Find Courses Find Online Courses Find Booklets Find Appraisal Essentials Basket 0

Hot Topics Blog

40 Step Test: Order in the Chaos?

Dr Neal Tucker - 22 Apr, 2020

It’s human nature to try and make order out of chaos. It’s no surprise that in the madness of the coronavirus pandemic we want to grab hold of anything which could provide that order.

The Roth test was suggested as a potentially useful tool. It was rapidly incorporated in to guidelines and then rapidly removed as it’s failings were highlighted. But we’re still looking to ease our uncertainty.

The latest crutch is the 40 Step Test – a simple test where patients walk 40 steps and we measure for oxygen desaturation. This is being widely used in hot hub car parks throughout the UK.

Where did this test come from?

You tell me…

The only current national guideline with any reference to the 40 Step Test is an A&E/AMU assessment tool for possible COVID-19 patients which can be found on NHS England’s website.

It is suggested to be performed in ‘well’ patients with otherwise normal vital signs before discharge home and if desaturation is present further review is required. There are no references and ominously there is an asterisk next to the test on the flow chart, which is not qualified anywhere in the document. If only we knew what it wanted to alert us about…

Is it a good test?

The truth is we simply don’t know. A quick literature search shows no data on the 40 Step Test. However, the Centre for Evidence-Based Medicine has now published an extensive review of similar tests that might help us.

The most applicable validated test for primary care is the 1-minute sit-to-stand (1MSTS) test. It is a simple test validated in patients with chronic lung disease and is effective at unmasking exertional desaturation. The 40 Step Test is slightly lower exertional stress but could be considered analogous, albeit with a probable reduction in sensitivity.

The most infection relevant exertional study is a 30 year old study in patients with HIV-related pneumocystis carinii pneumonia. PCP shares many similarities with COVID-19.

In this study, many patients with PCP and normal resting sats desaturated on exertion (a 10min cycle), although it’s also true that many didn’t so this is not a sensitive test. Interestingly, patients with other acute lung infections, e.g. pneumonia, did not have exertional desaturation.

The CEBM suggests that the 1MSTS, or the unvalidated 40 Step Test, could both be helpful in the assessment of COVID-19 but there are a number of caveats.

  • Firstly, the accuracy of these tests in COVID-19 is unknown.
    • They are likely to be fairly specific but not sensitive.
    • Desaturation likely indicates COVID-19 in the absence of chronic lung disease and greater severity of illness.
    • Even a small desaturation is important, and a drop of 3%+ indicates severe disease regardless of the amount of exertion required.
    • A negative test does not exclude severe disease and should not reassure us.
    • Regardless of oxygen saturations, the ability to eyeball a patient while doing some exertion can be very helpful for clinicians.
    • No test can predict against a rapid deterioration in COVID. Safety net, safety net, safety net.
  • Secondly, consider patient safety.
    • Particularly with remote use, e.g. via video consultation, how safe it is to encourage exertion in this unwell patient? No-one wants to precipitate an arrest on a video call…

No test is perfect. These tests are not concrete. These tests should not be used in isolation. They do have a role and can be helpful but are only one part of our assessment, helping to build the clinical picture. Our attempts to find order in chaos will continue, but for now remember our best assets remain our knowledge, our experience and our instincts.

Neal Tucker

Find this blog useful? You can quickly add CPD to your account by writing a reflective note about the post you have read.

Log in to your NB Dashboard and use the 'Add Reflective Note' button at the bottom of a blog entry to add your note.


COVID-19 and the ‘second week deterioration’

Dr Simon Curtis - 2 Apr, 2020

As we manage patients with COVID-19 remotely and try to avoid face to face clinical contact unless essential, understanding the natural course of the infection and the potential complications is vital to give appropriate safety netting advice. With the lack of central ‘top down’ guidance, GPs and primary care teams are doing an amazing job in working out largely for themselves how to manage COVID-19.

A key paper in the Lancet 2020 has already taught us that patients typically have a mild, viral illness for a week and then may clinically deteriorate with respiratory distress syndrome and sepsis in the second week of the illness. This was based on data from hospitalised patients in Wuhan. Respiratory failure from ARDS is the most common cause of ICU admission and death. We therefore need to carefully safety net our patients in primary care for a possible ‘second week deterioration’, with a special emphasis on dyspnoea.

Clinical experience from GPs here in the UK, anecdotally from us and widely shared by GPs across social media, is that high fever, cough and respiratory symptoms are commonly encountered in the second and sometimes third week of the illness. This has thrown up a common question: is this all viral COVID-19 or should patients be given antibiotics to cover possible secondary bacterial pneumonia?

These patients might fulfil NICE clinical criteria for community acquired pneumonia but of course this cannot be extrapolated to COVID-19 and we just have no published evidence to help us. However, our clinical experience and that of other GPs is that some patients with fever and cough into the second week can respond quickly to antibiotics and the persistent fever is driven by secondary bacterial infection. Some local guidelines such as Barnet Primary Care Guide COVID-19 recommend antibiotic cover (amoxicillin + clarithromycin, or doxycycline) for patients with moderate symptoms. Other local guidelines are not recommending this approach, highlighting the uncertainty. However, what is clear is that we don’t want to miss treatable illness and once people end up in hospital most are getting antibiotics for that reason. I have got some local expert ID advice here in Oxford which supports this ‘low threshold for antibiotics’ approach.

New knowledge for some of us is that there is accumulating evidence Lancet 2020 that a subgroup of patients with severe COVID-19 may deteriorate rapidly in the second week with a ‘cytokine storm’. This is cytokine release syndrome, a systemic response to the virus when the immune system trips into over-drive with a systemic hyper-inflammatory response leading to a flood of immune cells and inflammatory proteins into the lungs and other organs leading to respiratory distress syndrome and multiple organ failure. Crucially cytokine storm can occur rapidly, like sepsis, leading to a rapid clinical deterioration and death. It is thought that some of the younger, fitter people who become very unwell and die from COVID-19 may have a genetic ‘host’ factor that puts them at greater risk of a cytokine storm. The sooner this syndrome is recognised, the better the prognosis.

So, whilst there is still a lot of COVID-19 uncertainty, useful leaning points for us are that:

  • Whilst the vast majority of patients in the community will have a mild illness with COVID-19, we need to be aware of the risk of the ‘second week deterioration’ and safety net appropriately
  • In patients with on-going fever and respiratory symptoms and signs, it is not possible to clinically distinguish between viral and secondary bacterial pneumonia so have a low threshold for antibiotic cover
  • Patients may deteriorate rapidly due to either sepsis or the cytokine storm syndrome, requiring urgent admission, and this needs appropriate safety netting


Find this blog useful? You can quickly add CPD to your account by writing a reflective note about the post you have read.

Log in to your NB Dashboard and use the 'Add Reflective Note' button at the bottom of a blog entry to add your note.


Q&A from Hot Topics live webinar Saturday 25th April 2020

Dr Rob Walker - 29 Apr, 2020


  • Exercise and hypertension - how much should we be recommending?

A meta-analysis from the Br J Sports medicine in 2018 showed some reasonably convincing evidence that overall exercise produced consistent improvements in blood pressure, similar to that seen by common anti-hypertensive drugs. The studies, however, were inevitably heterogeneous, so picking out how much exercise is optimal, and which type was not possible. Both endurance and resistance based exercises showed benefits.

The Academy of Royal Colleges recommend physical activity 5 times per week, for 30 minutes each time in total (can be broken up into smaller 10-15 min chunks), of moderate-intensity = 150 minutes/week total.

What about lower intensity activity? Read my blog from earlier this year discussing more recent evidence suggesting even light levels of activity have mortality benefits.

  • Is an ECG recommended in patients with hypertension and what are we looking for?

Yes NICE recommend an ECG for newly diagnosed hypertensive patients. We need to assess for signs of LVH, and also to confirm the underlying rhythm.

  • What about home blood pressure monitors? How do patients use them, are they validated?

There is even more need for remote monitoring of BP at the moment. Home BP monitors are recommended by all international guidelines as an option for monitoring. The British and Irish Hypertension Society have excellent information about home monitors/which are validated/where to get them (click here). Patients should do twice daily readings for 7 days, discard the first day's readings then take a mean of the remaining readings to get your average. Remember you need to add on 5/5 mmHg to the home average to get an equivalent clinic reading (targets and treatment thresholds based on clinic readings).

  • In the Hygia trial did they really also take their diuretics before bed?

YES! The Hygia trial has been a practice changing piece of evidence for me, by showing better CVD outcomes if BP medications are taken before bed, rather than in the morning. A number of delegates quite rightly questioned if diuretics were also taken before bedtime in the trial, and yes they were. A large number of participants were on diuretics (mostly hydrochlorothiazide) and if in the bedtime arm took ALL medications before bed (they were not split to some am, some pm dosing). What about concordance I rightly hear you cry?! Concordance in the study was very good with this regime, but is obviously a somewhat idealistic trial setting. Ultimately the most important thing is that patients continue to take their medications, so if they cant tolerate them before bed, as may be the case with diuretics, then the pragmatic solution is to split when they are taken, and have the diuretics in the morning.

Lipids and stroke

  • How do the European guidelines differentiate between the risk groups, when deciding which lipid target to aim for?

The 2019 European lipid management guideline recommends basing lipid targets on risk category, briefly summarised:

Very high risk (e.g. documented atherosclerotic CVD, diabetes with end organ damage or ≥3 major risk factors, or T1 diabetes for >20 yrs, CKD stage 4) = LDL target <1.4 mmol/L

High risk (e.g. total cholesterol > 8 mmol/L, LDL > 4.9 mmol/L, diabetes ≥10 years or additional risk factor, CKD stage 3) = LDL target < 1.8 mmol/L

Moderate risk (e.g. T1 diabetes aged <35 or T2 diabetes aged <50 with diabetes duration <10 years) = LDL target < 2.6 mmol/L

The difficulty for us is the other way they assess risk is by 10 year CVD MORTALITY risk - this is different from all the CVD risk calculators we are used to using, which give 10 year CVD EVENT risk (not mortality) so the scores they use cannot be tallied with QRISK3 or ASSIGN….

  • If you’re working in Scotland should we use the ASSIGN CVD risk score or QRISK3?

The ASSIGN score is still recommended by SIGN for use in Scotland for CVD risk estimate, as it has the benefit it was based specifically on Scottish population data; the downside compared to QRISK3 is it doesn’t have a number of other risk factors incorporated into it that have been associated with increased CVD risk over the more recent past (e.g. AF, migraine, SLE, antipsychotic medications etc.) so if you have patients with these additional risk factors you should take these into account when assessing overall CVD risk.

  • What do we do with the fit, younger adults who have had screening and found to have elevated lipid levels, but low CVD risk score? Should we offer them statins?

I’m sure a question we have all been asked on many occasions…In the webinar, we presented data from the Lancet in 2019, which showed clear associations between worse CVD outcomes and raised lipid levels, especially if levels raised <45 years of age, with a ‘dose’ response (i.e. the higher the lipid levels the higher the CVD risk later in life). We know the traditional 10 year CVD risk scores underestimate lifetime CVD risk, but does this mean statins will reduce CVD risk if started in younger adults with raised lipids as a sole risk factor? Well, we simply don’t know - the Lancet study modelled the predicted improvement in CVD events if younger adults were started on statins with impressive results…however, this is only based on MODELLING and is not there same as real-world data showing statins do actually work. So we don’t have the answer to this question and need to discuss that uncertainty with patients that ask.

  • What about statin intolerance?

Whatever trial data says, we know lots of people who struggle to tolerate statins for a multitude of reasons. Current guidance suggests starting on high doses of potent statins for people after atherosclerotic CVD events, but what if they can't manage their 80mg of atorvastatin? Well, there is some evidence that re-starting at a much lower dose, and slowly titrating means many of these people will manage statins over time, so don’t lose hope! Ultimately we can assure patients that ANY dose of statin has been shown to be beneficial after CVD events, even ifs they cant tolerate higher doses.

  • What dose of statin/targets should we be aiming for with patients post-stroke or TIA?

Patients who have had atherosclerotic stroke or TIA are high risk for future events and should be offered high-intensity statins e.g. atorvastatin 20-80mg. We should aim to get their non-HDL cholesterol (total cholesterol-HDL cholesterol) at least 40% lower from baseline. We presented in the webinar new data from France which showed more intensive targets aiming for LDL < 1.8 mmol/L reduced the risk of further events, so this would be an alternative strategy. Statins are generally NOT recommended for haemorrhage stroke.


  • Personal Asthma Action Plans (PAAP) - are there PAAP apps patients can use? What do we put in the amber/red sections with respect to %PEFR drops?

I have not seen any app PAAPs yet (but no doubt there are some out there!), so suggest we base advice on the well respected Asthma UK which recommends taking a picture of the PAAP, once completed, and sharing it with friends and family as well. The decision about what level of PEFR drop to put in the amber/red sections will need to be personalised (I would suggest taking advice from your excellent respiratory nurses!). Just a reminder that the SIGN/BTS guideline assess a drop in PEFR to 50-75% of best/predicted = moderate acute asthma; 33-50% = severe acute asthma; <33% life-threatening asthma.

  • Can calcium channel blockers cause chronic cough?

This was flagged as a question by a few delegates and I’ve had the odd patient who has been recommended to stop their CCB by respiratory physicians as a potential cause of chronic cough. I’m not really sure of the mechanism, or how evidence-based this is (I could not find any data supporting this…) but it is listed as a rare side effect in the BNF.

  • Do ARBs cause chronic cough?

We all know ACEI is a common cause of chronic cough, and we are recommended to switch to an ARB if this happens but do ARBs also cause cough? The best data I could find was a meta-analysis in 2012 which suggests that the rate of cough with ARB was similar to that of placebo.

Hand osteoarthritis

We presented the HOPE study which suggests a 6 week course of oral steroids can give short term relief for severe symptomatic inflammatory hand OA, and this threw up a few interesting questions…

Could we use IM depo-medrone instead of oral prednisolone? Until the HOPE study, there was no evidence to support oral (or other routes) corticosteroids for hand OA, so it may work, but there is currently no evidence to support IM corticosteroids for hand OA.

Why not use NSAIDS instead of oral steroids? Quite right…this would be the recommended first-line treatment - both oral and topical NSAIDs have data supporting their use in hand OA. The HOPE study was very much looking at the severe end of the spectrum who may not have responded to usual first-line treatments.

What about long term NSAIDS for hand OA? Current guidance suggests if using oral NSAIDS, we should use them at the lowest effective dose for the shortest possible time - if considering long term oral NSAIDS we need to balance the benefits vs CV, renal and GI harms.

What about DMARDS for hand OA? Can methotrexate work? Some of our colleagues mentioned they have had some of their patients trialled on methotrexate….The evidence seems pretty limited and the latest EULAR review in 2018 suggested there was no evidence to support DMARDS in hand OA; a small study was published in 2019 that suggests methotrexate may reduce progression of joint damage (on MRI) in erosive hand OA, but it did not improve pain or function.

Mental Health

  • Which anti-depressants have the highest risk of withdrawal symptoms?

A recent literature review suggests risk of withdrawal side effects can be split into:

High risk - paroxetine, venlafaxine, TCAs

Moderate risk - citalopram, escitalopram, sertraline, duloxetine

Unclear (insufficient evidence) - mirtazepine

  • How long should we take to wean off antidepressants to reduce the risk of withdrawal side effects?

We presented some data which showed many antidepressants cause withdrawal side effects for much longer after stopping than originally thought (in some cases months). The Royal College of Psychiatry recommend that the time taken to come off antidepressants should be proportional to the length they have been on them e.g. if been on antidepressants for years it may take many months to slowly wean off them.

  • If we are considering switching to a long-acting drug (e.g. fluoxetine) due to withdrawal side effects from a short-acting drug (e.g. venlafaxine) how do we do this?

We had lots of questions about this, and thank you so much to all of you who sent in lots of useful information! This included GP notebook, the Maudsley guidelines, but there is good information in CKS on switching antidepressants, which was very recently updated in March 2020 (again thanks for the pointers!) - click here for CKS

  • What dose of quetiapine is used for augmentation in treatment-resistant depression and can this be started in primary care?

The Cochrane review we presented that showed some evidence for augmentation with quetiapine for treatment-resistant depression (at least in the short term) looked at studies with quetiapine doses ranging 150-300mg daily; whether we can/should start this in primary care will depend on a number of factors e.g. your own expertise, access to secondary care support etc.

  • What about lithium for treatment-resistant depression?

BMJ Best practice states that ‘lithium augmentation remains the best evidence-based approach’ for treatment-resistant depression, but is hampered by the narrow therapeutic window of the drug and that it would have to be started by secondary care.

Find this blog useful? You can quickly add CPD to your account by writing a reflective note about the post you have read.

Log in to your NB Dashboard and use the 'Add Reflective Note' button at the bottom of a blog entry to add your note.