unsuitable for the creation of an arterio-venous fistula (AVF) or implantable graft. A tunnelled dialysis catheter (Doublelumen PermCath) was inserted and haemodialysis commenced. Two months later he was admitted with pyrexia, rigors and a marked inflammatory response. He also complained of pain at the left sterno-clavicular joint. Repeated blood cultures, from separate sites, grew Staphylococcus epidermidis. Spiral CT scanning demonstrated evidence of osteomyelitis and clavicular destruction. The catheter was removed and peritoneal dialysis commenced. Antibiotic therapy with vancomycin and rifampicin was continued for a 6 week period with resolution of symptoms and inflammatory markers.Case 2. A 62-year-old Yemeni man presented to our emergency department with ESRF. Owing to symptomatic uraemia a tunnelled catheter was inserted and dialysis commenced. Three months later the patient developed high fevers, with raised inflammatory markers. Repeated blood cultures were negative. Despite replacement of the line and empirical antibiotic therapy he developed a shallow 4 cm ulcer at the catheter exit site. The catheter was removed, and the ulcer biopsied. The culture grew fully sensitive Mycobacterium tuberculosis. Clinical and radiological assessment found no other sites of tuberculosis. After initiating treatment and replacing the catheter, the ulcer healed and the patient improved.

Pharmacokinetic (PK) studies of oxaliplatin, using a dose regimen of 85 mg/m, are lacking. A PK model may be used in future studies to investigate the relationship between pharmacokinetics and dose limiting toxicity. The purpose of this study was to construct a population PK model to describe platinum (Pt) concentrations in plasma in 33 patients with colorectal cancer. The secondary objective was to determine the relationship between the amount of Pt in 24-hour urine and the amount of Pt in fractionated urine collection periods. Plasma and urine samples were collected from patients during their first oxaliplatin treatment course. Population PK analysis was performed with WinNonMix. The model that best described the Pt concentrations in plasma was a two-compartment PK model. The elimination clearance (CL) and the elimination clearance of the peripheral compartment (CL2) (median +/- SE) were 25.2 +/- 6.3 L/hr and 68 +/- 24.8 L/hr, respectively. The median volume of distribution (V1) was determined to be 41.6 +/- 9.4 L and the median volume of distribution of the peripheral compartment (V2) was 452.5 +/- 96.4 L. The relationship between the cumulative amount of Pt in urine in the first 12 hours compared with the amount of Pt in 24 hours urine was reflected by a correlation coefficient (r2) of 0.95. The cumulative Pt concentration in urine in the first 10 hours and the first 8 hours compared with 24 hours was reflected by correlation coefficients r2 = 0.93 and r2 = 0.897, respectively. This PK model could be useful in identifying predictors for PK and pharmacodynamic variability to individualize dosing. The results of this study suggest that fractionated urine samples can replace 24-hour urine collection.