The effect of circuit "down-time" on uraemic control during continuous veno-venous haemofiltration.

OBJECTIVE The term continuous veno-venous haemofiltration (CVVH) suggests a treatment without interruption. However, interruptions do occur and the duration of the haemofiltration circuit "down-time" may influence uraemic control. We conducted a prospective study to ascertain the percentage of operative "down-time" for CVVH in our intensive care unit and to test the hypothesis that it significantly affected uraemic control. PATIENTS AND METHODS Prospective data measuring the time spent off the filter in ten patients receiving CVVH were collected. Continuous veno-venous haemofiltration was performed at 2 litres per hour of ultrafiltration. Anticoagulation was maintained using unfractionated heparin administered pre-filter and infused at a rate to achieve a systemic APTT varying between 30-45 seconds. The circuit functional life was documented for each CVVH circuit as progressive cumulative hours of operation. The time off treatment was calculated for each 24-hour period. These data were then correlated with the change in plasma urea and creatinine concentrations for each 24-hour cycle. The APTT, INR, haemoglobin and platelet count were measured and levels were correlated with the filter duration. RESULTS Ninety three days of CVVH treatment were assessed in 4 female and 6 male patients. The mean circuit "down-time" in these patients for this period was 22% or 5.27 hours per day. The most common cause of circuit "down-time" was circuit clotting, followed by a need for radiological procedures, time spent in the operating theatre and catheter malfunction requiring replacement. There was a strong correlation between circuit "down-time" and increase in plasma urea (p = 0.0017) and creatinine (p = 0.0451) concentrations. Circuit "down-time" was also inversely correlated with the platelet count (p = 0.0048) but not significantly correlated with the APTT, INR or haemoglobin values. CONCLUSIONS In our study the average daily duration of an interruption in CVVH (i.e. circuit "down-time") represented > 20% of the potential operative time. There was a strong correlation between time without treatment and solute control during CVVH. The percentage of "down-time" may be a useful marker of operative quality during CVVH.