Cerebral edema causing death in children with maple syrup urine disease

Cerebral edema causing death in children with maple syrup urine disease James J, Riviello, Jr., MD, Iraj Rezvani, MD, A n g e l o M. DiGeorge, MD, a n d C a t h e r i n e M. Foley, MD From the Sections of Child Neurology and Endocrinology and Metabolism, St. Christopher's Hospital for Children, and the Departments of Pediatrics and Neurology, Temple University School of Medicine, Philadelphia, Pennsylvania Four children with the classic form of maple syrup urine disease (MSUD) died of cerebral e d e m a during an intercurrent infection that caused severe dehydration and acidosis. The diagnosis of MSUD had been established during the neonatal period in all four patients, on day 4 of life in three of them. All were in satisfactory control before the infercurrent illness. Two patients underwent peritoneal dialysis. Signs of brain-stem compression occurred after treatment, when biochemical abnormalities were improving. C o m p u t e d tomography of the head, which was done in two patients, revealed cerebral e d e m a ; one of these patients also had subarachnoid hemorrhage. Autopsy in one case revealed cerebral e d e m a with herniation. Our experience documents that cerebral e d e m a may occur in the older child with MSUD as well as in the neonate. The pathogenesis of cerebral e d e m a in MSUD remains unclear. Early treatment of dehydration and acidosis may prevent the catastrophic consequences that we have observed. (J PEDIATR4994;149:42-45)

In the most common form of maple syrup urine disease (classic M S U D ) , lethargy, poor feeding, vomiting, and alternating periods of hypertonicity and flaccidity appear within the first several days of life. If the disease is untreated, progressive neurologic deterioration, seizures, cerebral edema, coma, and death usually occur within the first month of life. 1 Although early diagnosis and treatment may avert or reverse these neurologic findings, including cerebral edema, mental and neurologic residua are common in those infants who survive the neonatal period. Therapy consists mainly of a diet low in branched-chain amino acids, but little is known regarding the long-term clinical course and lifespan of these patients. A mortality rate of 25% has been reported in a follow-up study of 12 children with M S U D who survived the neonatal period; however, the cause of death was not given. 2 Cerebral edema, Submitted for publication Sept. 26, 1989; accepted Feb, 13, 1991. Reprint requests: James J. Riviello, Jr., MD, Section of Pediatric Neurology, Maine Neurology, 180 Park Ave., Portland, ME 04102. 9/20/28994

42

previously unreported after the early months of life in children with MSUD, has resulted in the death of four girls among our 15 patients. CASE R E P O R T S The clinical data regarding diagnosis of MSUD and neurologic status of the patients are summarized in Table I. Patient 1. Symptoms of a mild upper respiratory tract infection developed in patient 1 approximately 1 week before admission to the hospital. Intermittent vomiting and decreased oral intake See related article, p. 35.

DKA MSUD

Diabetic ketoacidosis Maple syrup urine disease

I

occurred 4 days later, and she received rectally administered trimethobenzamide; no aspirin was given. She became progressively lethargic and was brought to the hospital after having a seizure. On admission, she was in a coma and had diffuse hyperreflexia; the pupils were in the mid position and reacted sluggishly to light. Dehydration was estimated at 15%. Kussmaul respirations

Volume 119 Number 1, Part 1

F a t a l cerebral e d e m a in M S U D

43

Table I. Patient information: Background Previous hospitalizations for control

Patient No.

Age (yr)

Developmental level

1

4

2

3

3

5

4

3

Mild to moderate intellectual deficit Normal gross motor skills Intellectual skills: 22 mo level at 3 yr Normal gross motor skills Moderate intellectual deficit Normal gross motor skills Intellectual skills: 9 mo level at 2V2 yr Gross motor skills: 15 mo level at 21/2 yr

Several One Several Several

Table II. Initial blood plasma values at presentation of terminal illness Patient Leucine Isoleucine Valine Glucose AST/ALT NH3 No. (/~mol/L) (#mol/k) (#real/L) (mg/dl) (U/L) (/~g/dl) 1 2 3 4

3970 6259 1644 4878

1389 1527 316 1705

1880 2820 600 250l

167 68 108 0t

24/18 62/34 48/52 ND/ND

ND 146 48 ND

pH 7.41" 7.31" ND 6.94

Pco2 Poa HCO3 Na (ram Hg) (mm Hg) (mm Hg) (mmol/L) 15.9' 15" ND 24

114 158 ND 206

11.8" 5 11" 5

133 139 137 133

AST, Aspartate aminotransferase;ALT, alanine aminotransferase;NH3, ammonia; Pco2, carbon dioxidepressure;Po2, oxygenpressure;HC03, bicarbonateradical; Na, sodium; ND, not done. *Indicates laboratory values obtained after administrationof intravenous fluids and sodium bicarbonate. tDetermined by Dexstrostixreagent strip.

were present, and she had coffee-groundsvomitus. Laboratory data at the time of admission are listed in Table I1. Treatment consisted of intravenous rehydration (Table III), penicillin and ampicillin, and peritoneal dialysis. The patient's mental state remained unchanged despite significant improvement in her metabolic condition. On the second hospital day, decerebrate posturing occurred. Respiratory arrest with fixed, dilated pupils ensued on the third hospital day, with subsequent development of diabetes insipidus and brain death. At autopsy there was marked swelling of the brain with herniation of the temporal lobes and cerebellar tonsils. Patient 2. Patient 2 had intermittent persistent vomiting of 3 days' duration. She received no medications. On the day of admission, coffee-grounds vomitus and decreased consciousness were noted, and she was taken to a local emergency department. Initial physical examination revealed coma with 15% dehydration and Kussmaul respirations. Intravenous fluid therapy was started, and the patient was transferred to our hospital. Initial laboratory data are listed in Table II. Treatment consisted of rehydration (Table III) and Intralipid lipid emulsion (KabiVitrum, Alameda, Calif.). Peritoneal dialysis was started 8 hours after admission. Forty-eight hours after admission, intermittent eye deviation, bruxism, and spontaneous posturing developed; phenobarbital was given for suspected seizure activity. Twenty-four hours later the patient suddenly had apnea, bradycardia, and dilated pupils with loss of corneal reflexes. The trachea was immediately intubated and lungs were hyperventilated. Computed tomography of the head showed cerebral edema. An intraventricular catheter was placed, and the initial intracranial pressure was noted to be 5 cm H20, with elevations as high as 56 cm H20 despite hyperventilation, dexametha-

sone, and mannitol infusion. The condition progressed to brain death with electrocerebral silence on the electroencephalogram, and support was terminated. Autopsy permission was denied. Patient 3. Seven days before admission to the hospital, fever and rhinorrhea developed in patient 3. She received no medications. Progressive dysarthria, ataxia, and lethargy occurred, and vomiting began 1 day before admission. On admission, the patient was stuporous and dehydration was estimated at 10%. Initial laboratory data are listed in Table II. After approximately 12 hours of intravenous rehydration (Table III), the patient became more responsive but remained lethargic; the left pupil reacted sluggishly. Twenty-four hours later, she was in eardiorespiratory arrest. Although successfully resuscitated, she died several hours later. Autopsy permission was denied. Patient 4. Patient 4, the sister of patient 2, had diarrhea 4 days before admission to the hospital, followed by persistent vomiting. She received no medications. On the night before admission, she awakened with labored breathing and was brought to our emergency department. Examination showed hypotension, coma, hyperpnea, and pupils that were 3 to 4 mm in diameter but equal and reactive to light. Dehydration was estimated at 15%. Initial laboratory data are listed in Table II. The patient was flaccid, and no focal neurologic signs were present. Therapy consisted of fluid resuscitation (Table III), tracheal intubation, and hyperventilation. Thirty minutes after the patient's arrival in the intensive care unit, pupillary asymmetry developed, and treatment with mannitol followed by furosemide was started. Computed tomography revealed cerebral edema with subarachnoid hemorrhage. The parents refused placement of an intracranial pressure monitoring device. Brain death ensued, with electrocerebral silence on the electroen-

44

Riviello et al.

The Journal of Pediatrics July 1991

T a b l e III. Amounts of fluid administered on days before cerebral herniation

Fluid levels Patient No.

1 2 3 4

Day t

Day 2

Day 3

2.9 L/m 2 2.7 L/m 2 4.l L/m 2 420 ml*

4.2 L/m 2 3.3 L/m 2

3.4 L/m 2

*Before herniation.

ccphalogram, and support was terminated. Aatopsy permission was denied. DISCUSSION The four deaths reported here were in children 3 to 5 years of age. All four had been adequately managed with relatively few problems until the episodes that resulted in death. The diagnosis of cerebral edema was conclusively established at autopsy in one patient, and was demonstrated in two by computed tomography. Although the diagnosis was not proved in patient 3, the clinical course was similar to that of the other three patients. Cerebral edema with increased intracranial pressure has been described in neonates with either classic or variant forms of MSUD 37 and is usually reversible with dietary therapy. Cerebral edema is also a complication of other metabolic disorders, such as diabetic ketoacidosis, Reye syndrome, hyperammonemic syndromes, certain amino acid disorders and organic acidemias, and disorders of lactate, pyruvate, and carnitine metabolism. Pseudotumor cerebri also has been observed in children with galactosemia, 8 galactokinase deficiency,9 and various endocrinopathies. The pathogenesis of cerebral edema in infants and children with MSUD, as well as in these other conditions, remains unknown. Neurologic deterioration in our four patients occurred after treatment of dehydration, although patient 4 was near death at presentation. This is reminiscent of the situation in DKA, in which rapid correction of hypovolemia and metabolic derangements may cause or aggravate preexisting cerebral edema; in particular, fluid administration of more than 4 L / m 2 per day has been postulated as a risk factor for the development of cerebral edema.l~ Two of our patients received >4 L / m z per day of fluid before clinical deterioration. Perhaps more conservative fluid administration in patients with MSUD and dehydration, as in DKA, may diminish exacerbation of cerebral edema. In addition, colloids have been advocated for fl,uid replacement in DKA because they remain in the intravascular space. 14

The clinical courses of our patients were similar to those of the patients of Naughten et al. 2 Presentation of all 12 of their patients was during the neonatal period, although in two patients the diagnosis had been delayed; the three childhood deaths occurred during intercurrent infections, and persistent hypokalemia was present in two of their patients. Although no other details were given (such as the development of coma, cranial nerve abnormalities, or autopsy findings), the clinical course suggests cerebral edema as the cause of death in each of these three children. The outcome of MSUD has greatly improved since its initial recognition, is In the classic form of the disease, neurologic prognosis is related to the age of the child at the time of diagnosis and to the adequacy of metabolic control. 16, 17 The prognosis should be best if treatment is initiated on the first day of life. At present, this goal can be achieved only in infants with an affected sibling 18 or in the event of prenatal diagnosis. However, our experience suggests that early dietary treatment may not prevent developmental and neurologic sequelae in all patients. It is disconcerting that three of our patients, although their MSUD was diagnosed during the first 24 hours of life and adequately controlled thereafter, were among the four children whose deaths were noted in this report. We are aware of two other unreported deaths during intercurrent infections in children with MSUD, one with documented cerebral edema. The pathogenesis of cerebral edema in MSUD and other metabolic conditions remains unknown. Status spongiosius is a common neuropathologic finding in MSUD. Altered amino acid metabolism may interfere with the function of the blood-brain barrier, which may predispose patients to the development of cerebral edema in association with an acute illness. During an intercurrent illness, such as occurred in our patients, dehydration with increased intracellular osmolality, a rapid fall in extracellular osmolality, excess antidiuretic hormone, and overhydration may exacerbate this edema. In addition, a buildup of toxic substances, such as a-ketoisocaproic acid, may aggravate cytotoxic edema. Activation of the N a + / H + exchanger, or antiport, has recently been hypothesized as a possible mechanism for the cerebral edema seen in DKA and other metabolic disorders, including MSUD. t9 Despite the underlying diagnosis of MSUD, the differential diagnosis must include Reye syndrome because these patients had an influenza-like illness followed by vomiting, cerebral edema, and death. Toxic encephalopathy, sepsis, meningitis, and encephalitis must also be excluded. The marked elevation of branched-chain keto acid values supports the diagnosis of a primary exacerbation of MSUD. Initial laboratory values in our patients do not support the diagnosis of Reye syndrome; no aspirin was given, and only

Volume 119 Number 1, Part 1

Fatal cerebral edema in M S U D

one patient received trimethobenzamide, making toxic encephalopathy unlikely; blood cultures and results of examination of the cerebrospinal fluid were negative. Our experience documents the frequent occurrence of cerebral edema in children with MSUD, a complication usually precipitated by acute illnesses associated with vomiting, dehydration, and worsening of biochemical control. Serious consideration should be given to early hospitalization and cautious rehydration in all children with MSUD in whom intercurrent infections develop in association with decreased nutrient intake or vomiting or both. An increased awareness of this aspect of MSUD should lead to earlier and improved management of these children, thereby avoiding the catastrophic consequences that we have observed.

REFERENCES 1. Snyderman SE. Medical and nutritional aspects of maple-syrup-urine disease. In: Koch R, Shaw KNF, Durkin F, eds. Maple syrup urine disease symposium: issues and perspectives. Rockville, Md.: U.S. Government Printing Office, 1979:18-33. 2. Naughten ER, Jenkins J, Francis DEM, Leonard JV. Outcome of maple-syrup-urinedisease. Arch Dis Child 1982;57~918-21. 3. Committee for Improvement of Hereditary Disease Management. Management of maple syrup urine disease in Canada. Can Med Assoc J 1976;115:1005-t3. 4. Mantovani JF, Naidich TP, Prensky AL, Dodson WE, Williams JC. MSUD: presentation with pseudotumor cerebri and CT abnormalities. J PEDIATP,1980;96:279-81. 5. Mikati MA, Dudin GE, Der Kaloustian VM, BensonPF, Fensore AH. Maple-syrup-urine disease with increased intracranial pressure. Am J Dis Child 1982;136:642-3. 6. Lungarotti MS, Calabro A, Signorini E, Garibaldi LR. Cerebral edema in maple syrup urine disease. Am J Dis Child I982;136:648.

45

7. RomeroFJ, Ibarra B, Rovira M, Natal A, Herrera M, Segarra A. Cerebral computed tomography in maple syrup urine disease. J Comput Assist Tomogr 1984;8:410-1. 8. Huttenlocher PR, Hillman RE, Hsia YE. Pseudotumor cerebri in galactosemia. J PEDIATR1970;76:902-5. 9. Litman N, Kanter AI, Finberg L. Galactokinase deficiency presenting as pseudotumor cerebri. J PEDIATR1975;86:410-2. 10. Duck SC, Wyatt DT. Factors associated with brain herniation in the treatment of diabetic ketoacidosis. J PEDIATR 1988; 113:10-4. 11. Harris GD, Fiordalisi I, Finberg L. Safe management of diabetic ketoacidemia. J PEDIATR1988;113:65-8. 12. Krane EJ. Cerebral edema in diabetic ketoacidosis [Letter]. J PEDIATR 1989;114:166. 13. Harris GD, Fiordalisi I, Harris WL, MosovichLL, Finberg L. Minimizing the risk of brain herniation during treatment of diabetic ketoacidemia: a retrospective and prospective study. J PEDIATR1990;117:22-31. 14. Hillman KM. Resuscitation in diabetic ketoacidosis. Crit Care Med 1983;11:53-4. 15. Menkes JH, Hurst PL, Craig JM. A new syndrome: progressive familial infantile cerebral dysfunction associated with an unusual urinary substance. Pediatrics 1954;14:462-6. 16. Naylor EW, Guthrie R. Newborn screening of maple-syrupurine disease (branched-chain ketoaciduria). Pediatrics 1978;61:262-6. 17. Clow CL, Reade TM, Scriver CR. Outcome of early and longterm management of classical maple syrup urine disease. Pediatrics 1981;68:856-62. 18. DiGeorge AM, Rezvani I, Garibaldi LR, Schwartz M. Prospective study of maple-syrup-urine disease for the first four days of life. N Engl J Med 1982;307:1492-5. 19. Van der Meulin JA, Klip A, Grinstein S. Possible mechanism for cerebral edema in diabetic ketoacidosis. Lancet 1987;2: 306-8.

FELLOWSHIPS Beginning in January 1992, fellowships available in pediatric subspecialties and those for general academic pediatric training will be listed once a year, in January, in THE JOURNAL OF PEDIATRICS. Each July, forms for listing fellowships available for the academic year beginning 18 months after publication will be sent to the Chairman of the Department of Pediatrics at most major hospitals in the United States and Canada. In addition, a copy of the application form will appear in the August and September issues of THE JOURNAL (please use the current form). Should you desire to list fellowships, a separate application must be made each year for each position. All applications must be returned to Mosby-Year Book, Inc., by October 15 preceding the listing year to ensure publication. Additional forms will be supplied on request from the Journal Editing Department, Mosby-Year Book, Inc., 11830 Westline Industrial Drive, St. Louis, MO 63146-3318/ 314-453-4317.