Compiled by Elizabeth Lindley1 based on contributions from Ray James2, Gary Wright3, Elisheva Milo4, Melissa Goench3, Alois Gorke5, Ruth Hyde3, Jackie Ross2, Marisa Pegoraro6, James Tattersall2, Esperanza Vélez7, Richard Ward3, Maurice Harrington2, Diane Thomson8, Anne Murphy9, Gordon Farquhar2, Jean-Yves De Vos10, Susan Hansen3, Gareth Murcutt2, Nic Hoenich2, Freddy Hardy10, Stanley Shaldon11, Hans Polaschegg12, Andre Stragier10, Franta Lopot13, Lenora Perkins3, Rosie Simmonds14, Christine Lansing3, Anthony Page15, John Agar14, Dick Cole2, John Sedgewick2, Lee Fischbach3, Waltraud Küntzle5 and Stephen Sandroni3.
1Department of Renal Medicine, Leeds Teaching Hospitals NHS Trust, UK; 2UK; 3USA; 4Israel; 5Germany; 6Italy; 7Spain; 8The Netherlands; 9Republic of Ireland; 10Belgium; 11Monaco; 12Austria; 13Czech Republic; 14Australia;15New Zealand.
The publication selected for discussion by the EDTNA/ERCA Journal Club in July 2005 was an abstract entitled 'Venous needle dislodgement during hemodialysis: An unresolved risk of catastrophic hemorrhage'. . The author, Dr Stephen Sandroni who is the Director of the Division of Nephrology and Hypertension at Allegheny General Hospital in Pittsburgh, kindly agreed to provide an extended paper for the Journal Club and to take part in the discussion. As you can see from the list of contributors, this paper generated a great deal of interest amongst the Journal Club members.
Synopsis of the discussion paper
The discussion paper can be downloaded in full from the EDTNA/ERCA website (see ref. 1). Dr Sandroni began by describing the case of an elderly gentleman whose venous needle became partially dislodged during a routine haemodialysis session. The machine did not alarm and the blood loss continued until a nurse at the adjacent station saw blood dripping onto the floor. By this time, the patient was unconscious having experienced a 50% blood loss. Resuscitation was unsuccessful.
The unit, which was small and designed so that the patients could be easily seen, was fully staffed with no new trainees present. The treatment was apparently routine and no obvious break with procedure occurred - the needles were taped securely and were plainly visible. The machine was checked but was not defective, yet no alarm sounded to alert the staff because of the limited venous pressure detection capabilities of dialysis machines. A needle can become partially or completely dislodged with resultant copious bleeding without triggering the venous pressure alarm.
The reasons for needles becoming dislodged include patients rotating their arms while asleep, rubbing the needle against their body or the chair with enough force to pull out even a taped needle, patients picking at the tape or needles even while awake and patients, staff members or equipment getting entangled in the blood lines. In most reported episodes there is no observed event that leads to dislodgement of the needle. Blood loss can go unnoticed for some time because it pools initially against the clothed body of the patient, or between the patient and the chair.
The actual incidence of this problem is not known with certainty, as there is a fear of reporting such events, but from informal contacts with other nephrologists, dialysis staff, and manufacturers, Dr Sandroni estimated that the 7-10 deaths reported annually in the U.S. from venous needle dislodgement is perhaps one third or one fourth of the actual total.
Dr Sandroni tried to find out what, if anything, was being done to prevent further tragic and avoidable deaths. He was unable to get the major manufacturers of dialysis equipment to say whether or not they are working on this problem, either because of proprietary secrecy or constraints of pending litigation, and he found no statements on the issue from renal leadership groups or relevant government bodies. From informal discussions, Dr Sandroni found recommendations to use enuresis pads for monitoring of blood leaks; but the pads he and others tried did not alarm when blood was poured on them. Increased frequency of monitoring would help to some extent, but could prove difficult within current staffing constraints.
In other industries where maintenance of flow in a system is critical, devices based on a range of principles have been developed to detect flow or pressure changes. Dr Sandroni suggested that renal care providers need to lobby more aggressively for an engineered solution to the detection of needle dislodgement. This is especially important for home dialysis where the goal for some patients would certainly be dialysis while asleep, alone. One solution would be to use variations of the old single needle techniques, where the machine will detect flow or air.
Experience of needle dislodgement and line disconnections
Needle dislodgements and line disconnections serious consequences are rare events. Ray James, Gary Wright, Elisheva Milo and Melissa Goench (who share at least 80 years experience in haemodialysis) reported having seen a few incidents of this type, but fortunately all were detected before the patients suffered any major complications. Melissa had witnessed needle dislodgements where the blanket stuck to the tape used to secure the lines, where the lines caught on the arm of the chair to which they had been taped and where the patient refused to comply with the taping policy, stating their needles never dislodged. She also knew of a catheter being pulled out as a nurse reclined a patient whose lines were taped to the chair.
Other participants in the discussion had seen incidents where the blood loss was catastrophic, if not fatal. Alois Gorke remembered an incident many years ago where a dislodged needle led to significant blood loss. The venous pressure alarm was not activated, but the fall in the patient's blood pressure eventually set off the arterial pressure alarm. By then, the patient was unconscious but was successfully treated by infusion. Ruth Hyde told us how a patient ended up in the emergency room having lost about third of his blood volume. His venous needle had become dislodged and it was found that the paper tape used did not stick well to his arm, which was very hairy.
Jackie Ross had also experienced a problem with taping in which a patient's lines had been secured to a pillow rather than between his thumb and index finger because he found this uncomfortable. During dialysis, the pillow fell off the bed and pulling the venous needle out. The machine did not alarm, but the patient was able to call the nursing staff. A similar incident was described by Marisa Pegoraro. A patient dialysing in a low-care unit, whose lines were taped to the chair, fell asleep and allowed his arm to slip off the chair so that the venous needle was pulled out. The machine did alarm, but almost too late.
Several participants reported ‘near-misses'. These occurred because staff could not see the blood leak. James Tattersall remembered one incident where severe blood loss from a dislodged venous needle, again without a machine alarm, was not observed because the patient was dialysing in reclining chair with solid arms and a base that hid the floor below the chair from view. The nurses did eventually notice blood pooling on the floor and the patient survived.
Jackie described how a sleeping patient whose needles had been taped securely had somehow managed to become entangled in the lines and pulled out the arterial needle. Although the arterial alarm sounded instantly, the staff, who were attending another patient, could not see what had happened because the patient's arm was under the sheet. When they investigated the alarm a couple of minutes later, they found the lines full of air and blood spurting from the patient's access. Esperanza Vélez remembered an incident in which a very elderly patient who was sleepy, and perhaps unaware what he was doing, was thought to have pulled out his venous needle. The patient was covered by a blanket and the needle was stuck in the fabric which may explain why the alarm did not sound. Staff noticed blood pooling on the floor and the patient survived, but only after receiving 6 units of blood and more than 10 minutes of CPR.
Richard Ward and Maurice Harrington both reported serious incidents where catheters had become disconnected from the blood line. In the incident Maurice described, the patient had a femoral catheter and the connector was covered by a sheet. The machine did not alarm and by the time a nurse observed blood dripping from the bed onto the floor, the patient had lost a significant amount of blood. The patient was confused and may have accidentally made the disconnection.
While Diane Thomson had not seen a needle become dislodged during more than 20 years in dialysis, she had seen leaks from connectors that were badly formed. Diane made the important point that her unit's vigilant reporting of all such incidents had led the manufacturers to modify their production methods and resolve the problems. Anne Murphy recently noticed deformed Luer lock connectors in steam sterilised blood lines. The lines did not lock correctly and there was a fairly large blood leak from the join. This was reported to the manufacturers whose investigation showed that the problem was due to the sterilisation procedure.
All the incidents described above occurred in the supervised setting of the dialysis unit. Venous needle dislodgement is considered to be a significant risk in home haemodialysis (HHD), especially when treatments are carried out overnight. However, there seem to be very few incidents in this patient group. Ray James, Gordon Farquhar, Jean-Yves De Vos and Susan Hansen had no experience of needle dislodgement in their HHD patients. Gareth Murcutt was aware of two disconnections but although the machine did not alarm, the patients realised what had happened and suffered no long-term problems. Anne Murphy remembered a case of severe blood loss and air embolism in a unattended home haemodialysis patient whose bloodlines had come apart in the days before Luer lock connectors were introduced. Fortunately the patient lived for many years following this disturbing episode.
Optimal venous pressure monitoring
Nic Hoenich explained that current haemodialysis equipment meets the requirements of the International Electrotechnical Commission standard 60601-2-16 . To meet the requirements of this standard, the equipment has to provide a system to protect the patient from extracorporeal blood loss to the environment during dialysis. The standard states that this can be venous pressure monitoring, and specifies that the protective system must be activated if the venous pressure falls below 10 mmHg which is below the pressure in the fistula .
During dialysis, the venous pressure alarms are set automatically as a window around the prevailing pressure, but the standard does not give requirements for the size of this alarm window and the manufacturers know that if they make the limits tighter, it is more likely that interruptions to dialysis will occur due to transient fluctuations in pressure.
Nic explained that during dialysis with a blood flow of 400 ml/min, the venous pressure might be 250 mmHg depending upon the type of needle being used. The pressure in the patient's fistula might only be contributing 20 mmHg to the total venous pressure. If the alarm window is set at ± 50 mmHg, dislodgement of the needle will not be picked up by the monitoring system.
Gary Wright recalled reading about a study done with human blood that found the pressure loss during venous needle dislodgement with a 250 to 300 ml/min blood flow was only 20mmHg. He pointed out that patient movement or coughing causes bigger pressure changes than that.
Freddy Hardy noted that when running at low pressures (where the low pressure alarm limit is the +10 mmHg required by the standard) there may be enough pressure in the circuit to stop the alarm triggering. Ray James confirmed this. His unit had found, after some experimentation, that there could be a venous pressure of over 10mmHg even when the line was disconnected from the needle.
Nic and Ray both felt that reducing the width of the alarm window and increasing the minimum preset from 10 to 20mmHg would make detection of a needle dislodgement or line disconnection more likely. But more accurate detectors and/or an alarm management algorithm would be needed to avoid alarms being generated by ‘normal' pressure fluctuations.
Gareth Murcutt suggested offsetting the venous pressure alarm window to +50/-20 mmHg and giving a longer initial override time for low venous pressure alarms to avoid nuisance alarms, as the pressure drop that would occur with needle dislodgement would persist. Another option would be to develop software that would allow the machine to ‘learn' to identify the normal fluctuations for an individual patient.
Gareth also pointed out that the Operator's Manual for the machines in his unit states that "it is important that the lower venous alarm limit is set to be as close as possible to the actual value of the venous pressure". However, his machines default to ±60 mmHg and if the alarm window is widened to ±100 mmHg for one patient, it does not return to the default for the next session unless the power is switched off. In busy units, Gareth thought it was unrealistic to expect staff not to run machines in the available mode that causes the fewest alarms and interruptions, even if this mode does not offer the best protection for the patient.
Stanley Shaldon was shocked that there could be unnecessary deaths from venous haemorrhage in 2005, when these did not occur 40 years ago with much more primitive monitoring systems. He felt that the use of wide venous pressure windows to avoid false alarms, and ‘high pressure' treatments with 16 gauge needles and high blood flows are making dialysis unnecessarily risky. Professor Shaldon considered the manufacturers' standards of safety described by Nic to be completely unacceptable. After all, he said, high performance car manufacturers are not allowed to get away with it - in my opinion high speed dialysis today is like driving a Ferrari F40 without brakes.
Prof Shaldon, who was probably the first to perform home HD with AV fistulae back in 1967  felt that home HD patients were often more obsessive in securing needles than staff. His advice to professionals would be to reduce the intrinsic resistance of the venous needle by using larger bore needles and keeping the blood flow below 300 ml/min. The low venous pressure alarm limit can then be set closer to the actual pressure so that a sensitive pressure monitoring device can detect the drop in pressure caused by needle dislodgement. It would be tragic if failure to make use of a simple, proven and safe monitoring system meant that patients were not permitted to dialyse at home without an attendant.
More sophisticated technical solutions
Hans Polaschegg surprised us by explaining that both Gambro and Fresenius had patented technology to detect venous needle dislodgement in the 1990's. The Gambro patent was filed first . It uses the pressure pulses produced by the peristaltic blood pump. These are transmitted to the venous pressure monitor ‘forwards' through the extracorporeal circuit or back through the arterial needle, the patient's access and the venous needle. If the venous needle slips out, the component of the signal transmitted via the access will disappear. This can be detected by analysis of the signal from the venous pressure monitor with no extra hardware. Dr Polaschegg confirmed that the method would work, at least with an air free, low compliance system, but he stressed that a protective system of this kind does not prevent blood loss, it can only prevent excessive blood loss.
The Gambro patent also described how the patient's own pulse could be used as a signal but Dr Polaschegg's measurements showed that this method is not sensitive enough at high blood flows. The sensitivity could be improved with more sophisticated software and, if it worked, it would also allow continuous recording of the patient's heart rate.
The wording of the Gambro patent does not explicitly include pressure pulses generated on the dialysate side. This allowed Fresenius to file a patent covering any pressure pulses produced by the balancing chambers or any other device on the dialysate side . Again the signal from the venous pressure chamber is monitored. If the venous needle slips out, the pressure wave reflected from the free end of the circuit will cause changes in the signal.
Dr Polaschegg felt that these inventions had not been implemented because the IEC standard for HD machines accepts standard venous pressure monitoring as a protective system against blood loss to the environment, even though there have been many reported incidents where the venous pressure alarm has failed to operate. With the current standard, the machine manufacturers marketing people will not support a change.
Ray James had concerns about the sensitivity of the patented methods due to the small pressure fluctuations involved, but he thought they had the potential to detect blocked or wet isolators and clotting in the extracorporeal circuit, all of which give rise to incorrect (static) pressure readings. Dr Polaschegg confirmed that pressure pulses can be used to detect wetting of the transducer protector. This could help prevent cross infection due to defective transducer protectors. In fact, he said, the technology to do this was patented and published many years ago but it had never been implemented, although it would only require simple software modifications.
Andre Stragier thought that there was a problem with any security system that is needed extremely rarely. Imagine a busy renal unit with 25 dialysis stations, he said. It might be expected that needle or catheter dislodgement will occur only once every 15 years but meanwhile perhaps 1000 false alarms or function check failures will have occurred. This will definitely upset both the nursing and the technical team, and the companies implementing the system would be blamed. By the time a real life-saving alarm is needed, it is possible that either the device will no longer be operational or the team will assume they have another false alarm
As an alternative or an adjunct to venous pressure monitoring, Franta Lopot and Gareth Murcutt wondered if information from the blood volume monitors built into many modern machines could be used to help identify catastrophic blood loss. An unusual steep increase in apparent blood volume would be expected due to plasma refilling and loss of erythrocytes. Jackie Ross thought that if the patients were attached to an oximeter that constantly monitors blood oxygen saturation and pulse, the rising pulse and falling oxygen saturation in a massive haemorrhage would cause the oximeter to alarm, alerting staff to the problem.
Lenora Perkins informed the group that she has developed a 'venous needle dislodgement sensor'. It is going through the US patent process and she hopes to find a manufacturer interested in implementing it in their machine.
External moisture detectors
The suggestion that enuresis pads which emit an audible alarm when wet could be a practical method for detecting haemorrhage was first made by Nic Hoenich. Gary Wright thought a portable, external, easily positioned device that detects blood, but does not alarm if the patient sweats, would give a more timely alarm than any protective system in the machine. It would have to work reliably and without producing nuisance alarms that cause staff to stop using it.
Rosie Simmonds seems to have found exactly what Gary described. Her unit in Geelong (Victoria, Australia) has been running a nocturnal home haemodialysis programme for the past 4 years  and now has over 20 patients dialysing for 8-9 hours, 5-6 nights a week in their own home while they sleep, including a couple of patients who dialyse alone. The Geelong programme is based on that of the Humber River Regional Hospital in Toronto  and following Toronto's lead, they included the use of moisture detectors at the AV fistula site when they started the programme
The device used in Geelong is the DRI Sleeper from AnzAcare Ltd . The staff advise patients to position the moisture detector between the needle sites, closer to the venous needle. For some patients they use a ‘velcro' sleeve that covers the needle sites and secures the lines and DRI-sleeper.
Rosie explained that they have continued to use DRI Sleepers for a number of reasons, the main one being the added safety and security that it provides for the overnight procedure. The patients and staff trust that the detector will quickly alert the patient if the needle becomes dislodged, or if there is leakage around the puncture sites, and find this very reassuring. Rosie has found an added bonus of using the moisture detector for people with diabetes as it will pick up excessive sweating and has alerted 2 patients to hypoglycaemic episodes that they were experiencing during sleep.
The DRI Sleeper positioned close to the venous needle.
Although the Geelong patients have not experienced needle dislodgement during nocturnal HHD, there have been cases of quite severe leaking from around needle sites. The moisture detectors detected this insidious loss, which would not be picked up by the machine, and alerted/woke the patients.
Christine Lansing confirmed that the DRI Sleeper, unlike the devices tested by Dr Sandroni, does respond to blood. She has had patients wake up for quite small needle leaks. Christine has found that wrapping the moisture detector in gauze (5 x 5 cm) prevents it being activated by sweat. Patients have to check the batteries and ensure that the alarm is not against clothing that might mute the sound.
The developer of the DRI Sleeper, Dr Anthony Page, told us that while Geelong were one of the early pioneers in using this device as a fistula alarm, around 25 hospitals in Australia as well as units in the USA and Canada, were also using the DRI Sleeper for this purpose. Alpha Consultants are now working with the team in Geelong and a technology development company in the UK to produce a purpose built sensor. The time scale for this project is around one year.
Susan Hansen thought it was probably not feasible for a dialysis machine to detect a blood leak external to its sensors, and that this was where an external blood sensitive device would be useful. Ideally the blood pump should be stopped in the event of needle dislodgement and, as Ray James pointed out, this would mean that the device would have to be attached to the patient and simultaneously electrically connected to the dialysis machine.
Single needle dialysis
Andre Stragier noted that, in principle, conventional single needle dialysis protects against massive blood loss due to needle dislodgement because the air alarm will be activated after the machine has pumped through the volume of the extracorporeal circuit up to the venous level detector.
Dr Sandroni's unit had no experience with the single needle mode as a home treatment at the time of the discussion but were soon to be trialing the new Allient machine, a sorbent-based dialysis system manufactured by Renal Solutions . The new machine will be used for long treatments in single needle mode. Instead of the usual peristaltic pumps, the Allient has a blood pump that consists of two chambers and a system of 4 valves. In two-needle dialysis, one chamber will be filling with blood from the patient, while the other pumps blood through the dialyser and back to the patient. For single needle operation, both chambers are filled with blood from the patient in the arterial phase, they then both empty, pumping the blood to dialyser and returning it to the patient. Unlike conventional systems, the Allient has an air detector in the arterial line so it should cut down the time required to detect a disconnection and stop blood loss.
John Agar from Geelong felt that single needle nocturnal HHD could be the way forward provided the noise problems associated with switching between pumping blood from and to the patient were managed as a quiet system is essential for this treatment. A dual lumen needle would solve the noise problem as it would allow continuous flow and the inevitable recirculation is acceptable with long nightly dialysis. Professor Agar thought he remembered that Quinton Inc made a dual lumen needle in the 1970's or 80's but he'd been unable to source any suitable devices on the market now.
Dick Cole, the UK-based European Representative for Renal Solutions, said that the noise level is the same for single needle and two needle mode in the Allient. In service mode with the covers open, the primary noise is from the pneumatic pumps, while the clamps make a very low level sound. In normal use, the door of the pump housing muffles most of the sound. Dick also said that a contract engineer had been employed to address noise and ensure that the machine can run in either mode without compromising the patient's ability to fall asleep.
Problems with Luer connectors
A number of incidents described by participants were due to accidental separation of the Luer lock connector between the needle tubing or catheter and the blood lines. Richard Ward explained that it was thought that some Luer lock connectors could be prone to separation because of torsional energy stored in the tubing by the twisting action used to make the connection. Separation will be more likely if the catheter is immobilised with only a short length of tubing between the point of immobilization and the connector.
Gareth Murcutt agreed that there was a problem with the security of the Luer lock connector. He thought that if garden hoses, which are often subject to torsional forces, were attached with Luer lock technology, they would come undone far more often than with the common hose-lock type connector. If the widely used Luer connector itself cannot be changed for logistic reasons, perhaps a simple safety guard could be developed?
Maurice Harrington also agreed that improvement in the Luer lock design was overdue. He was not convinced monitoring fluctuations in venous pressure would be a sensitive enough method to detect the disconnection of a catheter. One possible improvement might be some physical feedback to tell the user that it has been tightened sufficiently.
Jean-Yves De Vos commented that there are already special catheter connectors for use in dialysis (the Tego Connector ref 11), but the design of these connectors means they restrict blood flow. Jean-Yves wondered if it would be better to modify the Luer lock connectors fixed to the needle tubing with the silicone membrane as used in the Tego device so there would be no extra flow restriction and no additional connection.
Securing needles and lines
Poor or inappropriate taping, particularly taping the lines to the chair or pillow, was the most common reason for needle dislodgement in the incidents described by the contributors. Alois Gorke, Ruth Hyde, Marisa Pegoraro and Melissa Goench all had experience of such incidents.
When the needle dislodgement occurred in Alois's unit, they used 2.5 cm wide silk woven tape from rolls. Strips of tape were usually placed crosswise over the wings of the needle. Cotton swabs were often placed under the wings to fix the needle in a better position although this reduced the contact between the skin and tape. The tubing was usually fixed between thumb and index finger by a strip of tape over the tubing. When investigating the dislodgement incident, the staff found that the tape was not sticking properly to the skin on warm days or when the patient was sweating, and that the needles were not fully anchored behind the wings.
Alois's unit introduced the following procedure:
- Place a split sterile dressing (7.5 x 9 cm) over the needle with the full area stuck to the skin
- Place one or two strips of hypoallergenic, non-woven tape (with good ability to stick on skin) under the needle tubing to anchor the needle
- Place cotton swabs under the needle tubing ONLY if better placement is needed
- Use one or two strips of tape to fix the blood tubing to the wrist of the patient. The tape needs to surround the tubing fully, ideally it should be stuck back to back and then to the skin). This fixation has to resist accidental pulls and pulls by confused or muddled patients
- Check the position of needles and tapes during every round
"Fistula needles secured using the procedure developed in Alois Gorke's unit"
Ruth's unit also changed policy following the incident she described. They now put paper tape over the needle and secure it with a chevron or a loop, then place another strip of paper tape over the top of this. If the arm is hairy, or the tape is not sticking well, they wrap the tape around the arm and they no longer use folded tabs to make it easier to remove the tape. A 10 x 10 cm dressing is taped around the wrist like a bracelet to secure the lines. For upper arm access, the lines are clipped to the patient's clothes at the shoulder area. Ruth's unit also has a policy of taping the Luer connections to be sure they are secure.
Marisa identified two problems when reviewing practice after the incident in her unit: a tendency to follow patient's choice rather than a protocol, and the variable quality of the tape supplied by the hospital. Melissa pointed out that it is essential to tape the needles with anchors and a loop to prevent the needles from moving, and that clamping the bloodlines to the patient (not their environment) allows for patient mobility with minimal disturbance to the treatment.
Alois made the point that, although it is critical to patient safety, there is little research on secure taping procedures and materials in the renal nursing field. In fact, pictures in dialysis text books, journals and in slides shown at conferences often show insufficient and hygienically questionable taping.
John Sedgewick, the chair of the EDTNA/ERCA Education Board, noted that there is a need for clear, evidence based guidelines associated with the taping of needles. In addition to security, it would be necessary to look at infection rates, patient comfort and dialysis efficiency.
Monitoring patients during dialysis
Lee Fischbach felt that manufacturers will never claim that their machines can monitor for blood loss and it will remain the responsibility of the nursing staff to monitor for leaks. Andre Stragier confirmed that the US-based health equipment watchdog, ECRI, agreed with the manufacturers. In 1998 ECRI recommended that dialysis staff should be advised to "continually examine hemodialysis bloodlines during treatment if this is not already routine" and to "keep the entire venous line from being covered by anything that might prevent good visualization of the needle insertion so that it can be easily monitored". 
Several participants thought that there were drawbacks to introducing more and more advanced technology to protect the patient. Ray James and Elisheva Milo both felt that making machines do more seemed to lead to the nursing staff becoming less aware of what to look out for. As Elisheva said, technology cannot take the place of the nurse's eyes. Andy Johnson agreed saying that no matter what automation and monitoring we design into the equipment, there is no replacement for the most powerful computer of all - the one between our ears. Andy also made the point that but staff must be willing to question the technology and use their observations to confirm the accuracy of protective systems. Pressure monitors do not give reliable readings when the circuit is clotted or the transducer protector is wet.
Waltraud Küntzle pointed out that there must be sufficient trained staff to monitor patients and the surrounding area closely. Haemodialysis, she said, is really an intensive care treatment in an out-patient setting with all risks of an extracorporeal circulation, blood loss and electrolyte disturbances. Elisheva added that it was important that nurses knew how to identify patients who are at a greater risk of needle dislodgement or line disconnection.
Inability of staff to see the blood leak was a factor in several of the incidents described earlier. The importance of ensuring that the patient's access is not covered and can be easily seen by the nurses was emphasised by Freddy Hardy, Melissa Goench, Ruth Hyde and Esperanza Vélez,
Melissa and Ruth also stressed the need for repeated checks of all connections, prior to treatment and during dialysis. Jackie Ross described the procedure followed in her unit to prevent incidents such as needle dislodgement and haemolysis. Before commencing dialysis, the initiating nurse checks connections, lines and prescription. Immediately after the patient has commenced dialysis, a second nurse checks the patient, needles, tapes, connections, lines and prescription.
Needle dislodgements and line disconnections are rare and are usually detected quickly, but the discussion showed that renal care professionals are keen to minimise the risk
The suggested technical solutions ranged from better management of the existing venous pressure monitoring systems to the introduction of patented methods that detect needle dislodgement using signal analysis and Luer lock connectors that will not separate due to torsional force from the tubing.
For nocturnal HHD, the DRI Sleeper monitor appears to provide an effective way to alert a patient to blood loss and we heard that a purpose built fistula monitor based on this device is under development. Single needle dialysis could provide a safer alternative for long nightly treatment if noise levels can be reduced.
Many contributors felt that while technical advances would be welcome, most incidents could be prevented by education, secure taping and vigilance. It is important for dialysis units to adhere to protocols for checking taping and connections, and for ensuring that the patient's access is not covered. There is a need for good practice guidelines for securing needles and lines.
Use the dialysis machine's venous pressure monitoring system effectively. If you are concerned about the flexibility of the alarm settings on your machines, tell your company representative.
It is essential for staff to understand the importance of securing and monitoring needles, catheters and bloodlines.
We are very grateful to Dr Stephen Sandroni for highlighting the problem of venous needle dislodgement and participating in the discussion and to all the journal club members who contributed their thoughts and ideas.
We would also like to thank Lee Fischbach for circulating the technical comments to the International Electrotechnical Commission working group responsible for safety of dialysis equipment, and John Sedgewick for taking a proposal to develop guidelines for securing and monitoring needles and lines to the EDTNA/ERCA Education Board.