Abstract
There is a lack of awareness of acutely presenting inborn errors of metabolism in adults, of which the X-linked urea cycle defect ornithine transcarbamylase (OTC) deficiency is an example, many comparatively mild mutations having been identified. In male hemizygotes clinical manifestations and age at presentation vary and depend on the mutation. In female heterozygotes the clinical spectrum depends on the extent to which the abnormal gene is expressed. Milder versions of the defect may not cause clear clinical symptoms and may remain unrecognized until the person is subjected to an unusually high nitrogen load when they develop severe hyperammonaemia. During acute episodes liver enzymes may be normal or only slightly elevated and occasionally accompanied by coagulopathy, but the key finding is hyperammonaemia. Boys with these milder forms may exhibit abnormal behaviour and be diagnosed with attention deficit hyperactivity disorder. This case illustrates how late presentation of OTC deficiency in a non-specialist centre can be difficult to differentiate from drug abuse, psychiatric illness or encephalopathy. Failure to measure blood ammonia in adults with unexplained key symptoms – particularly prolonged vomiting without diarrhoea and altered mental state/hallucinations, or to recognize the significance of elevated blood ammonia without evidence of liver decompensation can lead to delayed or missed diagnosis.
Case history
A 24-year-old Caucasian man was referred to the accident and emergency (A&E) department with dehydration due to nausea and vomiting of one-week duration. He had not passed a motion for several days. He appeared disoriented and complained of tactile hallucinations (‘things creeping on his skin’). His only prescribed medication was salbutamol inhaler prn (pro re nata or as needed) for asthma, but he was taking methandrostenolone and a creatine containing food supplement in an attempt to body build. He also admitted to cannabis use on alternate days.
He was afebrile with normal blood pressure and regular pulse. Examination of his chest and abdomen showed no abnormality. His Glasgow Coma Score was 15 and he was judged to have normal speech and gait and normal power in his limbs. There was no neck stiffness.
Blood count, renal and liver profiles were essentially normal, but prothrombin time was 28.5 s (11–16s); INR (International normalized ratio) was 2.2 (0.9–1.2) and activated partial thromboplastin time was 47 s (29–37 s). Clotting factor assay revealed subnormal factor VII (35 IU/dL reference range: 50–200) and low factor V. The patient was re-hydrated, and feeling slightly better the next day, he was discharged. A urine drug screen revealed only cannabis. An ultrasound scan of the abdomen showed a liver of normal appearance with no sinister features.
Previous medical history
The patient's medical notes revealed a long history of childhood problems from birth to age 15 y, mainly due to asthma, growth retardation and behavioural problems. He was diagnosed with attention deficit hyperactivity disorder for which he was unsuccessfully treated with methylphenidate (Ritalin) for a short time. He had experienced severe headache preceded by episodes of facial twitching when aged 11 y. After the age of 15 y, his weight and height became normal and he was discharged. He ate a normal diet, but had recently become concerned about his slight stature and increased his meat intake, embarking on a course of bodybuilding and food supplements to improve his appearance. Crucially his symptoms had been considered to be due to a psychological or social disorder. Childhood laboratory investigations had included sweat tests (x4), thyroid function tests, full blood count, renal profile and karyotyping with exclusion of fragile X syndrome, all of which were essentially normal.
Progress
Three days later he re-presented to the A&E department in an acute confusional state unable to give a history and disoriented in time, place and person. He had been vomiting and hallucinating. Apart from the marked and progressive confusion, clinical examination was similar to his first attendance. He was afebrile and remained haemodynamically stable. Clotting studies were again abnormal (INR was 2.7), with slight increases in liver enzymes. Serum aspartate aminotransferase was 114 U/L (20–39 U/L), having risen from 57 U/L when first seen; alkaline phosphatase was 139 U/L (25–114 U/L); gamma glutamyl transferase was normal and bilirubin was 22 μmol/L (5–21 μmol/L). Plasma ammonia was 348 μmol/L (<48 μmol/L) with partially compensated respiratory alkalosis, pH 7.504, [H] 31 nmol/L. The possibility of an intracerebral bleed was considered. Results of a computed tomography (CT) scan of the head aroused suspicion of subarachnoid blood, but the area of hyperdensity seen could not be visualized in a follow-up scan.
Advice given to medical staff (from an external specialist centre for hepatology) was that the raised ammonia was likely to be a ‘red herring’ and that the priority should be to obtain a neurosurgical opinion in view of the suspected subarachnoid haemorrhage. No further ammonia measurements were made. The possible differential diagnoses considered were therefore intracerebral bleeding and drug toxicity. The laboratory suggested urine orotic acid and organic acid analysis, but a metabolic disorder had not been part of the work-up and the sample was not processed urgently.
The patient developed cerebral oedema and seizures. He became comatosed with fixed and dilated pupils. Life support was withdrawn on the fourth day after his re-admission.
Results
Results of the organic acid and orotic acid analyses were received some weeks after death. Urine orotic acid was 223 μmol/mmol creatinine (normal: <5 μmol/mmol creatinine) and urine uracil was 190 µmol/mmol creatinine (normal: <50 μmol/mmol creatinine). These were consistent with the urea cycle defect, ornithine transcarbamylase (OTC) deficiency. 1 The patient's mother had normal random urine orotic acid and uracil, but these rose to 11.1 μmol/mmol creatinine and 77 μmol/mmol creatinine, respectively, after a high protein meal. Molecular investigations revealed that she is a carrier of the OTC mutation 118C > T (Arg40Cys), which has previously been associated with fatal late-onset OTC deficiency in boys. 2 Extracted patient DNA, which had been stored at the time of karyotyping (in childhood), gave consistent results.
Discussion
Ammonia toxicity caused by urea cycle defects in adults and neonates has been recently reviewed and the incidence of clinical presentations suggested is 1/45,000. 3 A United States study estimates the prevalence of urea cycle defects to be at least 1/25,000 births, but suggests that the true incidence (if undiagnosed mutations associated with late presentation could also be included) is considerably higher. 4 Of the 341 genetic mutations for OTC deficiency described by the year 2006, 149 are associated with neonatal presentation. 5
Although hyperammonaemia is considered in sick neonates, plasma ammonia is infrequently requested in older patients, yet 66% of urea cycle defects were first reported well after the first month of life in the United States study. 4 The mutations associated with milder versions of OTC deficiency may not be recognized until the person is subjected to an unusual nitrogen load either as a result of increased intake, as in our patient, or due to a catabolic state associated with illness or surgery. OTC deficiency presenting in males or female heterozygotes in middle age and even old age continue to be reported. 6,7 For older patients, mortality rates are high at first presentation because symptoms of hyperammonaemia are not usually recognized in time for successful treatment. It is probable that some urea cycle defects are not diagnosed and therefore go undocumented. In this case, it was only by chance that laboratory data aroused suspicion at the clinical validation stage and the possibility of a urea cycle defect was pursued.
The coagulopathy in this patient, who was not on warfarin, with near normal biochemistry tests of liver function had attracted attention, but clinical symptoms of hyperammonaemia were not recognized, being attributed instead to possible drug use and he was allowed home.
Significance of hyperammonaemia with respiratory alkalosis
On representation three days later, plasma ammonia was ordered, but the link between hyperammonaemia, respiratory alkalosis and a metabolic defect was not made. Laboratory staff were unaware of the data obtained on the A&E blood gas analyser which was without IT connectivity. A second window of opportunity to treat the patient was lost. 8
When plasma ammonia should be measured
The importance of considering an inherited disorder of metabolism has been stated previously when adult patients present with features typical of hyperammonaemia in the apparent absence of liver decompensation. 9–11 Key signs include protracted vomiting without diarrhoea, and unexplained altered mental state. 12 The encephalopathy of hyperammonaemia usually develops over a few days, as in this case, offering an opportunity to reverse its effects. Survival rates are usually above 80% with latest treatment regimens, even if the patient is comatose on admission. 8
Plasma ammonia is not part of the work-up for investigating acute encephalopathy in the UK, so clinicians may need reminding to order its measurement. Spurious hyperammonaemia due to incorrect preanalytical procedure does occur and must be excluded not assumed. 9 Establishing plasma ammonia as part of a test panel used in a ‘confusion screen’ applied to children and adults where there is no apparent explanation for an acute confusional state could be the way forward.
Conclusion
Diagnosis of OTC deficiency in adults or children in non-specialist centres is difficult when milder versions of the mutation are present as the disease can be episodic and laboratory findings can be normal between episodes;
OTC deficiency can present at any age and clinicians may need to be reminded to order plasma ammonia when confronted with a patient with unexplained odd behaviour and/or protracted vomiting without diarrhoea;
The combination of elevated plasma ammonia without evidence of liver decompensation should arouse suspicion of an inherited metabolic defect. An associated respiratory alkalosis is also a key sign and the laboratory has a role in advising appropriate investigations;
Establishing plasma ammonia as part of a test panel for acute and inexplicably altered mental state may be the way forward as early aggressive treatment can reverse the encephalopathy of hyperammonaemia.
DECLARATIONS