Presentation: A 66 year old woman, was admitted as an emergency with one week history of profound weakness ('had to crawl to the fridge to get milk for tea'), lethargy, productive cough and hypokalaemia (GP results: Na 141, K 1.5, bicarbonate 47 (23-28), Creatinine 54 (μmol/l). Her medication: Amlodipine 5 mg od, Bendrofluazide 2.5 mg od, Aspirin 75 mg.
PMH: hypertension, moderate COPD and peripheral vascular disease. She smoked 20 cigarettes a day and drank 7 units of alcohol per week. Examination: Pulse 90/min, BP 158/66; JVP normal, moderate ankle swelling, decreased breath sounds consistent with COPD, generalised muscle weakness (4/5) with diminished, but symmetrical reflexes, plantars flexor.
Investigations: ECG: sinus rhythm, non-specific T-wave abnormalities, Na 141, K 1.7, bicarb 46, urea 3.9, creat 57, glucose 5.8, Calcium 2.11, Magnesium 1.10 (0.7- 0.91 mmol/l), Phosphate 0.79, albumin 31, Alk.Phosph 99, ALT 110, AST 374, bilirubin 22, CK: 17624 (CK-MB normal), free T4 24.3 (10-30 pmol/l), free T3 3.9 (2.5-5.3pmol/l), TSH 2.55 μiu/ml. Hb 15.2, WCC 8.8, Platelets 373. MSU: normal, urinary myoglobin: negative.
Clinical Impressions: We assumed that Bendrofluazide alone was unlikely to lead to such a profound hypokalaemia, in view of heavy smoking the possibility of ectopic ACTH-secreting bronchial cancer was raised. Chest-X-ray however was normal, 24-hour urinary free cortisol was 255 nmol/24hr (40-305), ESR 14 mm/hr. Further tests revealed low plasma renin activity (PRA), low aldosterone (recumbent PRA 0.4 (1.1-2.7 pmol/ml/hr), aldosterone 75 pmol/l (100-450)), as well as 24-hour urinary aldosterone of 2 nmol/24hr (10-50). Normal US abdomen. Following discontinuation of Bendrofluazide and addition of Potassium supplements, on 6th day post-admission, the results were as follows: Na 141, K 3.1, urea 2.9, creat 50, bicarb 36, CK 1482, AST 80.The cause of her profound hypokalaemia at presentation became more clear when she admitted consuming at least a packet of 'Barnips' throat sweats a day (probably more during episodes of chest infections). We eventually obtained a sample of these and established that they contained 3.8% oil of liquorice (~250 mg of deglycyrrhizinised liquorice extract/pack of 'Barnips' sweets).
Further progress: She was told to avoid excessive consumption of 'Barnips' tablets and was discharged on Aspirin, Amlodipine and Slow K 1200 mg tds. Three months post-discharge her Na was 137, K 5.3, bicarb 24, creatinine 58. Potassium supplements were stopped, U&E-s repeated one month later were normal.
Conclusions: Though liquorice preparations are no longer used in the treatment of peptic ulcer disease, there are reports that in susceptible individuals hypokalaemia can develop after chronic ingestion of about 500 mg of liquorice/day. We therefore concluded that in our patient such profound and life-threatening hypokalaemia was caused by chronic liquorice ingestion in the form of 'Barnips' cough sweets, in the setting of concomitant treatment with a thiazide diuretic.