Acute renal failure is a major complication of rhabdomyolysis. New membranes for hemodialysis have been developed with a high cut-off pore size allowing efficient removal of myoglobin. We report on six patients treated by hemodiafiltration with a high cut-off membrane (HCO-HDF) for myoglobinuric acute renal failure. Rhabdomyolysis was caused by infection in two patients, by a statin in one patient and a non-traumatic crush in another, and followed cardiovascular surgery in two others. Ten HCO-HDF procedures were performed. A high cut-off hemofilter was used, with citrate anticoagulation and postdilutional fluid substitution of 2-3 L/h, dialysate flow 500 mL/min, and blood flow within 250-300 mL/min. Albumin losses were replaced by infusion of human albumin solution, and the mean myoglobin reduction ratio was 77% (range, 62-89%). An excellent clearance of 81 mL/min (range 42-131 mL/min) was achieved. Nearly 5 g of myoglobin was removed into the dialysate collected in one of the procedures. A high rebound in serum myoglobin, on average to 244% of the post-procedure myoglobin level, was observed. The four patients alive at the time remained anuric for a week. Slow myoglobin elimination with a mean half-time of 39 h (range 19-59 h) was observed in that period. Highly efficient myoglobin removal by high cut-off membrane hemodiafiltration was demonstrated in our patients. Rapid redistribution from the extracellular fluid and sustained myoglobin release were suggested by the high rebound observed. Elimination of myoglobin within the body was shown in our study to occur slowly during the period of anuria.

Serum creatine kinase (CK) is routinely used as a marker in the assessment of rhabdomyolysis and acute myoglobinuric kidney injury (Mb-AKI), while the use of myoglobin is much less explored in this respect. We retrospectively analyzed the incidence of Mb-AKI (creatinine [Cr] > 200 μmol/L) and the need for hemodialysis (HD) in 484 patients (70.5% males) with suspected rhabdomyolysis, grouped according to peak serum myoglobin (A: 1-5 mg/L, B: 5-15 mg/L, C: >15 mg/L). The median peak myoglobin was 7163 μg/L. Both peak Cr and peak CK were significantly higher in group C. The incidence of Mb-AKI was 24.6% in group A, 38.6% in group B (P < 0.01 vs. group A), and significantly higher (64.9%) in group C (P < 0.001 vs. groups A and B). Fifty-one patients (10.5%) needed HD, the proportion increasing from 6.7% in group A, and 12.3% in group B (NS), to 28.1% in group C (P < 0.001 vs. group A, P 0.01 vs. group B), and reaching 36.8% with myoglobin >20 mg/L. Creatine kinase correlated with the severity of rhabdomyolysis, but less so with Mb-AKI. The peak Cr levels were not significantly different between patients divided by CK 60 μkat/L, or grouped into CK tertiles or quartiles. A significant proportion of patients with rhabdomyolysis experienced Mb-AKI, whose frequency increased in parallel with myoglobin levels. Myoglobin levels above 15 mg/L were most significantly related to the development of AKI and the need for HD. Blood myoglobin could serve as a valuable early predictor and marker of rhabdomyolysis and Mb-AKI.

Of the cases of acute pancreatitis, 1-7% are caused by severe hypertriglyceridemia and can be treated with plasma exchange (PE). We report on a large series of patients with acute hyperlipidemic pancreatitis (HLP) treated with PE. In the 1992-2008 period, 50 patients (45 +/- 8 years old, 92% male) with acute HLP were treated with PE, during which 1-2 plasma volumes were exchanged. Heparin was used as anticoagulant in 85% of the procedures, and citrate in the rest. Cholesterol and triglycerides were measured before and after PE. In the 2003-2008 cohort of 40 patients, we retrospectively recorded an Acute Physiology and Chronic Health Evaluation II (APACHE II) score at the first PE session, hospital mortality, and length of hospital stay. A total of 79 PE treatments were done, 1-5 per patient. The volume exchanged was 4890 +/- 1300 mL over a duration of 3.5 +/- 2 h. During the first PE, the triglycerides were lowered from 58.9 +/- 40.8 to 10.8 +/- 10.8 mmol/L, and the total cholesterol was lowered from 20.0 +/- 7.6 to 5.7 +/- 4.3 mmol/L. In 10% of the procedures the plasmafilter was replaced, and in 3% the filter was clotted. Hypotension occurred in 3% of PE and there was one case of gastrointestinal bleeding after PE with heparin anticoagulation. In the 2003-2008 cohort, the median APACHE II score was 5 (range 0-15), the median overall hospital stay was 18 days (range 3-142 days) and the hospital mortality was 15%. To conclude, in acute hyperlipidemic pancreatitis, one to two plasma exchanges effectively reduce the serum triglyceride level. There is a low rate of procedure-related complications. A mortality rate of 15% is considerable.

In some cases, long-term (>3 months) citrate anticoagulation is needed in maintenance hemodialysis patients due to a persistent bleeding risk. In this retrospective observational study, we present our experience and assess its safety and effects on mineral and bone disorder parameters. Sixteen patients (mean age 67 ± 15 years) were treated with long-term citrate anticoagulation. The indications were: recurrent gastrointestinal bleeding in nine patients, heparin-induced thrombocytopenia, retroperitoneal hematoma, chronic subdural hematoma, proliferative diabetic retinopathy, vascular malformations in the brain in one patient, and others in two patients. Metabolic complications and intact parathyroid hormone (iPTH) were analyzed. Citrate anticoagulation was performed for 4 months to 6.3 years (median 12 months). Ionized calcium was stable during the procedures; hypocalcemia (<0.9 mmol/L) was rare (2.1% of procedures), and there was one case of severe symptomatic hypocalcemia. There were no clinically significant acid-base disturbances and no clotting problems. In the short term (1-3 months after starting citrate), the iPTH increased in 73% of patients (from 325 ± 310 to 591 ± 793 pg/L, P = 0.11, N = 11). In the long term (1-2 years), an increase in iPTH was observed in 3/6 patients. The time period (before/after starting citrate) was a significant predictor of iPTH using main-effects anova (P < 0.001). To conclude, long-term citrate anticoagulation in chronic hemodialysis patients is safe. Mild hypocalcemia during dialysis with citrate anticoagulation may contribute to a short- and long-term increase in iPTH in these patients. Further studies on long-term citrate anticoagulation are necessary.

No abstract