EDITOR:

Fluid overload and hyponatraemia are central to the development of the TransUrethral Resection of Prostate (TURP) syndrome after transurethral prostate surgery [Reference Norris, Aasheim, Sherrard and Tremann1]. Hyperammonaemia and hyperglycinaemia can also be part of this syndrome [Reference Roesh, Stoelting and Lingeman2,Reference Norten, Allgen, Vinnars and Bedrelidou-Classon3]. Hyperammonaemia or hyperglycinaemia usually presents with central nervous system (CNS) depression although hyperglycinaemia exhibits a CNS excitatory action through its positive action on N-methyl D-aspartate (NMDA) receptors [Reference Kish, Dixon and Burnham4]. We describe a patient who developed CNS excitation following TURP surgery possibly because of hyperglycinaemia and responded to propofol sedation.

A 72-yr-old male weighing 65 kg and with a history of controlled hypertension for 15 yr underwent TURP surgery under a subarachnoid block. His medication (diltiazem 30 mg and metoprolol 50 mg each once daily) was continued till the day of surgery. He fasted overnight and was premedicated with alprazolam 0.25 mg on the night before and the morning of the day of surgery. On examination, his heart rate was 64 beats min⁻¹ and blood pressure 140/70 mmHg. He had a normal electrocardiogram (ECG) and haematology, biochemistry and coagulation profiles were within the normal limits. Transthoracic echocardiogram revealed an ejection fraction of 65% with normal pulmonary arterial pressure and elevated left ventricular filling pressure.

Following preloading with 500 mL of lactated Ringer’s solution, a subarachnoid block was performed with 2 mL of 0.5% hyperbaric bupivacaine and 25 μg fentanyl administered through a 25-G Quincke needle at the L3–L4 intervertibral space in the left lateral position, which produced a block to T_8. Surgery was started and 1.5% glycine was used for irrigation. The surgery lasted for 1.5 h and 8 L of fluid were used for irrigation intraoperatively. After the surgery, the patient was transferred to the post-anaesthesia care unit. His vital parameters were stable and he was conscious and oriented.

His initial postoperative course was uneventful. After 6 h, however, he became restless and apprehensive. He was initially given sedation with incremental doses of intravenous (i.v.) midazolam (5 mg in total) without effect. Pain was thought to be the cause, hence incremental i.v. morphine (6 mg in total) was administered also without effect. He did not have chest pain and the ECG was normal. His pupils reacted to light and were of normal size bilaterally; visual field examination was not possible because of agitation. Postoperative serum electrolyte showed marginally lowered serum sodium (133.7 mmol) and arterial blood gas analysis results were within the normal ranges. Meanwhile, the patient became more agitated and started to shout and talk irrelevantly. I.v. haloperidol 5 mg and midazolam 6 mg (in incremental manner) were given, but still the patient continued to be restless, agitated and started using abusive language. It was then noticed that mistakenly the surgeon had continued glycine as the irrigating fluid in the postoperative period, and hence the irrigation was changed to normal saline. During these 7 postoperative hours, 12 L of glycine had been used for irrigation (a total of 20 L in the whole perioperative period). We could not measure the serum glycine level; however, the serum ammonia level was found to be 91 μmol L⁻¹ (normal 5–50 μmol L⁻¹). A diagnosis of glycine toxicity was considered as the cause of the violent behaviour. The patient was given a bolus of 30 mg propofol followed by infusion of 25 μg kg⁻¹ min⁻¹. This produced an immediate response with progressive decrease in his agitation. The infusion was increased gradually over a period of 15 min to a maximum of 40 μg kg⁻¹ min⁻¹, which was continued for a further 3 h. He slept overnight and the next morning he had no agitation nor a recall of the events from the previous night. The serum ammonia level fell to 13 μmol L⁻¹ by the morning.

Although TURP syndrome lacks a stereotypical presentation, typically it produces central nervous system and cardiovascular system manifestations (altered mentation, tachypnoea, hypoxaemia, pulmonary oedema, convulsions and coma) because of hyponatraemia and fluid overload arising from excessive absorption of the irrigation fluid [Reference Norris, Aasheim, Sherrard and Tremann1]. Our patient did not have any clinical features of fluid overload and his serum sodium was not so low so as to produce features of classic TURP syndrome. The features of ammonia toxicity start developing at 500 μmol L⁻¹ [Reference Roesh, Stoelting and Lingeman2], so that in our patients it was too low to produce any symptoms of ammonia toxicity. Moreover, ammonia toxicity produces nausea, altered sensorium, drowsiness, lethargy and decreased alertness. Our patient was exhibiting violent behavior unresponsive to routine sedation.

The raised ammonia level, however, indicates that there was absorption of glycine. This may then be metabolized by the kidney and liver to produce glyoxylic acid and ammonia. Patients who do not have glycine as the irrigating fluid do not develop hyperammonaemia after prostate surgery [Reference Hamilton Stewart and Barlow5]. If the absorption of glycine is of prolonged duration it may accumulate although the serum ammonia level may not rise at least for 12 h [Reference Fahey6].

In the situation of continued glycine irrigation and high serum ammonia level, we presume that our patient was developing glycine toxicity symptoms, which include vomiting, nausea, headache, weakness and malaise through its inhibitory neurotransmitter activities. However, glycine may also act as an excitatory neurotransmitter on NMDA receptors, leading to the development of TURP encephalopathy [Reference Kish, Dixon and Burnham4]. We believe that our patient was developing early features of glycine encephalopathy. Midazolam and haloperidol do not act on NMDA receptors, therefore they failed to reverse such excitatory symptoms even with repeated administration. Propofol, however, is an NMDA receptor antagonist with effects on glycine receptors [Reference Dong and Xu7] and we believe that sedation with propofol was achieved not by its action on gamma-aminobutyric acid (GABA) receptors but with its effect on NMDA receptors. Therefore, we suggest that propofol should be considered in the management of excitatory manifestations of glycine toxicity following TURP surgery.