring consideration of tumour mobility on the chest wall, degree of skin involvement, and general condition of the patient. Clinical assessment of tumour diameter correlates poorly with extent of disease as assessed by mammography or histology and is subject to substantial interobserver variation. The clinical T-stage is thus rarely of any practical value to the surgeon. Does the T-stage influence decisions about adjuvant systemic therapy? For node-positive patients, the T-stage is irrelevant to decisions about adjuvant systemic therapy. For node-negative patients, the concept of "minimal breast cancer" has recently been introduced to describe those at especially low risk of distant metastases for whom adjuvant chemotherapy can be avoided, but this group of patients consists of those with ductal carcinoma in situ and lobular carcinoma in situ and invasive tumours less than 0-5 cm or 1 ’0 cm diameter.2.3 These cut-off points do not correspond to the T-stage cut-off point of 2 cm, and of course refer to the pathological tumour diameter rather than the clinical diameter. Does the N-stage influence treatment decisions? Clinical assessment of axillary node involvement is very unreliable, with false-positive and false-negative rates of over 30%.4 Pathological assessment of the axillary nodes, however, is often used to make treatment decisions about adjuvant
Planning of treatment
therapy, since adjuvant chemotherapy is recommended for premenopausal women who are node positive.5,6 The
cancer.
Does TNM staging assist in the planning of treatment? The clinical T-stage of a breast cancer does bear some relation to surgical decision-making, since smaller cancers may be suitable for
breast-conserving
surgery, whereas
versus
ADDRESS. Breast Unit, Royal Marsden Hospital, Fulham Road, London SW3 6JJ, UK (L C Barr, FRCS, Prof M Baum, FRCS).
Correspondence to Prof Michael Baum.
