ScottH Posted 20 April 2006 11:30 AM
Undetected Venous Line Needle Dislogement during Hemodialysis was published by ECRI in Health Devices, Nov 1998;27(11):404-6. I have read alerts regarding this issue from at least Gambro(Cobe) and B.Braun Medical and I don’t doubt every other manufacturer has written something as well over the years.
That said, this has been a growing problem along with the trend of increasing blood flow rates and is not a machine dependant hazard, it can occur with any manufacturers machine.
Because the hazard is not neccessarily a simple one to fully appreciate, it’s worth reviewing what pressures are where in the extracorporial circuit. Blood comes from the patient via the arterial needle or port and is measured for pressure pre-blood pump (Arterial). Depending on the machine, there may be another pressure read post-pump before the blood enters the dialyzer. The next blood pressure measurement (Venous) is read after the blood exits the dialyzer but before it enters the venous needle or port and is returned to the patient.
In a static system (blood pump off), the pressures (both arterial and venous) would be nearly 100% a reflection of the patient’s vital pressures. During dialysis however, we have a dynamic system through the extracorporial circuit that contribute multiple factors as a percentage of the total meter pressure, dramatically effecting the composition of both the arterial and venous pressures. These factors include:
Blood pump speed, Size of the needle lumin, Surface drag of the blood tubing material, Compliance of the tubing material, Blood viscosity and Temperature.
Remembering that with the blood pump off, there is little or no effects attributed to the A/V meter pressures other than the patients blood pressure, with increasing blood pump flow rates (Qb) the pt pressures contribution is quickly overwhelmed by these latter factors (particularly the needle lumin size). If the pressures contributed were to be graphed out, one would see not linear positive sloping lines but a set of positive curves and S-shapes, except for the pt pressure contribution which would negatively coorelated to the sum% of all the other factors.
Back in the years when Qb’s were 150 to 250 ml/min, it would not be abnormal to have the patient contribution to be 30 – 40% of the meter pressures. In such a situation, should a venous needle come out, the venous meter pressure might drop as much as 40%. With todays Qb’s hitting twice this, the pt’s pressure contribution could be as low as 05%. Just the tubing compliance paired with the pressure oscillations of a peristalic pump (all blood pumps are currently of this type) can contribute as much as this. That means, unless you want continuous venous pressure alarms while at this Qb, the alarm window would need to be wider (in this case) than the pressure drop from the venous needle disconnecting. Unfortunately, this means that the alarm will probably not trip should this occur.
From an engineering standpoint, this is a difficult problem because of the number of factors that contribute to the meter pressures and that most of them are quite variable under widely different circumstances. To date, no manufacturer has been able to overcome these obsticles. The strong consensus of opinion is that clinical operators must be continuously vigilant in monitoring the extracorporial circuit and that needles are secure and in place at all times, particularly in the case of any blood handling alarm.
Some food for thought… Even if the blood pump was set to a rate where a venous pressure alarm would occur should the needle come out, if the operator resets the blood pump (unaware the needle is out) the new alarm limits will be activated around the new lower pressure and obediantly pump the blood from the patient without the patient’s pressure contribution to set off another alarm and stop the catastrophy.
This last scenario is the one that keeps me up at night.
Please forgive the long post.
NKC Technical Services
Posts: 3 | Location: Seattle | Registered: 19 October 2005