Marked cerebrospinal fluid collection associated with hepatic failure

CASE REPOR TS

Marked Cerebrospinal Fluid Collection Associated with Hepatic Failure Takao Enomoto, MD, Masanori Hashikawa, MD, Tadao Nose, MD, Yutaka Maki, MD, Kenzo Hamano, MD and Norimasa Nakamura, MD

We report two cases with abnormal CSF accumulation in the cranium. One case had marked dilatation of the ventricular system, for which a ventriculoperitoneal shunt was carried out. The shunt was not beneficial and it turned out that the patient was suffering from ornithine transcarbamylase deficiency. The second case developed marked subdural effusion after fulminating hepatic failure and septicemia, for which drainage of the subdural space and subsequent insertion of Om maya reservoirs were carried out. In spite of successful CSF control, the outcome was bad, both cases being incapacitated due to severe encephalopathy probably caused by hyperammonemia. The results of serial computed tomographic follow-up are presented. Key words: Ornithine transcarbamylase deficiency, hepatitis, hyperammonemia, brain atrophy, subdural effusion. Enomoto T, Hashikawa M, Nose T, Maki Y, Hamano K, Nakamura N Marked cerebrospinal fluid collection associated with hepatic failure. Brain Dev 1988; 1 0: 389-92

Elevation of the serum ammonia level has been considered to be a cause of histopathological changes in the brain, although another functional theory has been proposed to explain the hepatic coma [1]. Recently, a marked anatomical change was reported in congenital hyperammonemia [2-4] and a mild change in acute hepatic failure [5]. The authors had the chance to experience two unusual cases and to follow them up by means of serial computed tomography (CT). To our knowledge, chronological changes observed on CT have seldom been reported. Therefore, we report these cases with particular reference to the CT results. CASE REPORT Case I A year and a half old girl was admitted to our hospital in October, 1984, because of vomiting and subscalp fluid

From the Departments of Neurosurgery (TE, MH, TN, YM) and Pediatrics (KH, NN), Institute of Clinical Medicine, University of Tsukuba, Tsukuba, Ibaraki. Received for publication: July 6,1987. Accepted for publication: July 23,1988. Correspondence address: Dr. Takao Enomoto, Department of Neurosurgery, Institute of Clinical Medicine. University of Tsukuba, Tennodai, Tsukuba, Ibaraki 305, Japan.

collection around a shunt device. She was born to unrelated healthy parents after 38 weeks' uneventful gestation, with vacuum traction. She weighed 2,650 g at birth and there was no asphyxia. A brother had died of an unknown cause 2 days after birth and there is an elder sister who is healthy at present. The patient's development had been quite normal until March 24, 1984, when she developed a fever followed by vomiting. A month later she developed gingival hemorrhaging and was admitted to hospital, where she was found to have impaired consciousness, tetraplegia, myoclonus, vomiting, coagulability disturbance and an elevated transaminase level. CT demonstrated diffuse brain edema (Fig 1). She underwent blood transfusion, but 2 months later CT showed marked ventricular dilatation. For this reason, a ventriculo-peritoneal (V-P) shunt was inserted and then the patient was discharged. Subsequently, she was admitted to our hospital with the complaints mentioned above. On admission she was found to have subscalp fluid collection along the shunt device. Her head circumference was 47 cm, which was the mean for her age. She was irritable and cried easily. She was spastic, tetraplegic and moved only her right arm slightly. There was no papilledema. Laboratory studies showed the serum transaminase levels to be slightly elevated, GOT was 36 and GPT 52. The blood cell count was normal. CT scanning showed marked dilatation of the ventricular system and less conspicuous cortical sulci (Fig 1). An RI cisternogram

showed ventricular filling but normal clearance. An Rl shuntgram revealed a complete block on the abdominal side. The ventricular CSF pressure, measured via punctured valve, was 120 mmH20. EEG was abnormal, with an admixture of sharp waves. Although the head size was normal, in view of the patient's repeated vomiting, irritability with spasticity and increased ventricular size with fluid escaping into the subgaleal space, we decided to revise the shunt. After the operation, the vomiting ceased but there was no improvement in the CT findings. In December, she was discharged home again and did relatively well until September, 1985, when she developed feeding difficulty and pneumonia, and so was readmitted. This time the serum ammonia level was elevated to 82011g/dl. Several kinds of amino acid loading tests and an open liver biopsy were carried out. She tolerated the procedure well. Histopathological examination revealed marked glycogen-accumulated hepatocytes and biochemical investigations disclosed a marked decrease in ornithine transcarbamylase (OTC) activity, to less than 10% of the normal value. A low protein diet, 1.5 g/kg/day, was prescribed, which has worked well so far. Currently, she is being followed up in the outpatient clinic but is severely incapacitated. Case 2 A 4-month-old girl was referred to our hospital for possible treatment for acute fulminating hepatitis. She was born on March 27, 1985, after 38 weeks' uneventful gestation and delivery to a mother who was positive for HBs antigen. She weighed 2,900 g and her head circumference was 29.5 cm. She showed good head control by age 3

months. At the end of the 3rd month her mother noticed yellow discolouration of her conjunctivae. She passed a white stool, and was found to have a palpable liver and elevated serum transaminase levels (GOT 2,060, GPT 928), the bilirubin level being elevated to 13.5 mg/dl. She was referred to us on August 5, 1985, with a tentative diagnosis of acute fulminating hepatitis. On examination on admission she was irritable but clear mentally. She was icteric and had some stigmas of subcutaneous hemorrhage. The liver was palpable 2 cm below the xyphoid process. She had a hepatic odour. The serum transaminase level started to settle down. The ammonia level was slightly elevated, being 161 I1g/ dl. The PT and APTT were prolonged_ She was positive for HBc antibody. The CT scan seemed normal, but she gradually became comatous. On repeated CT, low attenuation of the white matter was seen (Fig 2). The laboratory data levels except for those of serum ammonia and total bilirubin subsided, but her consciousness deteriorated, followed by fever, cyanosis, hypotension and gasping respiration. A blood culture yielded gram negative rods. In spite of intensive treatment for sepsis, the fontanelle became distended in September. CT revealed major subdural fluid collection and subarachnoid hemorrhage. The anterior fontanelle was widely open (5 x 5 cm) and distended. The head circumference was 44.5 cm (+1 SD). The fontanelle was punctured, with an opening pressure of 300 mmH20, yielding a yellowish-brown turbid fluid, whereas a lumbar tap produced clear CSF devoid of inflammatory signs. The bilateral subdural drainages were connected to external reservoirs. The protein content of the subdural fluid was 1,000 mg/dl and the cell count was 2,800/3. A total of about 500 ml of turbid fluid brained out every day.

Fig 1 Computed tomograms of case 1. A: Low density of white matter compressing the lateral ventricle on the right side, B: Progress of the edematous change leaving basal ganglia and posterior fossa structures unchanged, C and D: Resolution of edema and concomitant enlargement of the ventricular system. A shunt was inserted after D and revised after E without any beneficial effect 2 years later F.

390 Brain & Development, VallO, No 6,1988

Fig 2 Chronological review of computed tomograms of case 2.

The initial change was low attenuation of the white matter followed by massive subdural collection after an episode of septicemia.

The head circumference was at the +2 SD level. Metrizamide and RI cisternography, which were carried out in December, demonstrated good clearance but almost no RI activity was traced in the drain fluid. After confirming decreases in debris and the amount of discharge, the external drainages were converted to bilateral Ommaya reservoirs. As the reservoirs were tapped, the fontanelle became sunken and the abnormal increase in head circumference ceased. The child's countenance improved, she smiled at times, although she was still severely incapacitated. She was discharged home on August 26, 1986, after training in basic feeding. Currently, she is being followed up in the outpatient clinic. Her development level is estimated to be about 2 months although her chronological age is 1 year and 11 months.

DISCUSSION The central nervous system and the liver seem to be intimately connected through ammonia, a semi-physiological substance. In many liver diseases, serum ammonia increases as a final common product, which in turn provokes neurological deterioration both functionally [1] and structurally [2,5-7]. Congenital hyperammonemia is

a well-known but rare disease cuased by five different enzyme deficiencies, of which ornithine transcarbamylase deficiency is the commonest. There have been some reports dealing with CT findings [3, 8] and some with pathological findings [2-4]. The degree of the histopathological change varies, depending on the degree, duration and onset age of symptoms. Severe histopathological brain changes consist of marked ventricular dilatation and a thinning of the maximal parenchyme in the frontal lobes. There is collagenous thickening of the leptomeningeal layers. The neurons are lost, being replaced by gliosis and parenchymal microcyst formation. Although the degree of the changes varies, the transformation of glial cells to Alzheimer type II cells invariably occurs, even in the mildest form with normal CT findings. Therefore, CT findings also vary from case to case. Usually, the change starts in the acute encephalopathic state with edematous attenuation of the white matter in the frontal lobes; in the chronic state it is characterized by nonspecific partial or generalized ventricular dilatation [4]. Clinically, it is characterized by episodic vomiting, drowsiness, acute encephalopathy and failure to thrive [4]. In case 1, retrospectively, it is apparent that the signs and symptoms, and CT changes followed a typical course of the disease. Nevertheless, based on the combination of marked ventricular dilatation, a blocked shunt with sub scalp CSF leakage, irritability, opisthotonus and repeated vomiting in the absence of hyperammonemia, we tried to revise the shunt in the hope of some improvement, but in vain. Finally, it turned out that the OTC level decreased to less than 10% of the normal level. Usually, in cases of acute fulminating hepatic failure, the initial change occurring is brain edema, sometimes to the extent that it causes brain herniation [6, 7]. In a chronic state, brain atrophy has been reported [5], although it was not as severe as in our case. In case 2, the generalized attenuation of the white matter in the initial stage may have been caused by acute hepatic failure. The following massive subdural fluid accumulation cannot comprise hydrocephalus ex vacuo because of the moderately high intracranial pressure and the increasing head size with a bulging fontanelle. Therefore, we assume that the subdural effusion following septic shock and subarachnoid hemorrhage with secondary rupture into the subdural space, together with the basic process of brain atrophy, contributed to this unusual state of massive subdural fluid collection. Socalled external hydrocephalus was once thought to be the cause of this state but was excluded eventually in view of the small lateral ventricles and the failure to detect radioactivity in the drained fluid. The RI and metrizamide study suggested minimal communication between the subarachnoid and subdural spaces. Inflammatory exudation was considered to be one of the contributing factors in

Enomoto et al: Hepatic failure and CSF collection 391

massive subdural collection. After a considerable period of external drainage, the effusion became self-limiting, which discouraged us from carrying out a subduroperitoneal shunt operation. Now the patient is delayed developmentally, although her subdural effusion is improving slowly. In view of these two cases, we have to be aware of the pitfalls as to indication to operate, namely, that hepatic failure with or without hyperammonemia can often be associated with massive CSF collection with underlying organic brain changes.

REFERENCES 1. Ishikawa A, Ishiyama H, Enomoto T, et al. Hepatic coma and amino acids in the nerve endings of the central nervous system. Life Sci 1985;37:2129-34. 2. Bruton CJ, Corsellis JAN, Russel A. Hereditary hyperammonemia. Brain 1970;93:423-33. 3. Kendall BE, Kingsley DPE, Leonard JV, Lingam S, Ober-

4.

5.

6.

7..

8.

holzer VG. Neurological features and computed tomography of the brain in children with ornithine carbamoyl transferase deficiency. J Neurol Neurosurg Psychiatry 1983;46:28-34. Uchikoshi T, Nishimura H, Ishiko T, et al. Neuropathological study of 2 autopsy cases of hyperammonemia (ornithine transcarbamylase deficiency) occurring in female siblings (in Japanese). Shinkei Kenkyu No Shinpo (Tokyo) 1977;21: 417-25. Toda C, Chiba T, Matsuda Y, Inatome T, Inoh T, Fujita T. A case of brain atrophy after fulminant hepatic failure. Am J Gastroenterol 1983;78:446-8. Sutherland LR, Muller P, Robertlewis D Jr. Massive cerebral edema associated with fulminating hepatic failure in acetaminophen overdose. Possible role of cranial decompression. Am J Gastroenterol 1981 ;76:446-8. Ware AJ, D'Agostino AN, Combes B. Cerebral edema: major complication of massive hepatic necrosis. Gastroenterology 1971;61:877-84. Takayanagi M, Ohtake A, Ogra N, Nakajima H, Hoshino M. A female case of ornithine transcarbamylase deficiency with marked computed tomographic abnormalities of the brain. Brain Dev (Tokyo) 1984;6:58-60.

A Case of Dandy-Walker Malformation: Consideration on the Teratogenic Period and Sleep Structures Jun Kohyama, MD, Seiichi Watanabe, MD, Yoshihide Iwakawa, MD and Chikao Fukuda, MD

An ll-month-old girl suffering from Dandy-Walker malformation (DWM) associated with tetralogy of Fallot (TOF) is presented. There has been no report describing a case of DWM associated with TOF. Our case may raise a new discussion on the teratogenic period of these anomalies. A polysomnogram was obtained, and basic sleep components, such as the proportion of each sleep stage and the numbers of body movements and rapid eye movements, showed no significant differences from those in the controls, despite hydrocephalus, pachygyria of the cerebral cortex and a large low density area in the posterior fossa. Key words: Dandy-Walker malformation, tetralogy of Fallot, polysomnogram. Kohyama J, Watanabe S, Iwakawa Y, Fukuda C. A case of Dandy- Walker malformation: consideration on the teratogenic period and sleep structures. Brain Dev 1988; 1 0: 3 92-6

There has been, to our knowledge, no reported case of Dandy-Walker malformation (DWM) associated with tetralogy of Fallot (TO F), and also no previous sleep study on a patient with a cerebellar defect. We present here an 1I-month-old female who had these two congenital anomalies, and discuss the teratogenic period and

her sleep characteristics. CASE REPORT An 11-month-old female was admitted to our hospital because of developmental delay. She was the first child