Lesson of the Week: Cardiac arrest due to severe hyperkalaemia in patient taking nabumetone and low salt diet

Case report

This 70 year old woman had suffered from seropositive, erosive rheumatoid arthritis for about 17 years, mainly managed with non-steroidal anti-inflammatory drugs and simple analgesics. She developed hypertension about three years before presentation but had not been taking any regular antihypertensive treatment. Six years previously she had been investigated for haematuria but no lesions were found in the urinary tract. Recently she had been reassessed after a flare-up of her arthritis, and her naproxen was changed to nabumetone 500 mg twice daily. She had been taking it for one month when she was admitted for further assessment of her rheumatoid disease. Mild renal insufficiency was found, with a serum creatinine concentration of 148 μmol/l and a potassium concentration of 5.7 mmol/l. Dipstick urine analysis showed protein and blood, and urine culture showed coliform organisms sensitive to trimethoprim. She was kept on nabumetone and no new drugs were introduced, except for trimethoprim at the time of her discharge.

A week later she suffered a cardiac arrest at her general practitioner's surgery. Immediate resuscitation was started and continued in the ambulance to the casualty department. Electrocardiography showed ventricular tachycardia, and she eventually reverted to sinus tachycardia. At the time of her resuscitation she was found to have a serum potassium concentration of 9.7 mmol/l with impaired renal function as shown by the following results: sodium 138 mmol/l, urea 18.0 mmol/l, creatinine 280 μmol/l, and bicarbonate 9 mmol/l. One hour after resuscitation she was awake and talking but the serum potassium concentration was still raised at 8.3 mmol/l. She was treated with further intravenous insulin and dextrose, calcium chloride, sodium bicarbonate, and 250 mg of frusemide. Urine output increased to 200 ml/h and serum potassium concentration fell progressively. Her subsequent progress was satisfactory and the serum potassium concentration returned to near normal values (K=5.4 mmol/l). Some renal impairment remained (serum creatinine 148 μmol/l).

After her acute episode was over further questioning revealed that her general practitioner had advised her to start a low salt diet for her hypertension and as a result she had been using a low sodium formulation (Lo Salt) for cooking, and also taking extra orange juice as a vitamin C supplement. As she had felt progressively less well and anorexic in the week before her cardiac arrest she found she was able to eat only bananas.

Discussion

We do not consider that nabumetone (administered and excreted in an inactive form but converted in the liver to the active metabolite) was the main cause of our patient's dangerous level of hyperkalaemia, but it may have played a part.1 Indeed, this patient probably had intrinsic renal problems, as evidenced by her history of haematuria and biochemical renal insufficiency which may well have been associated with her hypertension or rheumatoid disease. Patients with rheumatoid arthritis have an increased prevalence of renal disorders,2 often caused or aggravated by their drugs.3 4 5 6 7 Personal communication with the manufacturers revealed that there was a single report made to the Committee on Safety of Medicines of hyperkalaemia associated with nabumetone in the United Kingdom.

The important point here is that this patient's hyperkalaemia and renal insufficiency were further aggravated by her low sodium diet which contained abundant potassium (estimated to exceed 100 mmol/day (see table)), and her increased risk for hyperkalaemia because of renal impairment further compounded by the intercurrent urinary tract infection. In this patient serious problems arose from a combination of her mild renal insufficiency associated with hypertension, the urinary tract infection, her drug treatment and unwitting intake of a high quantity of potassium salts in her diet, as well as the low salt (but high potassium) preparation.

Our case raises the issue of potassium supplementation deliberately for health gain (often by individuals without medical consultation or advice), or, inadvertently, because people are not aware of the potassium content of health products or salt substitutes and the consequent dangers associated with unrestricted consumption.8 Potassium supplementation is particularly risky in susceptible patients with underlying diseases of the heart or kidneys; those on certain drugs, such as potassium sparing diuretics, angiotensin converting enzyme inhibitors, and non-steroidal anti-inflammatory drugs; and also the elderly.

A number of “low” salt products available with brand names such as “Lo Salt,” contain varying proportions of potassium salts in addition to sodium chloride. “Lo Salt,” for example, contains 66% of potassium chloride. Low salt preparations are often recommended for patients with oedema and may have a place in the management of hypertension. Doctors who recommend low salt diets need to be aware that many of these patients may also be taking high amounts of potassium. The problem is compounded by the fact that potassium overloading may go unnoticed for a long time as hyperkalaemia has few or only vague clinical warning symptoms, and serum potassium concentrations may rise to dangerous concentrations before presentation, as happened in this patient. This event in our patient was almost fatal, and it could readily recur unless a regular check is kept on prescribed drugs, as well as over the counter medicines and products from health shops.

Doctors should thus be careful about their advice on low salt diets, especially for those high risk subjects with underlying cardiac, hepatic, or renal insufficiency, those taking non-steroidal anti-inflammatory drugs, and elderly people. Manufacturers should also clearly state the risks of their preparations in susceptible subjects: the warning should appear prominently on the label.

We thank Mrs Yvonne Bradburn, dietitian and Mrs Joan Kay for typing the manuscript.