Polyuria following an overdose.

Case report detailing the occurrence of diabetes insipidus in a 42-year-old man admitted to the intensive treatment unit (ITU) following an overdose. Whilst on ITU, he was sedated with propofol. Cessation of treatment with propofol coincided with resolution of the polyuria. Animal studies suggest a theoretical mechanism for propofol as the causative agent, but this phenomenon is not commonly seen in humans.

INTRODUCTION

This is an unusual case of a 42-year-old gentleman, admitted to the intensive treatment unit (ITU) following an attempted suicide. He was observed to develop significant polyuria following admission to ITU and this polyuria resolved when propofol was stopped with no further treatment. This is a rare occurrence of this side effect with this drug, considering how frequently it is used. It may be the case that it occurs more frequently than is known about at present since it may be under-reported, which is why we feel this case report and the discussion it may stimulate are important.

CASE REPORT

A 42-year-old man with a known psychiatric history was admitted in the morning to ITU for respiratory and circulatory support following a mixed overdose. He had taken 50 mg of ramipril, 150 mg of olanzapine, and 200 mg of citalopram. He had taken no alcohol, paracetamol, aspirin, or recreational drugs and had not suffered any head injury.

On ITU he was sedated with 120 mg/h infusion of propofol 1%.

In the afternoon, he was noted to be polyuric with a urine output greater than 500 ml/h and his urine was noted to be progressively more dilute [Figure 1].

Figure 1

Catheter bag showing progressively more dilute urine since admission

Urine and serum osmolarity were measured. Urine osmolarity was 160 mosmol/L (normal range 300–1000), serum osmolarity was 304 mosmol/L (normal range 280–300), demonstrating inappropriately dilute urine production. His blood glucose level was normal.

He was given intravenous fluid to replace urinary losses and was monitored. The following morning, sedation was stopped and he was extubated successfully. Shortly thereafter, urine concentration returned to normal.

DISCUSSION

Medline was used to perform searches with combinations of the term “diabetes insipidus” and the drugs taken in overdose. There were no results found for a Medline search of “ramipril” and “diabetes insipidus” or “citalopram” and “diabetes insipidus.”

The search “diabetes insipidus” and “olanzapine” gave three results: one case report detailing the case of a 17-year-old who had taken an overdose of olanzapine and developed cranial diabetes insipidus whilst in ITU;[1] a case report of a patient treated for diabetes insipidus secondary to lithium in a patient also taking olanzapine;[2] and a pilot study of the use of olanzapine in a palliative care setting in which one of the 24 patients developed diabetes insipidus.[3]

Olanzapine is well absorbed via the oral route and its half-life is approximately 33 h,[4] making it highly unlikely to be the culprit in our case where the polyuria resolved completely less than 36 h following ingestion of the overdose.

On reviewing the notes, it was found that the resolution of the polyuria seemed to occur shortly after sedation with propofol was stopped. A further search of “diabetes insipidus” and “propofol” was performed. This found one case report linking the use of propofol as an anesthetic with diabetes insipidus.[5]

The mechanism of action of propofol (2, 6-diisopropylphenol) as an anesthetic is not fully understood,[6] but it enhances GABA (gamma-aminobutyric acid)-mediated inhibition of ADH (anti-diuretic hormone/arginine vasopressin) release in rats,[7] which is a potential mechanism for transient neurogenic diabetes insipidus in humans. The onset of the action of propofol is 30 s and its action is short-lived. The case report linking olanzapine and diabetes insipidus does not state what sedation was used whilst the patient was on ITU.[1]

CONCLUSION

This case report shows that it is possible for propofol to induce polyuria in humans as it does in animal studies. Why this does not occur more frequently, given how commonly propofol is used in the ITU setting, is unclear and should be the subject of further investigation.

It is possible that the use of other drugs in conjunction with propofol may have led to the under-recognition of the association with propofol.

Propofol is often used for the induction of anesthesia rather than maintenance, so polyuria may be less pronounced with smaller doses, attributed to other drugs or to the administration of IV fluids intra-operatively.