Medical device incident investigations, November 2003
The TGA has received a report of a fatality associated with the use of a Graseby MS-16 syringe driver distributed by Smiths Medical Australasia. A patient was using the syringe driver for delivery of analgesia at the time of her death. The cause of death was found to be overdose. TGA tests concluded that the device was working within specifications, but highlighted that the device is not tamper-resistant and that the bolus dosing button can be depressed repeatedly without restriction or adequate alarm. At least two other similar events have occurred in Australia. In each case there was evidence to suggest that a degree of user error or tampering was involved.
Users should be aware that some older syringe drivers like the MS-16 and MS-26 do not incorporate the safety features found in current, state-of-the-art pumps and drivers. Also it is important that once these devices are set up and the infusion started, it must not be tampered with or adjusted by anybody other than someone who has been correctly trained in the use of the device.
Graseby has designed a "lock-box" specifically for use with the Graseby MS-16 and MS-26 syringe drivers. The lock-box prevents unauthorised access to the syringe and bolus facility during an infusion.
- Ensure users and carers (including patients and relatives in the home-care setting) are aware of the purpose and use of the device.
- All users must be fully trained in the use of the device.
- Provide instructions to patients, relatives or other carers to not interfere with the device once the infusion has commenced.
- Consider fitting the lock-box in all cases where the likelihood of tampering or over infusion is judged to be high and/or the consequences of such an occurrence is severe. For example, the lock box should be considered where the device is to be used in home care unsupervised.
The TGA received a report from a NSW hospital stating that sticking had inhibited the mechanism in one of their Kerrison Cervical Rongeurs. When the instrument was dismantled in order to fix the problem, debris was found on the sliding surface. Similar debris was found in several of their rongeurs. The instruments were not designed to be dismantled for cleaning.
Kerrison rongeurs of this design have been widely used for many years. This is the first time that this sort of incident has been reported to the TGA for this sort of device. However, this may simply be due to the fact that it was seen and never reported, or that it was occurring often and not noticed.
The sponsor* for this device sells a rongeur that is dismountable and is therefore easier to clean. It is preferable for all reusable instruments to be easy to clean. However, in this sort of situation it is largely up to the hospitals to decide what the best alternative in instruments may be, and this choice should include consideration on whether the hospital has the expertise and resources to adequately clean the instruments.
In this case, the sponsor of the instrument has sent out a Safety Alert (14/4/2003) to all their customers reminding them of the cleaning procedures for the instruments, and advising them that dismantlable rongeurs are available. In their letter the sponsor also offered to supply dismantlable rongeurs at a discount price to those customers who had purchased non-dismantlable rongeurs in the past.
* The name of the sponsor has been omitted on purpose - The TGA believes that this issue is generic and naming specific devices or suppliers would detract from the general message.
It is important to ensure that reusable instruments are thoroughly cleaned before sterilisation. The sterility of an instrument that is less than "squeaky-clean" cannot be guaranteed, regardless of the rigour of the sterilisation process. Sometimes, the design of the device makes it very difficult for it to be adequately cleaned and this should be considered before the instrument is purchased/re-used.
The issue of cleanability of reusable instruments generally is currently being addressed by a Therapeutic Goods Committee working group, which includes TGA, State Health Infection Control and Industry representatives.
Please continue to report this type of issue to both the TGA and the sponsor or manufacturer. These reports can lead to improvements in both products and cleaning procedures.
Two near-incidents have been reported by a hospital involving Micro Mist Aerosol Nebulisers manufactured by Hudson RCI, USA. The incidents involved connection of the nebuliser connector directly to a standard endotracheal tube connector. If these two devices are placed together, an almost totally airtight system is formed (apart from oxygen being administered via the nebuliser), thus not allowing the patient to exhale or sufficiently inhale resulting in physiological difficulties and subsequent serious adverse effects.
This incorrect connection is possible because the inner diameter of the nebuliser connector accepts the outer diameter of an endotracheal tube connector.
It should be noted that human error was a contributing factor in the incidents that occurred in the hospital. The manufacturer has had only one previous incident of this nature in the tens of millions of nebulisers produced for worldwide distribution. However, the manufacturer has acknowledged the potential for serious injury if the nebuliser is used in this manner.
As a result the sponsor, in conjunction with the TGA, issued a Safety Alert in which hospital staff are warned about the possibility of a misconnection. In addition, the manufacturer has changed the design of the nebuliser, making it physically impossible for it to be connected to a standard endotracheal tube.
The manufacturer has altered the design of the connector so that misconnection is less likely. Stocks with the new design will be in Australia late this year. However, stocks of the old version of the connector will take some time to completely deplete from hospital stocks.
In the meantime, check that you have received the Safety Alert about the possible misconnection of endotracheal tubes to Micro Mist Nebulisers and bring this to the attention of all staff responsible for the care of patients who may need to use both types of device.
In consultation with the TGA, Datex-Ohmeda issued a Safety Alert regarding the maintenance of Datex-Ohmeda Calibration Gas Regulators AUSTL No.76213
|755530||(with a rotary control valve)|
|755533||(with a rotary switch, now obsolete product, superseded by 755534)|
|755534||(with a push button)|
Datex-Ohmeda has identified a few calibration gas regulators that have become faulty over a period of time in a way that the customer might not readily detect. It is possible that over time very small aluminium and rubber particles enter the inside of the regulator resulting in it becoming partially blocked. The aluminium and the rubber particles both originate from the calibration gas cans.
Using a partially-blocked calibration gas regulator will result in an incorrect monitor calibration.
The consequence is that the monitor readings will be too high compared to the actual gas concentration. However, Datex-Ohmeda has not received any such complaints in Australia or New Zealand.
In February 2002, Datex-Ohmeda introduced new annual maintenance instructions concerning the calibration gas regulators. Regulators supplied to customers after February 2002 had the new instructions included. These new instructions are applicable to all regulators currently in use, including those that were supplied prior to February 2002. Please ensure that maintenance staff follow the new maintenance instructions.
For technical support, please contact Datex-Ohmeda's National Service and Support Centre on 1300 322 229.
Potential for disconnection exists when non-locking intravenous set or accessory connections are used
The problem of disconnection with luer slip connectors on intravenous lines is still a serious concern. The potential exists for disconnection when non-locking or luer slip connectors are used. There is the possibility of serious patient injury or death if the intravenous line disconnects due to blood loss, air embolus or interruption of critical therapy. This is particularly true when luer slip connectors are used on lines that have a high pressure, such as arterial lines, or those lines that feed directly into the central circulatory system, such as central venous catheters.
The use of luer slip connections on IV lines that connect to central venous lines, arterial lines or other critical applications where accidental disconnection could result in serious patient injury or death is discouraged.
All add-on devices should use a luer lock system. Tape should not be used to secure add-on lines.
The TGA has received a report from the Bioengineering Division of the Department of Medical Physics at Royal Perth Hospital regarding difficulties that may be encountered while inserting the ceramic acetabular liner into the shell of the Trident system. A laboratory study conducted by the RPAH unit concluded that under certain conditions the shell could incur a diametral deflection due principally to its "thin wall" design.
The Trident Acetabular System is a two-piece component system that is assembled during surgery. The alumina in the insert is completely covered by titanium backing. The ceramic inserts gain fixation within the shell by means of mating tapers and rotational stability is achieved by means of a series of interlocking barbs and scallops.
The role of the titanium backing is to protect the alumina insert by completely supporting all the outside surface of the insert and by providing a protective lip so that the alumina rim is not chipped by the neck of the femoral stem at large angular deflections of the hip.
The shells are designed to maximise fixation in the peripheral lunate region. The dual sphere and terraced geometry of the shell allows more of the load to be preferentially distributed to the denser bone in the lunate region because the rim is nominally 1.8 mm larger than the diameter of the shell. This results in interference fit at the rim of the acetabulum, which reduces the risk of stress shielding and cup protrusion. A key point to this report is that, by reaming the acetabulum to the nominal diameter of the shell, the surgeon is actually "under-reaming" by approximately 1.8 mm.
Because the acetabular shell is relatively thin, the shell can become distorted during intra-operative impaction of the implant into dense hard sclerotic bone, so much so that the liner will become difficult or impossible to put in. The report received by the TGA related to difficulties experienced with alumina inserts. However the same shells can also accept polyethylene inserts. Polyethylene inserts clip into place into a channel inside the shell. Polyethylene is more compliant than the equivalent alumina inserts and can be "made to fit" even if the shell had been distorted during insertion. The consequences of such intra-operative distortion when polyethylene liners are used are impossible to assess at this stage.
The surgical instructions included the following words:
"1.8 mm of interference fit is not always necessary when dense hard bone is encountered. Over reaming by 1 mm will lead to an interference of slightly less than 1 mm, while still achieving a secure fit." James A. D'Antonio, M.D.
Over the past 2 years, Stryker Australia has received 3 complaints of the ceramic insert being difficult or unable to be inserted in the presence of hard sclerotic bone and, although in each case there were no subsequent recurrences, in light of those experiences, Stryker has agreed to amend the surgical instructions to include the following recommendation.
Reaming progresses in 1mm increments until final sizing is achieved (1-2mm under size of cup for the Trident Hemispherical; line-to-line reaming for the Trident PSL HA) Surgical judgment is used to assess bone stock, amount of interference, and proper amount of under-reaming as desired. Note the Trident PSL HA Acetabular Shell contains a 1.8mm peripheral press fit built into the shell as marked (e.g. 52mm = 53.8mm). When implanting the Trident PSL HA shell, 1.8mm of interference fit is not always necessary when dense, hard, sclerotic bone is encountered. In this situation it is recommended to over-ream by 1mm, thus leading to interference fit of slightly less than 1mm. This can prevent potential problems that may typically occur in dense bone such as acetabular fracture, failure to fully seat the implant, or slight deformation of the titanium shell, making seating of the insert more difficult.
The amended surgical instructions will be available very soon. Ensure that you receive these instructions and any previous versions are discarded and/or destroyed.
Use caution when inserting the Trident shell. Over reaming by 1 mm is recommended when dense, hard, sclerotic bone is encountered. This will still achieve a secure fit. In any case some care during impaction is required to avoid distortion of the relatively thin acetabular shell.
Device incident reports 01/04/2003 to 30/06/2003
Number received: 105
|Not Device Related||11|
|Hospital Supply Service||5|
|TGA News article||4|
|No Further Action||42|
|Problem Not Confirmed||3|
|Refer to GMP||9|
- The problem causes are not mutually exclusive. For example, a material deficiency may have led to a mechanical malfunction or a biocompatibility problem.
- Every report received by IRIS receives a risk analysis by the Scheme Coordinator and is discussed by a panel of technical and clinical professionals. In the case of reports that are "Not Investigated" the panel has made a decision that further investigation of the particular event is not necessary at that time. However, these reports are logged into the database for future reference and the trend of reports is monitored. In making their decision, the panel considers whether any similar reports have been received previously.