SHOW / EPISODE

From AI to Reality: Navigating Multimorbidity in diabetes with NICE Guidelines

11m | Oct 17, 2023

This episode makes reference to guidelines produced by the "National Institute for Health and Clinical Excellence" in the UK, also referred to as "NICE". Please note that the content on this channel reflects my professional interpretation/summary of the guidance and that I am in no way affiliated with, employed by or funded/sponsored by NICE.

My name is Fernando Florido and I am a GP in the United Kingdom. In today’s episode I look at a fictitious clinical case of a patient created by Chat GPT to see how the NICE guidelines could apply to it.

 

I am not giving medical advice; this video is intended for health care professionals, it is only my interpretation of the guidelines and you must use your clinical judgement.  

 

There is a YouTube version of this and other videos that you can access here: 

There is a YouTube version of this and other videos that you can access here: 


The patient was created using the following Chat GPT prompt:

 

A)   Provide a fictitious patient. Details that you should include are:

1)   patient's medical information including:

·      name

·      age

·      sex

·      ethnicity

·      BMI

·      blood pressure

2)   medical history- you must include:

·      either one or two of the following poorly controlled conditions:

·      type 2 diabetes

·      hypertension

·      dyslipidaemia,

·      Asthma or COPD

·      Any number of other medical conditions of your choice, along with whether they are well controlled or not – the medication for these conditions should appear in the next section “medications”

  1. Medications given:

·      indicate whether the patient is currently taking medication for each medical condition or not.

·      If medication is prescribed for a condition, indicate the specific drug(s) and their dosages. You may choose to prescribe one, two, or three drugs for each condition as appropriate.

3)   State whether the patient tolerates the medication well or not, and if not, describe the side effect(s) and their severity.

4)   blood test results (give a bulleted list but do not number them):

·      HbA1c expressed in % and mmol/mol

·      renal function tests to include creatinine (expressed first in µmol/L and then in mg/dL), eGFR, urea, sodium, and potassium (expressed in UK units first and then USA units)

·      lipid profile expressed both in mmol/L first and then in mg/dL.

·      If the patient has asthma, give the peak flow reading expressed as a percentage of their best or expected reading.

·      If the patient has COPD, give the FEV1 reading expressed as a percentage of the predicted reading

·      include any other relevant test results for the patient, expressing them in both UK and USA units. If the patient has hypothyroidism or takes levothyroxine medication, provide the results of their thyroid function tests, including both T4 and TSH levels, in both UK and USA units. Also, include the normal range for these investigations.

B)  Provide the patient's cardiovascular risk using the QRISK2 tool, calculated as a percentage of the likelihood of experiencing a cardiovascular event over the next 10 years.

C)   At the end of the patient information, ask: 'What treatment recommendations would you make?' – do not make recommendations yourself

D)  Do not include a disclaimer that the patient is fictitious.


The NICE hypertension flowcharts can be found here: 


The full NICE Guideline on hypertension (NG136) can be found here: 

 

 

The full NICE Guideline on Type 2 diabetes (NG28) can be found here: 

 

·      https://www.nice.org.uk/guidance/ng28

 

 

The full NICE guidance on cardiovascular disease risk reduction (CG181) can be found here:

 

·      https://www.nice.org.uk/guidance/cg181


The full NICE guidance on asthma (NG80) can be found here:

 

·      https://www.nice.org.uk/guidance/ng80

 

The full NICE guidance on depression in adults (NG222) can be found here:

 

·      https://www.nice.org.uk/guidance/ng222

 

The full guidance on obesity (CG189) can be found here:

·      https://www.nice.org.uk/guidance/cg189

 

 

Thumbnail photos-

From: www.freepik.com

·      Image by Freepik: Image by a href="https://www.freepik.com/free-photo/excited-young-friends-running-beach_9057871.htm#page=2&query=45%20year%20old%20obese%20woman&position=10&from_view=search&track=ais"Freepik/a

 

 

Intro / outro music: Track: Halfway Through — Broke In Summer [Audio Library Release]  

 

  • Music provided by Audio Library Plus  

 

 

Transcript 

 

 

Hello everyone. My name is Fernando and I am a GP in the United Kingdom. Today we are looking at a fictitious patient generated by Chat GPT with a view to looking at the NICE guidelines from a practical perspective. 

 

If you would like to know the Chat GPT prompt that I have used, you can find it in the episode description. 

 

Right, so let’s jump into it.

Patient Information:

·      Name: Sarah Davis

·      Age: 45

·      Sex: Female

·      Ethnicity: Caucasian

Examination:

·      BMI: 32 (Obese)

·      Blood Pressure: 153/91 mm Hg

Medical History:

·      Type 2 Diabetes

·      Hypertension

·      Dyslipidemia

·      Asthma

·      Obesity

·      Depression

Medications:

·      Metformin 1000 mg twice daily

·      Gliclazide 80 mg once daily

·      Amlodipine 10 mg once daily

·      Indapamide MR 1.5 mg once daily

·      Atorvastatin 20 mg once daily

·      Salbutamol inhaler as needed

·      Sertraline 50 mg once daily

Medication Tolerance:

·      Sarah tolerates her medications well. She has not reported any significant side effects.

Blood Test Results:

·      HbA1c: 8.5% (69 mmol/mol)

·      Renal Function:

o  Creatinine: 90 µmol/L (1.02 mg/dL)

o  eGFR: 91 mL/min/1.73m²

·      Lipid Profile:

o  Total Cholesterol: 5.2 mmol/L (200 mg/dL)

o  LDL Cholesterol: 3.0 mmol/L (116 mg/dL)

o  HDL Cholesterol: 1.2 mmol/L (46 mg/dL)

o  Triglycerides: 2.5 mmol/L (220 mg/dL)

Cardiovascular Risk (QRISK3):

·      Sarah's 10-year cardiovascular risk is estimated at 18%.

So, what treatment recommendations would we make?

Although there is a lot that we can do from a lifestyle perspective, we will focus on the drug treatment only.

 

Right, let’s start with her diabetes, which is poorly controlled with an hba1c of 8.9% or 74 mmol/mol. Sarah is on metformin 1gr BD and gliclazide 80 mg OD and we know she is at high risk of cardiovascular disease with a QRISK3 of 18%.

 

NICE says that for step 2 treatment after metformin, if a patient is at high risk of CVD, we should choose an SGLT2 inhibitor. But this patient has been given gliclazide instead. We would like to see if there is a reason for this and whether an SGLT2 inhibitor was tried before but not tolerated.

So, assuming that there have been no issues, I would recommend starting an SGLT2 inhibitor.

SGLT2 inhibitors are also associated with a degree of weight loss, and given that Sarah is obese, this will be good for her.

Gliclazide on the other hand is associated with weight gain, which is the last thing that Sarah needs. The question now is whether we should stop gliclazide and, if so, when.

In my opinion, gliclazide should be definitely stopped and substituted by another diabetic agent which is weight neutral, for example, a DPP4 inhibitor.

A more difficult question would be when to make this switch. Doing it straightaway has got the disadvantage that, if a side effect develops, we may not know which drug is the culprit. Also, we do not know how much the HbA1c will drop with the SGLT2 inhibitor so there could be a risk of overtreatment.

Right, a balanced decision that you can make depending on your clinical judgement.

I think that I would start the SGLT2 inhibitor first, so, I would start her on something like dapagliflozin 5 mg OD initially and, if tolerated, I may increase it to 10 mg daily fairly quickly, possibly within 3 or 4 weeks to get the maximum effect as soon as possible.

Research studies have shown that the introduction of an SGLT2 inhibitor can lead to a drop in Hba1c of around 1%. Although this could vary depending on the individual patient, Sarah’s HbA1c is poor at 8.9% so even after the SGLT2 inhibitor, she is likely to require a third diabetic agent.

NICE says that if dual therapy with metformin and another oral drug is not enough, we should consider either triple therapy by adding a DPP‑4 inhibitor, pioglitazone or a sulfonylurea or starting insulin. Because of the risk of weight gain, I would try to avoid gliclazide and insulin.

Given that she is on gliclazide 80 mg daily, that is, a quarter of the maximum daily dose of gliclazide, I would probably switch Sarah to a DPP4 inhibitor as soon as she is tolerating the full dose of dapagliflozin. So, we could stop gliclazide and start her on, for example, sitagliptin 25mg daily, which is also a quarter of the maximum daily dose of 100mg.

But I do not think that there is a right or wrong approach in terms of the gliclazide switch. The decision about the timing is very dependent on the individual patient and your clinical judgement. Switching early has some more uncertainty about treatment response whereas switching later may mean that the patient remains on a potentially unsuitable drug for longer.

We have to make sure that we inform the patient that SGLT2 inhibitors have been associated with an increased risk of DKA, toe and lower limb amputations and Fournier’s gangrene.

So, on starting the SGLT2 inhibitor:

1.   We will educate her about DKA symptoms and when to seek advice.

2.   We will also advise against ketogenic diets to lose weight while on an SGLT2 inhibitor, as this also increases the risk of DKA.

3.   We will improve foot care and

4.   We will advise patients to seek urgent medical advice if they experience pain, erythema or swelling in the genital or perineal area, as this may precede Fournier’s gangrene or necrotising fasciitis.

 

If we were managing her diabetes according to the European or American guidelines, we would be talking a lot more about GLP 1 receptor agonists, given that they are also recommended for patients with high CVD risk and they are much better in terms of weight loss. We are following the NICE guidelines, which are more restrictive with GLP1 receptor agonists, but we can always use our clinical judgement to deviate from the guideline if we think that it is appropriate for a particular patient.

 

Now that we have dealt with her diabetes, let’s look at her hypertension.

 

Unless we have specific concerns, we do not need to arrange an ABPM or HBPM because NICE says that we can use clinic blood pressure measurements to monitor drug treatment for people already diagnosed with hypertension. However, NICE also say that in diabetes, we should check both the sitting and standing BP because of their higher risk of postural hypotension, especially if there is autonomic neuropathy.

 

 

Sarah’s BP is 153/91 and NICE says that the target BP for people under the age of 80 is below 140/90. So, her BP is high.

 

She is on amlodipine and indapamide, which is not in keeping with NICE guidance. NICE says that, for people with diabetes of any age and any ethnicity background, an ACEI or ARB should be used as first line treatment.

 

So, I would have a look to see if an ACE inhibitor or ARB has been tried before and then stopped because of side effects like a cough or angioedema. Both of these side effects can happen with ACEIs and they are normally managed by switching to an ARB. But, although much rarer, these symptoms can also be a side effect of ARBs

 

Assuming that there is no previous problem, I would start her on an ACE inhibitor, for example lisinopril 2.5 mg daily monitoring her renal function and titrating up according to response.

 

If the target BP is achieved at a lower dose than the maximal dose of the ACE inhibitor, then I would recommend stopping indapamide and continue titrating up the ACE inhibitor dose to compensate. Thiazide like diuretics can worsen diabetes so stopping indapamide and replacing it with a higher dose of lisinopril would be the right thing to do.  

 

Once indapamide has been stopped, if the target BP is achieved before the ACE inhibitor is at the maximal dose, we could consider reducing the dose of amlodipine while we increase the ACE inhibitor further, because of the benefits that ACE inhibitors have particularly in diabetes.

 

Now, let’s look at her hyperlipidaemia. Sarah has no history of cardiovascular disease so she is on atorvastatin 20 mg daily for primary prevention.

 

NICE says that we should offer atorvastatin 20 mg for the primary prevention of CVD if they have a QRISK score of 10% or more.

After atorvastatin 20mg has been started, and unlike secondary prevention, there are no specific lipid targets in primary prevention but, if we judge the person to be at higher cardiovascular risk, we will consider increasing the dose of atorvastatin in order to achieve a greater than 40% reduction in non‑HDL cholesterol.

In Sarah’s case, both her cholesterol and QRISK3 score are still high on atorvastatin 20 mg, so we could consider her to be at higher risk and, if the 40% drop in non HDL cholesterol has not happened, we could increase atorvastatin to 40 mg and monitor her blood tests.

Let’s now move on to her asthma. Sarah is on step 1 treatment, that is, on-demand Salbutamol inhaler as needed. If Sarah has infrequent symptoms, this is fine, but we should check that she has no symptoms that could indicate the need for maintenance therapy, for example, asthma-related symptoms 3 times a week or more, or symptoms causing waking at night. In that case, we should offer a low dose of an ICS, for example standard beclomethasone 100mcg, one or two inhalations twice daily.

 

The next issue is Sarah’s depression, for which she takes sertraline 50 mg daily. We don’t know how long she has been taking them for but NICE recommends that SSRIs are taken for at least 6 months (and for some time after symptoms remit). So we should assess how long she has been on it and if she wishes to continue treatment or whether the time has come to consider stopping after a gradual reduction of the dose.

 

It is worthwhile mentioning that the BNF says that SSRIs should be prescribed with caution in diabetes because SSRIs can affect diabetic control. We are advised to monitor blood glucose when starting or stopping an SSRIs.

 

Finally, let’s have a look at her obesity. Her BMI is 32. Apart from the obvious dietary and lifestyle changes, NICE recommends a number of pharmacological treatments for obesity such as orlistat, liraglutide and semaglutide. They all have different BMI thresholds and I would advise you to look at the different criteria before prescribing.

 

Except orlistat, liraglutide and semaglutide can only be given for obesity by a secondary care service. However, as discussed earlier and based on our clinical judgement, we could deviate from the NICE guideline and give a GLP1 receptor agonist for her diabetes instead of an SGLT2 inhibitor.

 

We have come to the end of this episode. Remember that this is not medical advice and it is only my summary and my interpretation of the guidelines. You must always use your clinical judgement.

Thank you for listening and goodbye.

Audio Player Image
Diabetes in Primary Care
Loading...