Karl Dunstan

MK NHS}

Milton Keynes
University Hospital
NHS Foundation Trust

Mr Tom Osborne
HM Senior Coroner
Milton Keynes Council

Coroners. Office@milton-keynes.gov.uk

15 August 2025
Dear Mr Osborne
Regulation 28 Report following an Inquest into the death of Mr Karl Dunstan

| am writing following receipt of a regulation 28 report dated 25 June, following on from
the Inquest concluded on 24 June 2025. Mr Dunstan died from a pulmonary embolus
secondary to a deep venous thrombosis the day after his admission to hospital.

| was sorry to learn of Mr Dunstan's death, and it is very evident from the inquest pack
that its nature and timing has had a profound impact upon a close and loving family.
My thoughts are with them.

| shall come on to address the substantive content of the regulation 28 report, but | do
find it necessary to comment on the text of your narrative conclusion before doing so.
Specifically, you assert that a failure to undertake a D-dimer test and therefore to
obtain an urgent CT pulmonary angiogram (to make a definitive diagnosis of a
pulmonary embolus) meant that thrombolysis was not started when he later collapsed
(in cardiorespiratory arrest). You imply that thrombolysis would perhaps have altered
the sad outcome in describing ‘missed opportunities [which] more than minimally
contributed to his death’. For avoidance of doubt, we consider that whilst a
different course of events and actions might potentially have influenced the
ultimate outcome, it would not — on the balance of probabilities — have done so.
Mr Dunstan’s management — although we naturally wish it had been different given
what subsequently transpired — was reasonable at each decision point. The Trust does
not make any admission of breach of duty. When asked, the clinicians attending the
inquest as witnesses expressed some disappointment that they had clearly not been
able to communicate some key points to you.

In terms of the substantive content of your regulation 28 report, you assert that Mr
Dunstan should have had both a D-dimer and a CT pulmonary angiogram (CTPA).

| would make the following points:

e Confirmed pulmonary embolus is not a rare diagnosis in hospital practice but
nor is it common.

sl

Asaw
pa
particy

paching hospital we conduct education and research to improve healthcare for our Chief Executive: Joe Harrison CBE
ts During your visit students may be involved in your care. or you may be asked to . elas .
pate in a clinical trial Please speak to your dector or nurse if you have any condemns Chair: Heidi Travis OBE

MK NHS

Milton Keynes

University Hospital
NHS Foundation Trust

e Pulmonary embolism is a notoriously difficult clinical area: in order to pursue
investigations towards a diagnosis, the clinician needs to be sufficiently
concerned about the likelihood of a positive diagnosis (as ionising radiation to
the chest carries real risk), and the diagnostic approach adopted changes
according to the clinical context and the level of suspicion. Risk stratification
scores (such as the Wells score) should only be applied when the diagnosis in
question is felt to be a real possibility. Likewise, ubiquitous use of D-dimer in
any patient presenting to the Emergency Department with symptoms or signs
which could be consistent with (but were not particularly suggestive of)
pulmonary embolus would not be appropriate. [NE(the Canadian
professor who designed and validated the scoring system bearing his name)
highlights these challenges:

The model should be applied only after a history and physical [examination]
suggests that venous thromboembolism is a diagnostic possibility. It should not
be applied to all patients with chest pain or dyspnea or to all patients with leg
pain or swelling. This is the most common mistake made. Also, never... do the
D-dimer first [before history and physical exam]. The monster in the box is that
the D-dimer is done first and is positive (as it is for many patients with non-VTE
conditions).

e Departments across the NHS (and indeed globally) use validated screening
scores to determine the pre-test probability of a pulmonary embolus. Where the
pre-test probability is low, a negative D-dimer test can be helpful in providing
further assurance that significant pulmonary embolus is unlikely. A positive D-
dimer is non-specific (see below).

e Clinicians will often establish longs lists of ‘differential diagnoses’ with some
being more likely than others in a specific patient presentation. A combination
of the natural history of the presentation (how the signs and symptoms evolve),
response to initial treatment and investigative tests are used during the course
of the admission to firm up on the diagnosis / diagnoses. Some conditions on
the list of differential diagnoses may be considered possible but very unlikely —
to be reconsidered later if one of the more likely diagnoses is not confirmed.

e Not every patient in whom a pulmonary embolus may feature as part of the
differential diagnosis needs or should have a CTPA. This is for a number of
reasons including resource availability and the risks of the test (significant
ionising radiation and the potential for contrast reaction). Pulmonary embolus
ought to be near the top of the list of differential diagnoses, rather than at the
bottom of that list, fora CTPA to be pursued. For clarity, it is my understanding
that the treating consultant did not at any point during her involvement consider
that pulmonary embolus was the most likely diagnosis. Her view was that
Ongoing infection (or an infective complication such as lung abscess or
empyema) was much more likely. This view was reinforced further after a
positive response to initial treatment when reviewed (once aware that the CTPA
had not been authorised).

SSS

Chief Executive: Joe Harrison CBE
Chair: Heidi Travis OBE

MK INHS

Milton Keynes

University Hospital
NHS Foundation Trust

e The circumstances in which to use scoring systems +/- D-dimer are somewhat
nuanced. D-dimer only really has a role when the clinician considers that the
diagnosis of pulmonary embolism is a significant possibility — somewhere
between possible and probable.

e The Wells Score is primarily designed for patients presenting acutely to an
Emergency Department with de novo symptoms. It is not particularly useful in
hospitalised patients. Given the chronicity of Mr Dunstan's presentation (with
two previous courses of antibiotics in the community), it could be argued that
he had more in common with a hospitalised cohort (rather than patients
presenting with de novo symptoms).

e D-dimer is considered to be an ‘acute phase reactant’. That is to say that levels
will often be elevated in association with infection or inflammation and D-dimer
is not specific for venous thrombosis. The consultant was also aware of a raised
ferritin (another acute phase reactant) which was being looked into by Mr
Dunstan's GP.

e When aCTPA is requested at MKUH, decision support software (iRefer) is used
to ensure that the clinician is not requesting an inappropriate or unnecessary
test.

e When a CTPA is requested at MKUH, the pre-test probability (Wells Score) is
considered and, where the probability is low, a D-dimer is sought. A low
probability request accompanied by an elevated D-dimer allows the
radiographer to authorise the investigation.

e Inthe absence of a positive D-dimer, the clinicians looking after the patient are
at liberty to speak to the duty radiologist and the investigation may then proceed
without the need for a positive D-dimer (or indeed with a known negative D-
dimer).

e In this case, the actual Wells Score (as evident in statements provided to the
inquest) was low at 1. This equates to a 1.3% chance of patient in the
Emergency Department population subsequently being confirmed to have a
PE. The request card stated a Wells score of 4 as the requesting resident doctor
Stated that PE was felt to be the most likely / equally likely diagnosis — it is likely
that the granularity and nuance of the consultant's thought processes would not
have been communicated in detail (and a Wells score of 1 versus 4 did not
impact on how the request was managed in radiology).

e The consultant did not consider that there was clinical urgency for the requested
CTPA to be carried out. Indeed, the scan was not solely or indeed primarily
being requested to evaluate the vasculature. Other reasons for the request
were to evaluate the lung fields themselves, on account of the long course of
Mr Dunstan’s symptoms (to rule out malignancy or another underlying
condition, or to demonstrate a septic collection — abscess or empyema — in the
context of the clinical picture of ‘slow to resolve’ infection) and the relatively
poor quality of the plain chest X-ray (in part on account of body habitus). The
possibility of an abscess or empyema was higher in the differential diagnosis
that a pulmonary embolus: the pulmonary angiogram element (the times
intravenous contrast) was essentially an add-on as a CT was being requested.

Leslee

As a teaching hospital, we conduct education and research to improve healthcor
Patients During your wsit students may be involved in your care, or you may be asked to . saa
Participate ino clinica! trial Please speak to your doctor or nurse if you have eny concerns Chair: Heidi Travis OBE

Chief Executive: Joe Harrison CBE

MK INHS)

Milton Keynes

University Hospital
NHS Foundation Trust

Evidently, the nuanced rationale around the imaging request is not fully
apparent from the request form and its focus on Wells Scores and D-dimers.

e Importantly, the medical team (including the consultant) was aware that the
radiographer had not been able to authorise the scan (due to a low Wells score
and absence of a D-dimer) and knew that they could seek authorisation by
discussing with the duty radiologist. The consultant did not do that as Mr
Dunstan had objectively improved over the course of the day in response to
initial treatment adding weight to the primary diagnosis of infection / dehydration
being the cause of Mr Dunstan's presentation and symptoms. The chance of
pulmonary embolus being the driving diagnosis was, at this point, felt to be even
lower than at presentation. The rationale for the CT chest (including the
angiographic component) was now even weaker — and the CT chest could
potentially have been cancelled or deferred to the six-week point. At this time
(in the afternoon), PE was not really a diagnosis which the team was actively
wanting / needing to exclude. Furthermore, and in pragmatic terms, Mr
Dunstan was being given prophylactic doses of low molecular weight heparin
which were relatively high (on account of body weight) and close to treatment
doses.

In light of this case, we have:

e Discussed the case at the Trust’s Grand Round (29 January 2025) to gain
alternative views and perspectives.

e Held discussions between departments at the hospital (medicine and radiology)
to review and optimise the screening and triage systems for various groups of
patients (outpatient, acute presentations, inpatients).

e Liaised with neighbouring Trusts to understand the approaches which they
currently use in this scenario.

We consider that our current approach is consistent with other NHS organisations and
broadly strikes the right balance between the advantage and risks of CT pulmonary
angiography.

We do plan to undertake an audit to look at pick up rates (of pulmonary embolus)
versus the Wells score and D-dimer. Clearly it is important to understand whether the
request and scoring systems are being used appropriately. Specifically:

e Are the rates of positive PE diagnosis for patients undergoing CTPA at MKUH
broadly in line with that which would be predicted by the recorded Wells scores?

e Is the patient group in whom the Wells score is being utilised appropriate
(outpatients and de novo ED attendances)?

e Are requestors placing undue emphasis on aspects of the history, or being
tempted to exaggerate, in the knowledge that the scoring system is in use.

—_—_——————————— LL

“ rondcl edtolon and ie search $0 ny moron © healthea
patents During y 1 students may be involved tn your core. or you may
participate in a clinical inal. Please sped bk aoa doctor of nurse st you howe ony

$a teaching ho:

Chief Executive: Joe Harrison CBE
Chair: Heidi Travis OBE

MK INHS)

Milton Keynes

University Hospital
NHS Foundation Trust

We plan to trial a system for six months (and evaluate) whereby if a CTPA cannot be
approved by the radiographer, it will be brought to the attention of the duty radiologist.
If the radiologist is satisfied (from the request narrative) that a D-dimer is not required,
he/she may authorise the study. If the radiologist is not satisfied, efforts will be made
by the radiographer to contact the requester by bleep / telephone to inform them of
this and to invite them to undertake a D-dimer or to discuss further investigation with
the duty radiologist as appropriate (in addition to the current system of the primary
communication being electronic).

| trust that this response is helpful.

Yours sincerely,

Chief Executive: Joe Harrison CBE
Chair: Heidi Travis OBE