Prevention of Future Deaths report 2021-0233

Dear Ms Persaud, 

Thank you for your letter dated 19/10/21 highlighting further concerns about the use of incorrect 
filters. We acknowledge the concerns raised by the consultant anaesthetists in your findings and will 
continue to monitor the situation through our regular safety signal and surveillance activities. This 
includes reviewing multiple data sources, including our own database of reported incidents, reports 
in journals and other media, as well as routine stakeholder engagement. Where necessary we will 
issue safety messages to health care organisations, patients, and the public. 

As discussed in our letter dated 31st August 2021, we contacted eight of the largest manufacturers 
of this device type that sell devices in the UK. We asked these manufacturers to conduct a search of 
their own databases for any similar incidents that had been reported to them in the past five years. 
As well as identifying if a patient was injured, or if there was the possibility of a patient being 
injured. We asked the manufacturers to include incidents that were outside of the jurisdiction of the 
MHRA, such as incidents that had occurred in the wider European Union. All of the manufacturers 
responded, a total of four incidents were identified. Two of the reported incidents occurred outside 
of the UK. This data should be considered in context that in this same 5-year period millions of filters 
were used in the UK  

This lack of reported incidents could indicate that this type of issue does not occur as frequently as 
believed. It is also possible that for the reasons discussed in our previous letter this type of incident 
may be under reported. To attempt to address this and encourage reporting the MHRA will engage 
with the medical device safety officers (MDSO) network at the next scheduled meeting in February 
2022, to raise awareness of possible incidents involving filters and encourage reporting of such 
incidents to the MHRA.  

We acknowledge your concerns with the lack of clarity of labelling of these devices. Annex 1 
(Essential Requirements) of the UK Medical Device Regulations 2002 sets out the labelling 
requirements for all medical devices. Manufacturers are required  to label devices with a range of 
identifier information which includes in section 13.3.b the details strictly necessary to identify the 
device and the contents of the packaging especially for the users; and 13.4 If the intended purpose of 
the device is not obvious to the user, the manufacturer must clearly state it on the label and in the 
instructions for use.  

MHRA will write to known manufacturers of filters registered with MHRA and the Association of 
British HealthTech Industries (ABHI) and UK Approved Bodies, to ask them to conduct a review of the 
labelling of filter devices against the regulatory requirements, taking into consideration the findings 
of the inquest, and making improvements where identified  

The MHRA fully encourages users to be trained on the use of any medical device before using it, as 
detailed in our published guidance Managing Medical Devices, however it is not within MHRA’s 
remit to provide training or to enforce it. There are a number of national organisations that carry out 
medical device training, for example Health Education England and The National Association of 
Medical Device Educators and Trainers (NAMDET). Training resources may also be available via the 
Royal Colleges’ and other professional organisations/societies. 

It is worth noting that existing evidence suggests colour coding to distinguish devices can have little 
or no impact. This issue should be explored further by HSIB. 

 We will contact HSIB and pass on your recommendation that an investigation into the possible 
confusion over filter types is carried out. In this communication we will suggest that they involve 
both the Royal College of Anaesthetists and The Faculty of Intensive Care Medicine in their 
investigation. We will also support the HSIB in any investigation that they undertake into this matter    

I hope this information has been helpful, if you have any further questions please feel free to contact 
me.  

Yours sincerely, 

Chief Safety Officer 

Directorate - Directorate Division

Miss N Persaud 
Her Majesty’s Coroner 
Walthamstow Coroner’s Court, 
Queens Road, 
Walthamstow 
E17 8QP 

Medicines & Healthcare products 
Regulatory Agency 

10 South Colonnade 
Canary Wharf 
London
E14 4PU 
United Kingdom 

gov.uk/mhra 

Dear Miss Persaud, 

Please find a summary below detailing the role of the MHRA Devices Division and an outline of the types of 
assistance that we can offer. 

The Medicines and Healthcare Products Regulatory Agency is an executive agency of the Department of Health. 
As part of this, the MHRA Devices Division is responsible for protecting the public’s health and safeguarding the 
interests of patients and users by ensuring that medical devices and equipment meet appropriate standards of 
safety, quality and performance and that they comply with relevant UK medical device directives. Details on all 
the business areas of the agency can be found on the website Medicines and Healthcare products Regulatory 
Agency - GOV.UK (www.gov.uk). 

One major area of the MHRA Devices Division’s responsibilities is the investigation of adverse incidents involving 
medical devices. An adverse incident is an event involving a medical device that produces, or has the potential to 
produce, unwanted effects impacting on the safety of patients, users and other persons. These incidents may 
arise from defects in the device design or manufacture, its operating instructions, user practice or conditions of 
use. 

Where adverse incidents involving medical devices have been reported to other authorities, including the Police, 
Coroners and the Health and Safety Executive (HSE), we are able to provide the following assistance to that 
authority: 

1. Liaise with the manufacturer of the medical device(s) and request that they investigate and test the 
device(s). As part of the manufacturer’s investigation they may choose to have the device independently 
tested at a laboratory with specialist equipment.  We can then request a report from the manufacturer 
summarising their findings, including the root causes of any device issues (if established), and provide this 
to the instructing authority. 

2. Witness the above testing/inspection of device(s) by the manufacturer or its representatives in the UK 
only. Please note, we do not have the resources or specialist equipment to undertake our own 
testing/inspection of medical devices. Normally the manufacturer will undertake testing at their premises, 
which may be located overseas. 

 
 
 
 
 
 
 
 
 
 
  
 
 
 
 
 
 
 
 
  
 
 
 
 
 
 
 
 
 
 
 
 
   
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 3. Produce a summary letter with comments on the manufacturer’s report (if applicable). The summary 
letter may also provide details such as the number of similar reported incidents on our database, and any 
related safety actions that may be relevant to the device in question. 

4. Help identify an independent medical device test facility, provided that there is no need for proprietary 
knowledge of unique design or software. Costs incurred will be at the instructing authority's expense. 

5. Attend the coroner's court to give evidence, if instructed. 

In addition, we may issue safety information to the health service or advise the manufacturer to make design 
changes to their product to prevent similar occurrences in the future. 

To assist us in an effective investigation, please advise us as soon as possible if you have any specific 
requirements. 

With regards to the regulation 28 report dated 07/08/2021 and the MHRA’s response to the request for comment 
received from the DHSC, to date (16/08/2021) the MHRA has not received incident reports for either of the two 
incidents mentioned in the Coroner’s report. This makes it difficult to carry out a thorough investigation or to 
comment on specifics. For a complete investigation to take place the MHRA would need to be provided details via 
the yellow card reporting scheme. It would be helpful if you provide: 

•  The name of the legal manufacturer and model name of the filters in question. 
•  Details of the devices with which the filters were used including legal manufacturer and model name. 
•  As detailed a description of the incident as possible. 
• 
•  Whether the filters involved were returned to the manufacturer for inspection. We accept that this may 
not be possible at this point due to the length of time that has passed from the date of the incident. 

If available, the lot and batch numbers of the filters involved. 

•  Any additional information reported would greatly assist our investigation. 

With this information the MHRA could undertake the investigation procedure described above with the help of 
the manufacturers. 

The MHRA has no record of any similar reports being received in the past ten years. It is possible that this is not a 
common incident and could be the result of the unique conditions of the past 12 months. It is also possible 
however that this type of incident is not often reported to the MHRA. There are a number of possible reasons for 
this. 

It is a commonly held misconception that the MHRA is not interested in incidents that could be classed as “user 
error”. This is not the case however, as if multiple reports of “use error” with a device or device type were 
received it could be a signal that, despite the manufacturer’s best efforts, the device is not user friendly when 
placed into practice. In these circumstances the MHRA would work with the manufacturer to determine the root 
cause of the issue. This root cause can then be addressed either as an iterative design improvement or a field 
safety corrective action by the manufacturer, depending on circumstance. 

Another possible reason for non-reporting of this type of incident is that the healthcare professional may notice 
the error prior to patient treatment or soon after treatment commencing and correct the issue, resulting in no 
patient harm. In this situation users may have thought that the incident was “too trivial” to report to the MHRA. 

It is also possible that this type of incident has occurred and not been noticed by healthcare professionals as no 
abnormalities or patient harm has occurred. 

It should be noted that health care professionals are under no obligation to report any incidents to the MHRA, or 
device manufacturers. Device manufacturers are required to report incidents that meet the criteria for vigilance 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  
 
 
 
 
 
 
 
 reporting as described in the Medical Device Directives and MEDDEV 2.12-1 rev 8 to the MHRA. However, 
manufacturers are only able to report incidents where they are informed of them.   

With regards to the introduction of a standard on filter colours. At the current time there is no agreed 
classification of all the filter types, which means that manufacturers can design and make any filter “type” to suit 
the design of their devices and perceived clinical need. There are not necessarily filters that fulfil the exact same 
function and specification between manufacturers. This diversity would make agreeing a standard and universal 
colour coding extremely difficult. 

If an agreement over the classification could be reached manufacturers would be under no obligation to stick to 
the agreement, leading to continued deviation. This has been shown in other areas of critical care and 
anaesthesia such as with Guedel airways and anaesthetic drugs.  

An attempt to create a standard for filter types and colour coding would require input from multiple organisations 
both domestically and internationally. This would include the manufacturers of the devices, the International 
Standards Organisation (ISO), the British Standards Institute (BSI), the MHRA and other international competent 
authorities. The drafting and implementation of a standard would be a protracted piece of work, possibly taking 
multiple years, and there is no guarantee that this would take precedence over other priorities across these 
organisations. 

The introduction of an international standard could reduce the confusion surrounding the different types of filter 
and possibly result in a safer environment for patients in the UK and worldwide. However, due to the time 
required to implement the introduction of a new standard this would not be an immediate solution to the 
problem experienced in the report. It is possible that the introduction of interim solutions could improve patient 
safety until a standard could be introduced. These interim measures could include actions such as: 

•  A reduction in the number of different filter types used to avoid the possibility of confusion. 
• 

Filters being prepared away from the clinical area reducing the reliance on front line staff needing to 
know the difference between multiple filters. 

•  Where multiple filter types are required it would be better for clinical areas to use filters from one 

manufacturer. This is because manufacturers tend to be consistent to their own internal colour coding 
•  Where possible an alternative manufacturer, preferably with a similar colour coding, should be identified 

as a back-up supplier. 

We acknowledge there is a concern that confusion over the type of filter used in ventilation could potentially lead 
to patient harm. However, at the current time there is insufficient evidence to show that a standardised colour 
scheme would reduce the risk of patient harm. The MHRA will work with manufacturers, other regulators, NHS 
England and Improvement and other stake holders to fully explore the effects that enacting the kind of actions 
detailed above would have on patient safety. We will also work with these groups to improve reporting of this 
type of incident and build our knowledge base surrounding the risks with the use of multiple filter types and any 
issues surrounding their labelling.  Any emerging evidence relating to possible risks associated with these devices 
will be carefully reviewed and, if appropriate, regulatory action will be taken if any serious risks were confirmed 

Thank you in advance for your assistance in this matter. 

Yours faithfully, 

Director of Devices 
MHRA, 10 South Colonnade, Canary Wharf, London E14 4PU

Prof. 

________________________________________________________________ 

10 October 	2021 

Ms Nadia Persaud 

Senior Coroner 

Walthamstow Coroners Court 

Queen’s Road 

Walthamstow 

London 

E17 8QP 

Dear Ms Persaud, 

Re HME/Filters 

Further to last week’s inquests and your request for comments in relation to HME/Filters, I have set 

out the following background, issues and potential solutions. 

Background and current issues 

1.  When a patient is connected to a ventilator or anaesthesia machine, the tube that is in their 

trachea (orotracheal tube [via their mouth] or tracheostomy [through their neck]) bypasses their 

natural processes (in the nose and pharynx) for providing humidification to their upper airway 

and lungs. Furthermore, unlike room air, medical gases are completely free of water. Thus, 

some form of artificial humidification is required when a patient is receiving mechanical 

ventilation. 

2.  Such humidification can be provided in three main ways: 

a.  Passive heat and moisture exchange (HME) devices placed at the patient end of the 

breathing system connected to tracheal tube. 

b.  Active humidification through a device placed in the breathing system next to the 

ventilator. 

1 

 
	
 
	
	
	
	
	
	
	
	
	
	
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
   
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 c.  Anaesthesia machines typically include a soda lime cannister to absorb carbon dioxide – 

this process also provides humidification, but this is insufficient for patients who require 

prolonged artificial ventilation. 

3. 

Intensive care ventilators will typically have a disposable bacterial/viral filter placed where the 

expiratory limb of the patient breathing tubing attaches to the ventilator - this is to protect the 

ventilator from becoming contaminated by infection. Sometimes the same type of filter is 

attached between the ventilator and the inspiratory limb of the breathing tubing. Whether one 

or two of these filters are used, they should be attached directly to ventilator.  These 

bacterial/viral filters do not provide any humidification. 

4.  Many HMEs also include a bacterial/viral filter and are known as HMEFs. These would also be 

placed in the patient end of the breathing system. 

5.  HMEs, HMEFs and bacterial/viral filters come in many shapes and sizes but are essentially 

indistinguishable. They will normally have writing on them to indicate what they are but this can 

be fairly subtle. They are usually coloured, and the manufacturers generally use one colour for 

each type, but there is no national or international standard for these colours, and they vary 

between manufacturers. 

6.  Several techniques are used to closely monitor a patient’s breathing when they are receiving 

artificial ventilation. A standard of care is to measure continuously the carbon dioxide (CO2) that 

a patient breathes out. This requires a CO2  sampling tube to be attached to the breathing 

system close to the patient end of the breathing system. 

7.  Many HMEs and HMEFs include a port specifically designed to attach a CO2 sampling line – this 

prevents the need for a separate connector in the breathing system specifically for the CO2 

sampling line. 

8.  Currently, most patients receiving artificial ventilation in the ICU will be receiving humidification 

from an active system. HMEs should not be used with active humidification because the HME 

will quickly become water-logged, which severely restricts or even prevents adequate 

ventilation of the patient. Where active humidification is used, attachment of one or two 

bacterial/viral filter filters to the ventilator will prevent it from being contaminated. These filters 

should not be placed at the patient end of the breathing system because they will reduce the 

amount of humidification reaching the patient and they can also become water-logged. When 

active humidification is used, the CO2 sampling line will need a dedicated connector to enable 

sampling from the patient end of the breathing system. 

2 

 
 
 
 
 
   
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
   
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
   
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  
 9.  Many manufacturers produce bacterial/viral filters that incorporate a port for CO2  sampling (see 

Intersurgical, Drager and GE brochures). I can think of no reason for these to be used in 

anaesthesia or intensive care settings, although it is possible that they are aimed at resuscitation 

situations, particularly out-of-hospital, when short term ventilation is provided in an emergency 

(negating the need for humidification). 

10. In my opinion, that the bacterial/viral filters included a sampling port contributed to the 

confusion recently experienced at the Nightingale Hospital London. This was compounded by 

non-standardised colours. 

Potential solutions 

11. Colour coding of filters should be standardised. For example: 

a.  Bacterial/viral filter with HME function – yellow 

b.  HMEF – green 

c.  HME only – blue 

d.  Another possibility is to have one colour for bacterial/viral filters, one for HMEs and then 

use both colours (in stripes?) for the HMEFs. 

12. Unless there is an essential reason for including a sampling port on bacterial/viral filters, 

manufacture of such devices should cease. If they are considered essential (perhaps in a 

resuscitation situation), they should be cleared marked (packaging and the device itself) to 

indicate that they provide no humidification. 

13. I think there is an opportunity to enhance the education about HMEs and filters. Given the 

incident at NHL, it may be appropriate to publish a National Patient Safety Alert on this topic 

(https://www.england.nhs.uk/patient-safety/patient-safety-alerts/) 

Prof. 

10 Oct 2021 

3

14 December 2021 

Dear Ms Persaud, 

Re: Regulation 28: Report to Prevent Future Deaths in the matter of Mr Kishorkumar Patel and Mr 
Kofi Aning

Thank you for your letter of 18th October providing further details and conclusions following the 
inquest touching the death of Mr Patel. You have highlighted a number of additional concerns 
that have been raised relating to the causes underlying the incorrect use of breathing circuit filters 
and suggested further actions to reduce the risk of future harm to patients.  

We confirm our support for standardisation of labelling/colour coding of breathing circuit filters 
and heat and moisture exchange filters (HMEF) between manufacturers. We highlight that it is not 
within our power to implement beyond referring the issue to the Medicines and Healthcare 
products Regulatory Agency to consider whilst recognising the challenges in implementing such a 
change.  

We would again highlight the unprecedented circumstances arising from the need to develop a 
remote surge ICU during the early stages of the COVID pandemic which are likely to have had a 
significant contribution to the errors that occurred.  The type of filter (with an integral port for 
capnography) that was mistaken for an HMEF is designed for use with anaesthetic machines and 
there would be no reason for such a filter to be used in ICU. We are not aware of previous 
incidents being highlighted from our regular review of national incident reporting in critical care. 
However, we have been in communication with NHS improvement to suggest that a formal 
analysis of the national reporting and learning system database (NRLS) should be undertaken to 
assess the frequency of incidents arising from incorrect used of breathing circuit filters. If this is 
shown to be a significant problem, we would support this being highlighted through a national 
patient safety alert. 

We can confirm that the key lessons will be highlighted to our respective membership through the 
Safe Anaesthesia Liaison Group’s Patient Safety Update and FICM Safety Bulletin. These will 
include: 

•  HMEFs and plain filters may be confused as they can look similar and the labelling may not 

be clear. 

•  Plain filters with sampling ports appear to be of particular risk for being mistaken for HMEF, 

are designed only for use in anaesthetic machines and should not be available in an ICU.   

  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  
 
 
 
 
 
 
 
 •  All members of the multi-disciplinary team with responsibility for managing ventilated 

patients must be aware of the difference between the plain filter and HMEF in use on their 
unit and their correct placement in the ventilator breathing circuit.  

• 

There must be a system in place to ensure that regular checks of the ventilator breathing 
circuit are undertaken. The check should ensure that an appropriate form of 
humidification is being used. 

We hope that this action will satisfy you that we are taking appropriate steps to ensure that 
anaesthetists and intensivists are aware of these issues, and that these steps should make future, 
similar adverse events less likely to occur. 

We would be happy to respond to any questions that you might have. 

Yours Sincerely 

Dr 
President, Royal College of 
Anaesthetists    

Dr 
Dean, the Faculty of Intensive Care  
Medicine