The phosphorus connection--a puzzling business

Abstract

A 40 year-old woman was admitted because of profound weakness for the past 4 days. She also claimed to have a severe headache for which she had ingested large doses ( 8 g) of aspirin. She steadfastly denied nausea or vomiting and did not complain of abdominal pain. Past medical history revealed generous alcohol consumption and a cardiomyopathy, ostensibly alcohol induced. She had intermittent atrial fibrillation for which she received phenprocoumon; however, her most recent ejection fraction was 48%. On physical examination, her temperature was 38 C, her heart rate was regular at 84 beats/min, her respiratory rate was 20/min, and her blood pressure was 80/60mmHg. The blood pressure did not fall with upright posture. The neck veins were not distended, the chest was clear, the heart was not appreciably enlarged, and the abdomen was not tender. Bowel sounds were diminished but present. The liver was enlarged at 3 cm below the right costalmargin.Therewasnoperipheral oedema. Her haemoglobin was 11.4 g/dl, and the haematocrit was 36 vol%. The leukocyte count was 11 400/mm, the INR was 8.4, and C-reactive protein (CRP) was 280mg/l. Plasma sodium was 132, chloride 87, potassium 5.5, magnesium 1.49, total calcium 1.49 and phosphorus 7.76 (all mmol/l). Plasma creatinine was 1290 mmol/l, urea 66mmol/l, lipase 131U/l, amylase 94U/l, aspartate aminotransferase U/l, alanine aminotransferase 42U/l, g-glutamyl transferase 150U/l, total protein 63 g/l and albumin 23 g/l. The urine showed granular casts, erythrocytes and leukocytes with protein þ1 without ketones. The urine osmolality was 335mOsm/kg H2O, Na <10mmol/l, Cl <10mmol/l and K 44mmol/l. The urine creatinine was 17mmol/l and urea 112mmol/l. The fractional excretion of Na was <1% and that of urea was 13%. Arterial blood disclosed pH 7.24, PaCO2 29mmHg, PaO2 105 mmHg, HCO3 12mmol/l and lactate 0.75mmol/l. Toxicological studies revealed a plasma salicylate level of 49mg/dl, while ethanol, methanol and ethylene glycol were undetectable. The electrocardiogram was unremarkable with the exception of a prolonged QT interval attributable to a lengthened ST segment (Figure 1, top). An abdominal ultrasound disclosed an enlarged liver of increased density. The kidneys were normal in size without obstruction. No free abdominal fluid was detected. We then measured creatinine kinase and myoglobin. Both were only slightly elevated. Plasma parathyroid hormone and 25-OH-vitamin D levels were within the low-normal range. We were confronted with a patient who presumably had pre-renal failure, anion gap metabolic acidosis, hypocalcaemia and profound hyperphosphataemia of unknown cause. We did not believe that her salicylate level explained the anion gap, the calcium or the phosphate levels. At the time, we felt we could not rule out a tumour-lysis syndrome and therefore acute haemodialysis coupled with volume expansion was performed. Three dialyses were done on three consecutive days. Thereafter, she had gained 5 kg in weight, her calcium increased to 2.31 and her phosphorus decreased to 1.1mmol/l. Her electrocardiogram (Figure 1, bottom) now showed a normal QT interval. Her CRP decreased to 145mg/l. She continued to deny abdominal pain. In the process of pursuing the idea of an occult neoplasm, a magnetic resonance imaging (MRI) study of the abdomen was carried out (Figure 2). Because of the patient’s acute renal failure, we did not perform computed tomography (CT) with contrast agents. To our considerable irritation and surprise, the MRI scan showed clear evidence of acute pancreatitis with secondary haemorrhage. The aetiology of the hypocalcaemia, the hyperphosphataemia and the pre-renal azotaemia suddenly became clear. Less obvious was why her acute pancreatitis failed to present with pain and enzyme elevations. We initially thought that a massive acute phosphate load might have caused hyperphosphataemia in our patient. Since exogenous causes, such as phosphate-containing enemas, could Correspondence and offprint requests to: Andreas Birkenfeld, Franz Volhard Clinic, Wiltberg Strasse 50, D-13125 Berlin, Germany. Email: birkenfeld@fik.charite-buch.de