Acute delirium in a critically ill patient may be a wolf in sheep’s clothing

Abstract

Acute delirium is a commonly encountered problem in the intensive care unit (ICU), which has a myriad of causes and contributes to poor outcomes. We present the case of an alcoholic critically ill patient who developed prolonged acute ICU delirium wrongly diagnosed as sedation and alcohol withdrawal. Protracted vomiting, swallowing disorders and continuous aspirations prevented him from enteral feeding and discontinuation of mechanical ventilation. After several days, it became clear that the patient had been misdiagnosed. Fortunately, nystagmus and ophthalmoplegia then allowed the recognition of Wernicke’s encephalopathy, confirmed by cerebral MRIs. After thiamine supplementation, his state improved but he was discharged only on day 32. Wernicke’s encephalopathy is an acute reversible neuropsychiatric emergency, which is falsely considered as uncommon, and is largely misdiagnosed, especially in critically ill patients. Thiamine should be systematically given to all critically ill alcoholic patients, especially those with protracted vomiting.

BACKGROUND

Wernicke’s encephalopathy is an acute reversible neuropsychiatric emergency, which is falsely considered as uncommon, and is largely misdiagnosed, especially in critically ill patients. The second pitfall illustrated by this case report is in the belief that patients admitted to hospital are given the total amount of enteral feeding prescribed by doctors in charge. This usual misconception may lead to profound vitamin B₁ deficiency in the weakest patients, especially those presenting an acute condition requiring high metabolic demands on already depleted vitamin stores. Consequently, Wernicke’s encephalopathy may develop as a hospital-acquired disease in a critically ill patient creeping along like a wolf in sheep’s clothing.

CASE PRESENTATION

We report the case of a 43-year-old man admitted to the intensive care unit (ICU) for deep cervical stab wounds following a family altercation and resulting in a haemorrhagic shock and an extensive cellulitis of the neck. He had no evident past medical history except alcohol abuse (60 g of daily alcohol intake). He was immediately sedated and placed under mechanical ventilation by the pre-hospital emergency medical team that ensured control of active bleeding. At hospital admission, resuscitation started with volemic expansion by massive transfusion and early surgical haemostatic procedures consisted in ligatures of the right external jugular vein and facial artery and draining of the cervical field. The first blood sample revealed an alcohol level of 1.4 g/L. Four days later, his state had improved allowing sedation to be stopped. The patient then presented an acute delirium with protracted vomiting and successive phases of agitation and drowsiness, initially diagnosed as alcohol, benzodiazepine and morphine withdrawal. He was given antiemetic drugs, low dose of opiates, benzodiazepines and then neuroleptic were tried but no improvement was noted. With such a therapeutic escalation, it was impossible to know if all these medications may have worsened his neurological state. At this stage, enteral feeding was frequently stopped during the night and nurses had to use nasogastric tubing to suction regurgitation and prevent the risk of aspiration. Medical staff were not immediately aware of these problems with enteral nutrition. After 10 days, the patient became markedly cachectic, he had lost 12 kg (weighed 51 kg for 1.75 m tall) and also presented severe muscle wasting. Prolonged delirium and swallowing disorders were responsible for continuous aspiration and prevented him from weaning the ventilator. Tracheostomy was necessary after 2 weeks of mechanical ventilation. At this time, he developed horizontal nystagmus with ophthalmoplegia.

INVESTIGATIONS

Laboratory evaluation revealed normal sodium, chloride, potassium, phosphate and magnesium levels. Liver enzymes (aspartate aminotransferase 35 IU/L; alanine aminotransferase, 43 IU/L; alkaline phosphatase 110 IU/L), serum bilirubin level (11 mg/L) and ammonaemia (20 μmol/L) were within normal ranges. His elevated gamma-glutamyl transferase level (112 IU/L) was related to his chronic alcohol consumption. Cerebral MRI showed arguments in favour of the clinically suspected diagnosis of Wernicke’s encephalopathy (figs 1A–D).

Figure 1.

Axial fluid-attenuated inversion recovery (FLAIR) MR images showed abnormal high signal intensity along the third ventricle (A) and aqueduct (B) as pointed out by the white arrows. Coronal T₂-weighted (C, D) MR images confirmed this abnormal hyperintensity of the periaqueductal grey matter suggestive of the Wernicke’s encephalopathy.

DIFFERENTIAL DIAGNOSIS

In the differential diagnosis of acute delirium in an alcoholic malnourished critically ill patient, sedative withdrawal, delirium tremens, hepatic encephalopathy, hypophosphataemia related to refeeding syndrome and anoxic brain injury have been mainly considered.

TREATMENT

He was given parenteral nutrition with vitamin B₁ (thiamine) supplementation.

OUTCOME AND FOLLOW-UP

After 5 days of vitaminotherapy, his neurological and respiratory state improved allowing the discontinuation of mechanical ventilation and decannulation. At this time, his Mini Mental State Exam reached a normal score of 28/30. He was transferred to a respiratory ward at day 32. He was discharged from hospital after 1 month of rehabilitation without any persistent neurological disorder.

DISCUSSION

In a patient, one minor disease may hide another dangerous one like a wolf creeping along in sheep’s clothing. In the present case, the medical staff had been mistakenly convinced that sedative withdrawal was the only diagnosis explaining all the patient’s symptoms. This case highlights the vigilance needed in the diagnosis and treatment of an acute ICU delirium, which may hide a life-threatening neurological emergency such as Wernicke’s encephalopathy. Acute delirium is a commonly encountered problem in the ICU, which has a myriad of causes and contributes to poor outcomes.¹ Its prevalence may reach 80% in medical or surgical intensive care units.¹ Despite the fact that many cases of Wernicke’s encephalopathy have already been published, it may remain unrecognised until the full clinical triad of signs (ophthalmoplegia, confusion and ataxia) help to make the diagnosis. Unfortunately, this represents only 10% of patients with Wernicke’s encephalopathy.² Wernicke’s encephalopathy often follows a treacherous course. So, as reported by Torvik et al, we may be failing to make the diagnosis of Wernicke’s encephalopathy in up to 90% of patients suffering from vitamin B₁ deficiency,³ such as alcohol misusers, malnourished subjects or patients presenting protracted vomiting.⁴ This is a strong argument to advise a high level of suspicion when caring for malnourished, alcoholic or vomiting patients at high risk to develop the disease. From a physiological point of view, Wernicke’s encephalopathy is characterised by an initially reversible brain lesion caused by overwhelming metabolic demands on cells that have previously depleted intracellular thiamine. Thiamine is essential because thiamine-using enzymes participate in breakdown of glucose and carbohydrate metabolism, generating adenosine triphosphate (ATP) and providing energy for the cells.⁵ This explains why Wernicke’s encephalopathy is often precipitated, in alcohol misusers, by the stress of an intercurrent illness or any acute condition placing increased demands on already depleted thiamine stores. The Royal College of Physicians (London, UK) has recently reported guidelines on managing Wernicke’s encephalopathy, emphasising the need for early presumptive diagnosis and systematic vitamin B₁ supplementation in all alcohol abusers admitted to hospital.⁶

As illustrated in this case report, in clinical practice enteral feeding is rarely sufficient to ensure rapid restoration of central nervous system thiamine levels. Even in an experienced centre, only 50% of the enterally fed patients achieved a successful intake despite an enteral feeding protocol.⁷ Discontinuation of enteral feeding often results from diagnostic or therapeutic procedures and from multiple unknown reasons too.⁸ In the present case, as a result of a misplaced ambition to achieve safer patient care, nurses often interrupted enteral feeding during night. So, while the doctors thought that the patient was fed correctly he was actually starving every night.

Fortunately, our patient left the hospital without any persistent memory disturbance. Approximately 80–90% of alcoholics with Wernicke’s encephalopathy treated with inappropriately low parenteral doses of thiamine develop irreversible Korsakoff’s psychosis—a long-lasting debilitating condition characterised by confabulation and short-term memory loss.⁶ Some 25% of them require long-term institutionalisation.⁹ Finally, concerning our patient, appropriate thiamine supplementation has kept the wolf from the door.

LEARNING POINTS

• Acute delirium in alcohol abusers does not correspond necessarily to withdrawal symptoms.

• Enteral feeding is not always achieved as the doctor expects.

• Wernicke’s encephalopathy may develop as a serious hospital-acquired neurological complication in a critically ill patient.

• Thiamine supplementation must be systematically added to the nutrition regimen of all alcoholic patients admitted to hospital to prevent Wernicke’s encephalopathy.