Dear Mr Oliver; write with reference t0 your letter of August 2016. The Medicines and Healthcare products Regulatory Agency (MHRA) is an Executive Agency of the Department of Health: The aim of the MHRA Devices Division is to take all reasonable steps to protect public health and safeguard the interests of patients and users by ensuring that medical devices meet appropriate standards of safety, quality and performance and that they comply with the relevant Directives of the European Union: The MHRA liaises with manufacturers to ensure that their products are as safe a5 reasonably possible, and that all information necessary to allow the device to be used correctly and safely is provided. Where required, we also take appropriate action with manufacturers and issue advice to users as & result of our investigations into adverse incidents_ Synopsis of incident detalls Mr Matthews died from an infective exacerbation of Chronic Obstructive Pulmonary Disease on 19 February 2016_ He was receiving oxygen at the time of his death, with an Oxylitre flowmeter used to regulate and measure oxygen flowrate. The flowmeter was later found to be damaged: forensic post mortem autopsy concluded that the reduced oxygen flow emanating from the damaged flowmeter did not significantly contribute t0 his death. On the 1 August 2016, you contacted the MHRA requesting that we look into whether MHRA action could be taken to prevent future deaths. Your main concerns regarded a lack of clarity and completeness of guidance relating to pre-use checks and general maintenance of flowmeters_ The

Medicines and Healthcare Products Regulatory Agency Device Description The Oxylitre F1602 flowmeter is a device used to measure the flow rate of therapeutic oxygen administered to patients. It is designed to measure oxygen flow rate from 0 _ 15 Imin. The device comprises of a chrome-plated brass body with BS (British Standard) probe and an outer tube used to keep the unit "gas tight" (appendix 1). Manufacturer's investigation On 12 April 2016 MHRA received an adverse incident report from Head of Non-Clinical Risk Management at Darlington Memorial Hospital (Hollyhurst Road, Darlington DL3 GHX): The hospital's report stated "Crack in flowmeter tube causing Oz leak of indeterminate volume: Device was reported as 'unable to turn off" Following receipt of the incident report received on 13/04/16, MHRA requested that Oxylitre investigate the event On 9 August 2016 MHRA were advised by Oxylitre that they had inspected and tested the F1602 flowmeter to determine whether oxygen flow had been compromised Oxylitre noted that a noticeable crack was visible in the outer tube (appendix 2). Upon testing, the manufacturer identified that when plugged into a gas source the flowmeter supplied oxygen at the maximum flow rate (15 Ilmin): Therefore, the manufacturer's analysis concluded that the device was still capable of supplying sufficient oxygen to the patient: MHRA did not witness the manufacturers examination of the F1602 flowmeter; but has reviewed the manufacturer's report (appendix 3). The particular unit in question is approximately 15 years old (exact date of manufacture cannot be established as serial number has been partially removed appendix 2) The manufacturer has confirmed that the device had exceeded the recommended service life of 10 years. According to your report, a review by the clinical engineering department of Oxylitre flowmeters in use at Darlington Memorial Hospital found that a further 2 devices available for use were cracked: MHRA analysis of instructions for use (IFU) Oxylitre provided MHRA with a copy of the IFU for the F16O2 flowmeter (appendix 4). The IFU states "No leaks are permissible on the device" , and advises the user t0 visually check the device for cracks before use. A user who identifies a crack should discard the product prior to patient use. It would appear that the hospital had not perfored pre-use checks in line with the manufacturer's IFU. The IFU states that; only qualified servicing personnel should perform tests to detect leaks, therefore implying that additional specific instructions referring to pre-use leak tests are unnecessary: We have undertaken a comparative assessment of various manufacturers' IFUs for similar devices to determine whether there are any insufficiencies in the pre-use checks description for the Oxylitre flowmeter: The information collected indicates that the Oxylitre IFU provides sufficient guidance for the user to perfomm pre-use checks and is in line with alternative products. Although the IFU appears sufficient; we have forwarded your concerns regarding clarity completeness of guidance to Oxylitre for their consideration in their next revision: Additional Guidance General guidance can be found in the MHRA's publication Managing Medical Devices (appendix 5). The purpose of this document is to outline a systematic approach to the acquisition, deployment, maintenance, repair and disposal of all medical devices. It is intended primarily for people in hospital and community based organisations who are responsible for the management of reusable medical devices to help them promote safe effective use of medical devices. document, which can be applied to oxygen flowmeters, states that a healthcare organisation's medical device management policy must cover the provision of maintenance and repair of all and and The

Medicines and Healthcare Products Regulatory Agency medical devices, including reconditioning and refurbishment: The healthcare organisation is responsible for ensuring their medical devices are maintained appropriately: Similar reports received by MHRA Oxylitre have confirmed that no similar incidents or complaints regarding the F1602 flowmeter have been reported to them. A review of the MHRA's adverse incident database has shown that since 2006 we have received no adverse incident reports involving cracked flowmeters Conciusion The MHRA received an adverse incident report concerning a cracked flowmeter which was in use by a patient who subsequently died, The manufacturer's investigation identified that the cracked flowmeter was still capable of supplying oxygen at the maximum flowrate_ The manufacturer of the oxygen flowmeter is fully compliant with the Medical Device Directive 93/42 EEC, and the device was functional despite exceeding the manufacturer's recommended service life. The MHRA has not identified a systemic problem with cracks associated to Oxylitre flowmeter: The manufacturer's Instructions for Use appear sufficient including details of pre-use inspection: MHRA has brought the Coroners concerns to the attention of the manufacturer: We have also requested that the manufacturer evaluate whether additional clarity in information could be incorporated at the next Instructions for Use review: The MHRA is continuing to monitor this situation and will investigate any further incidents that we receive If the MHRA can be of any further assistance in this matter, please contact us.

Dear Mr Oliver am writing in response to the Regulation 28 letter issued on August 2016. The content of the Regulation 28 letter has been given due consideration and an action plan in place to reduce any risk of future harm from a similar incident to patients in our care_ Mr Matthews was a gentleman who died from an infective exacerbation of chronic obstructive pulmonary disease on 19 February 2016. A concern was raised regarding equipment failure potentially contributing to the cause of death. An oxygen flowmeter had suffered a crack and was defective. have discussed the Regulation 28 with the Head of Clinical Engineering for the and he has provided further clarity as follows: The Flowmeter found to be damaged was still in use by the ward staff. However the flowmeter still delivered Oxygen but at an undefined rate due to the small crack in the plastic outer tube of the flowmeter body: Under pressure of the supply gas (nominal 4 bar; 0z) it is reasonable to 1 assume the gas escaping to atmosphere via the crack was variable ad hence related to the flow setting on the device caused by any back pressure level. Noise associated with the damagelleaking of 02 gas would most likely be evident to the user when operating the device but again probably related to the flow setting on the device itself. The actual crack damage would not get worse due to increased flow as such, but rather the 0 gas 'lost' to atmosphere would increase by escaping through the crack in the outer tube. 8

Chief Executive Darlington Memorial Hospital Hollyhurst Road 2 Darlington County Durham DL3 6HX put Trust

County Durham and Darlington NNHS] NHS Foundation Trust Overall maximum pressure within the flowmeter is limited to the supply gas pressure from the wall outlet (nominal 4 bar = 6OPSI approx ). Equipment suspected of damage should not be used. It should be removed from use, reported and a replacement device, in good condition, substituted. This is part of national and Trust policy: This issue is cleardly documented in issued guidance from MHRADoH: The Oz flowmeter device in question is a very simple device with one mechanical moving part (on/off flow knob): Everything else is sealed by design and in normal conditions the flow indicator is proportional to the gas flow being set with the flow knob by the clinical user: Oxygen is a and administered accordingly: In good condition these devices last for many years and are unlikely to fail due to natural component degradation. They require no planned technical intervention other than basic in- use functional checks by users. Damage is rectified by technical repair or replacement once the device is reported as defective. Replacement would normally be planned 'as required' by the users as a small value item (typically €60-E90) , or following major damage that renders a repair uneconomical AlI these devices work in the same way from the control knob action (anti-clockwise for 'ON' and flow increase versus clockwise for decreased flow and 'OFF') Following review of this incident ! can confirm that the damage report was not confined to within the Trust and apologise that this information was not fully shared with you at the time: The incident was reported to Adverse Incident Centre at MHRA and they asked the manufacturer to investigate our report of damage. The manufacturer responded to our report and MHRA responded to the Trust on 25 2016, and stated that found the manufacturer response acceptable and that they had closed the incident, as follows: MHRA Ref: Your Ref: Thank you for your report in connection with the following: Device: Flowmeter; Manufacturer: Oxylitre Ltd asked the manufacturer to investigate your report: Their response is as follows: "We found that there was a crack in the outer tube of one of the twin flowmeters: When plugged into & gas source with the control knob tumed off, the bobbin (which indicates flow) is positioned at the top of the inner tube. 1 When tured on the flowmeter would still supply the maximum flow rate. The customer indicated noticed the crack when tuming off the flowmeter; so regardless of the crack the flowmeter was continuing to supply oxygen The crack in the of flowmeter outer tube (approximately 30mm in length) was likely to have occurred by an impact As soon as the unit is plugged into an Oxygen Teminal it would have been immediately obvious that it was leaking 2 through the crack due to the very audible sound. Therefore we do not understand how clinical staff would deliberately use a defective unit: However; it is 8

Chief Executive Darlington Memorial Hospital; Hollyhurst Road, 2 Darlington, County Durham DL3 6HX drug the May they We they top

County Durham and Darlington NHS] NHS Foundation Trust possible that the damage occurred whilst in use, but as indicated the device would still deliver Oxygen Note Oxygen tube fitting is missing from the base of the other flowmeter: The retured unit is 15 years old. We could replace the cracked outer tube as a repair; but we do not support items that are over 10 years old and would normally offer a service exchange device. This has been indicated to the person who reported the incident as of 25.4.16. There is no other particular action Oxylitre can take since there is no design fault or any other problem with this model of device. We have no idea how this item could be associated with the death of a patient. As indicated, it would still supply Oxygen to a patient regardless of the damage found. Also as indicated we cannot understand how an obviously faulty device would be used in a clinical environment to administer Oxygen to a patient We can only conclude that because there is a link; however tentative, to a patient's death an adverse incident was raised regardless if there was zero contribution due to the damaged item. (MHRA) have reviewed the manufacturer's conclusions and consider them acceptable so we won't investigate any further: You raised concerns that none of the information provided by the Trust has addressed the audible indication of damage to the device, and the fact that it was not detected by that means. A medical devices newsletter was circulated in April which outlined that the MHRA (2015) advises routine maintenance ensuring that the device continues to function correctly: It may include regular inspection and care, as recommended in the manufacturer's user information or Trust policy. Tasks may include: Checking that devices work correctly before use Regular cleaning Specific daily, weekly checks Noting when a device has stopped working properly or when obvious damage has occurred, and then discontinuing use Contacting the relevant servicing organisation Also included in the message was a reference to audible evidence of a gas leak from the device; 1 'when this type of damage occurs, staff may hear a hissing sound of the gas escaping when in use and see visible damage. PCF Medical Devices @ Newsletter Issue 20 / 8 WWW: cddft nhs.uk Chief Executive Darlington Memorial Hospital, Hollyhurst Road 2 Darlington, County Durham DL3 6HX We

County Durham and Darlington NHS] NHS Foundation Trust have ensured that this issue has been raised at the Senior Nurse Leadership Group and all Care Group Governance Meetings within the organisation: AlI Associate Directors of Nursing have discussed this with clinical staff in all departments to reinforce the importance of checking oxygen flowmeters which includes any detection of audible hissing indicating a leaking of gas from the device. Actions All oxygen flowmeters across the Trust have been checked by the Clinical Engineering Department and any faults logged and reported to Department Managers All Equipment ControllersiDepartment Managers to perform a full weekly check of all their flowmeters and report any damage to their local Clinical Engineering departments. A checklist has been devised by the Medical Devices Nurse and will be discussed at the next meeting of the National Association of Medical and Educational Trainers (NANDET) (next meeting 21/09/16) before being utilised across the organisation: Medical Gas Flwmeter checklist d hope the response addresses the concerns raised and provides assurance of the processes that have been in place to prevent recurrence of this issue