William G. Herrington

Metformin: the safest hypoglycaemic agent in chronic kidney disease?

Metformin is the first-line oral agent in the treatment of type 2 diabetes and has many established benefits, including the reduction of macrovascular complications of diabetes. Its prescription in patients with renal impairment is limited by concerns relating to the theoretical risk of lactic acidosis, a fear which is perpetuated by numerous case reports in which it is implicated. Critical review of this literature calls into question the validity of these claims, with metformin usually acting as an ‘innocent bystander’ in acutely unwell patients with conditions well recognised to precipitate lactic acidosis such as sepsis or hypovolaemia. In fact, the evidence supports the safe use of appropriate doses of metformin in patients with chronic stable renal impairment, and highlights the important possible greater risks of the alternatives, most notably severe hypoglycaemia in patients taking sulphonylureas and/or insulin and fluid retention in patients taking a thiazolidinedione. Other traditional contraindications to metformin use such as heart failure are also being re-evaluated, as the benefits of metformin in these patients are increasingly recognised. Physicians should weigh this evidence carefully before deciding to withdraw metformin therapy in their patients with stable chronic kidney disease.

Are arterial and venous samples clinically equivalent for the estimation of pH, serum bicarbonate and potassium concentration in critically ill patients?

Diabet. Med. 29, 32–35 (2012)

Association of long-term administration of the survivin mRNA-targeted antisense oligonucleotide LY2181308 with reversible kidney injury in a patient with metastatic melanoma.

A 57-year-old man with metastatic melanoma was treated with the survivin inhibitor and antisense oligonucleotide LY2181308 as part of a First-in-Human Dose trial. After 18 months of treatment, he developed kidney injury and the treatment was discontinued. At 9 months and before the development of kidney injury, LY2181308 concentrations were 8- to 10-fold higher relative to median predicted values, but within the targeted exposure considered to be safe. However, at 17 months, 28 days after stopping LY2181308 therapy, LY2181308 concentration exceeded the predicted range by 38-fold. His decreased kidney function was slow to improve after stopping treatment. A kidney biopsy showed signs of acute tubular injury with regeneration. Complete recovery of kidney function occurred 6 months after treatment was stopped. The relationship between high exposures and slow LY2181308 clearance with the gradual improvement in kidney function after stopping the antisense treatment suggests that the oligonucleotide was related to the kidney injury. Based on this case report, kidney function should be monitored frequently in patients receiving long-term treatment with antisense oligonucleotides that specifically target survivin, particularly when they receive concomitant angiotensin-converting enzyme inhibitors or nonsteroidal anti-inflammatory drugs.

Metformin use in chronic kidney disease: new evidence to guide dosing.

Metformin controls blood glucose, does not cause hypoglycaemia or weight gain and has been shown to reduce the long-term complications of diabetes, including macrovascular disease.1 Since metformin is renally cleared, there is a genuine risk of metformin accumulation and associated lactic acidosis in chronic kidney disease (CKD). We have previously argued that there is disproportionate fear surrounding the safety of metformin in CKD, with important side effects from alternative agents ignored. Both the risk of hypoglycaemia with sulphonylureas or insulin and risk of heart failure with thiazoladinediones are expected to increase as CKD progresses, and are at least as large as the risk of lactic acidosis.2 Moreover, the mortality from metformin-induced (i.e. pure type B) lactic acidosis appears substantially smaller than the 50% observed with other more common causes of lactic acidosis.3nnNevertheless, the current UK’s National Institute for Clinical Excellence guidelines advise that metformin should be reviewed if the serum creatinine exceeds 130 µmol/l [or estimated GFR (eGFR) is below 45 ml/min/1.73 m2], and stopped when creatinine rises above 150 µmol/l (or eGFR falls below 30 ml/min/1.73 m2).4 Dosing recommendations from other bodies also advise a renal function threshold after which metformin becomes unsafe, but agreement is rare,5 with most not providing …

Neutrophil gelatinase-associated lipocalin in kidney transplantation: A review

Neutrophil gelatinase-associated lipocalin (NGAL) is a protein expressed by kidney tubular cells in response to ischemia, but may also be an early indicator of immunological rejection, calcineurin inhibitor toxicity, obstructive nephropathy, subclinical tubulitis or infection. Although there is currently no evidence to support the routine serial measurement of blood or urinary NGAL to detect subclinical acute tubular injury, NGAL has the potential to provide useful information to those that care for kidney transplant recipients (KTRs). First, high urinary or serum NGAL concentrations shortly after transplantation are a predictor of delayed graft function and are associated with reduced graft function at one year. Secondly, among KTRs with previously stable graft function who then suffer acute graft dysfunction, a high urinary NGAL predicts graft loss at one year. If further refined, diagnostic tests based on NGAL levels may provide future useful clinical tools.

Does prophylactic anticoagulation reduce the risk of femoral tunneled dialysis catheter-related complications

Purpose To determine the incidence and predictors of femoral tunneled dialysis catheter (TDC)-related complications and whether prophylactic anticoagulation is associated with reduced catheter-related deep vein thrombosis (CRT) or prolonged patency. Methods A retrospective review of femoral TDCs inserted for maintenance hemodialysis in patients from two dialysis units that have used two different strategies to reduce thrombotic complications. One center routinely considered all femoral TDCs for prophylactic anticoagulation, whilst the other restricted anticoagulation to TDCs that had required repeated treatment with urokinase locks to maintain patency. Survival analyses were performed to establish complication rates, identify predictors of complications and assess the effect of prophylactic anticoagulation use. Results Of the 194 femoral TDCs identified, 178 (92%) were associated with at least one complication. Approximately three quarters did not provide adequate small solute clearance; one half were not in use by three months; one quarter had at least one catheter-related infection (2.3 per 1000 catheter days); and one in ten developed a CRT (1.1 per 1000 catheter days). Prophylactic anticoagulation was not associated with significant improvements in rates of catheter occlusion, CRT, catheter-related infection or dialysis adequacy. A previous ipsilateral femoral TDC was identified as a statistically significant predictor of a CRT (adjusted hazard ratio 3.7 [95% confidence interval 1.4-9.8]; P=.007). Conclusions Femoral TDCs are associated with poor patency rates and high complication rates; reusing femoral veins for TDCs should be avoided where possible, and this study provides no evidence to support routine prophylactic anticoagulation in all patients with femoral TDCs.

Kidney failure with a diagnostic chest radiograph

An 87 year old woman presented with breathlessness, which was worse on exertion and when she was lying flat. She had a longstanding history of hypertension, hypercholesterolaemia, and cigarette smoking. She had been referred to a nephrologist two years earlier and was found to have a creatinine of 270 μmol/L, with 2+ proteinuria on urine dipstick, a normal albumin (45 g/L), and no oedema. She was lost to follow-up before investigations were complete. On re-presentation, she described new shoulder and lower back pain. Her pulse was 80 beats/min and regular, blood pressure was 130/60 mm Hg, and jugular venous pressure was raised with bibasal inspiratory crepitations and peripheral oedema. Urine dipstick showed 3+ proteinuria and a trace of blood (recent protein:creatinine ratio 850 mg/mmol; reference value <15 mg/mmol). Her urea was 35 mmol/L (2.5-6.7), creatinine 357 μmol/L (54-145), albumin 32 g/L (35-50), haemoglobin 80 g/L (120-150; normocytic picture), and kidney bipolar length was 10 cm on both sides. Serum and urine protein electrophoresis was performed and immunoglobulins, antineutrophil cytoplasmic antibodies, antineutrophil antibodies, and complement components were measured. After a week of diuretics her urea and creatinine increased to 53 mmol/L and 601 μmol/L, respectively, with ongoing oedema. She started haemodialysis. A chest radiograph obtained at re-presentation is shown in fig 1⇓.nnnnFig 1 Chest radiograph

Contents Vol. 118, 2011

Clinical Nephrology Guidelines J. Cunningham, London G. Eknoyan, Houston, Tex. A. Khwaja, Sheffield Clinical Appraisal/Evidence Based Nephrology A.K. El-Sherif, Ismailia R.J. Glassock, Laguna Niguel, Calif. A. Meyrier, Paris Global CKD G. Remuzzi, Bergamo N. Perico, Bergamo R. Atkins, Melbourne, Vic. Clinical Trials D. de Zeeuw, Groningen F. Locatelli, Lecco D. Wheeler, London Continuing Nephrology Education R. Barsoum, Cairo M. Field, Sydney, N.S.W. C. Zoccali, Reggio Calabria Clinico-Pathological Conferences T.H. Jafar, Karachi Editor-in-Chief

The hyponatraemic hairdresser: highlighting the differentials

In May, 2008, an 18-year-old trainee hairdresser presented to her general practitioner with a 2-month history of weight-loss and polyuria. In response to comments from her peers about her weight, she had partially starved herself during the preceding week and consumed large volumes of soft drinks. There was concern about an eating disorder and blood tests were taken. However, before the results were available, she became confused and developed tonic-clonic seizures at home. She was intubated in the emergency department and transferred to the intensive care unit. Cranial imaging and lumbar puncture were normal, but her serum sodium was 99 mmol/L, creatinine 780 μmol/L, and creatine kinase 88 860 U/L. Paired serum and urine osmolalities were both low at 264 mOsmol/L and 242 mOsmol/L, respectively, with an inappropriately high urine sodium of 58 mmol/L. With continuous intra venous rehydration her sodium improved to 124 mmol/L, but her creatinine reached a plateau at 380 μmol/L. 3 weeks after admission, she was well and was discharged home. At her fi rst nephrological review 1 week later she was dehydrated with postural hypotension, a serum sodium of 121 mmol/L and creatinine of 653 μmol/L. There was no medical or family history of kidney disease. She was underweight, with a body-mass index of 18 kg/m2. Her adrenal and thyroid function were normal. She had a normochromic anaemia, hyperphosphataemia, hyperparathyroidism, and low grade proteinuria. Her kidneys measured 8–9 cm on ultrasonography and were highly echogenic. Paired osmolalities again showed a dilute urine relative to plasma with a high urine sodium content. Despite rehydration there was no improvement in her blood or urine chemistry and haemodialysis was started. The diff erential diagnosis included resolving acute tubular damage from rhabdomyolysis or chronic renal failure with renal salt-wasting. Renal biopsy was consistent with nephronophthisis–part of the medullary cystic kidney disease complex (fi gure). There was no evidence of acute damage secondary to rhabdomyolysis. Her condition was stabilised on haemodialysis, and she was managed as an outpatient. When last seen in September, 2009, although on haemodialysis, her residual renal function means she continues to waste sodium; her serum sodium is between 125–130 mmol/L. Nephronophthisis is the most common genetic cause of childhood end-stage renal failure. It can be caused by mutations in various genes including the NPHP genes which encode proteins involved in the function of cilia in renal tubules. Diff erent mutations lead to presentations at diff erent ages from infancy to adolescence. Progressive renal tubular cystic degeneration causes chronic renal failure and renal salt-wasting, often without hypertension. Our patient does not have mutations in her NPHP1 or NPHP3 genes so is likely to have a mutation in one or more of the many causative genes. Her initial episode of polydipsia on a background of chronic salt-wasting resulted in the presentation of extreme hyponatraemia causing seizures and rhabdomyolysis. Of note, eating disorders have been associated with chronic tubulo interstitial nephropathies and decreased concentrating ability. The syndrome of inappropriate anti-diuretic hormone secretion can also lead to low plasma osmolality and hyponatraemia, but urine osmolality is typically higher than plasma osmolality in this situtation. Renal salt-wasting is an important and easily overlooked cause of hyponatraemia; extreme hyponatraemia—in the absence of diuretic therapy—is uncommon, even with an eating disorder. A salt-wasting nephropathy is an important diff erential diagnosis in this context. The treatment is fl uid and salt supplementation rather than fl uid restriction. A high urinary sodium (>20 mmol/L) is an easy screening investigation for a salt-wasting nephropathy.