About this blog: information for everyone else

Hi there! I’m a GP and this blog consists of the notes I take on journal articles, as part of my continuing professional development. It is intended purely for my own use in recording and remembering the things I learn so that I can refer back to them in the medical setting. While anyone is welcome to read it, please do not take it as any kind of substitute for seeing your own doctor for any medical-related queries or problems.

Posted in About this blog | Leave a comment

About this blog: information for appraisers

I use this blog to record my learning points/reflections arising from my CPD. It’s a very handy way of doing it, as it lets me refer back to it easily and use the search function and category menu to find the notes I’ve made on a topic when needed.

My CPD typically consists of a mixture of online modules, articles which I come across in the course of routine journal reading and find useful, answers to DENs that I’ve looked up, and the usual regular courses in CPR and (on a three-yearly basis) child safeguarding. Of the online modules, some are compulsory; the rest I pick based on DENs I identify, either specific (“I wonder what the answer is to this question that arose with this patient?”) or general (“I really need to brush up on my knowledge of skin cancers”).

To read my overall notes for this year, just click on the picture at the top to get to the home page; you can then read the posts in reverse date order (newest first). If you want to check the list of posts on CPD that isn’t from learning modules (as CPD credits for those are recorded separately), those are under the category ‘Credits ____’ with the current year. You can find this by scrolling down the category menu on the right (under the archive menu). I will normally also put in a direct link to this category in my appraisal record.

Hope that’s clear, but happy to discuss any questions at appraisal.

Posted in About this blog, Credits 2016, Credits 2018 | 2 Comments

Onset of action of antidepressants

For twenty-five years I’ve accepted without question the standard teaching that SSRIs take 2 – 3 weeks to kick in and often longer. Then, due to a family situation, I actually went and looked it up, and discovered that the research doesn’t support that at all. It actually appears that they start having an effect within 1 – 2 weeks if they’re going to have an effect at all.

This meta-analysis showed that if Mirtazapine hasn’t kicked in within the fortnight, it’s very unlikely to do so later. This one, looking at SSRIs, has some nice little graphs (Fig 3) showing the same thing, and also identified that improvement starts in the first week and can continue for up to six weeks. This one has fewer helpful figures in the abstract but backs up the general idea. Just in case I end up using tricyclics or MAOIs for some unlikely reason, this study looked at some of the older drugs and found basically the same thing. This editorial has a summary of the research, though I think the above links cover the main points.

Very helpful; means that, if a tablet isn’t working, I’ll be able to recognise that earlier and move onto the next thing.

Posted in Credits 2019, Medication, Psychiatry | Leave a comment

Anticoagulation in people with a falls risk

I’ve been very wary of anticoagulating elderly frail people at risk of falls – it feels very antithetical to ‘First, do no harm’ – but it seems I might have been being overcautious. I looked this up as part of deciding what to do about anticoagulation in a patient of mine who is at high risk of both falls and CVA.

The NICE guidelines on this are clear: Do not withhold anticoagulation solely because the person is at risk of having a fall. Curious, I looked up the evidence base for this one, but it only said ‘The GDG agreed that there was no link between falls and bleeding risk. They were concerned that excessive emphasis was often place on falls as a risk factor and wanted to ensure that patients were not denied anticoagulation treatment for this reason.’ (pg 165 of the evidence basis for the guideline)

For some solid evidence, I found DonzĂ© et al, Am J Med 2012; 125(8): 773 – 8, a prospective cohort study which found that high falls risk in people on warfarin wasn’t actually connected to increased risk of major bleeds. I suppose the reason is that most falls don’t result in a serious laceration or head injury.

The most helpful discussion was here; went through several studies and came up with the tidbit that, weighing up the risk from falls, a person on anticoagulants would have to have at least 295 falls a year before their risk of a major bleed from a fall would outweigh their risk of CVA from not taking an AC. (This was based on warfarin; the article pointed out the figures could be even higher than that for NOACs.)

So, good to know; I’ll start my patient back on the NOAC.

Posted in Anticoagulants, Cardiovascular, Credits 2019, DENs, Medication | 3 Comments

Withdrawal of CANH in patients without capacity

The BMA guidance.

General points:

  • Where capacity is in doubt, the principles for establishing it are the usual ones, including the importance of doing everything possible to maximise capacity (looking at different ways of communicating and of allowing the person to communicate, trying at different times where capacity might be fluctuating)
  • In people without capacity, start from the strong assumption that CANH is what they would want, but be prepared to overturn this in the case of evidence to the contrary.
  • The decision of whether CANH is appropriate in a person without capacity should be made before it is started and then reviewed at intervals.
  • If the person does not have capacity, first step is to check for a valid and applicable ADRT (remember, it must cover CANH situations to be valid here). If none, check whether there is someone with LPA for H&W as the decision will fall to them if the POA gives them the power to make it and unless anyone concerned has doubts that they’re acting in the person’s best interests. If not, then it becomes a best-interests decision by the doctor in charge of the person’s case (GP or senior clinician) in consultation with the patient if they’re able to contribute any opinion, any involved people (family, care staff, anyone whom the person has previously named as being someone to consult on decisions of that type, and anyone else who could reasonably be described as ‘interested’; it’s very broad) and with a second-opinion doctor from a relevant specialty. If the people involved can’t agree, it goes to Court of Protection. (All decisions on withdrawal used to go to Court of Protection, but it’s now been explicitly decided that if everyone concerned agrees that withdrawal of CANH is the best-interests decision then taking it to C of P isn’t necessary.)
  • It’s good practice to call a best-interests meeting when making the decisions above. Obviously, document meetings in detail.
  • Be honest when there’s uncertainty about the prognosis. If in doubt, seek advice from someone with experience in the field.
  • Look realistically at the person’s quality of life, and about how it can best be maximised.
  • Where there’s disagreement about the best interests, look at getting further clinical opinions, having further discussions, and/or using a medical mediation service. If all of that really isn’t working, then it’s Court of Protection time. People who disagree with the clinician need to be supported as to how disagreements are brought to the C of P.
  • If a clinician has a conscientious objection to carrying out the decision (whether this is withdrawal or continuation of CANH), they need to find a colleague who will do it.
  • Best-interests decisions to provide or to continue CANH should be reviewed at least every 12 months; if the patient is less stable they should be reviewed every 6 months, and if the clinical picture changes significantly or new information comes to light then a review should be called.
Posted in Credits 2019, Medicolegal | Leave a comment

Intrauterine contraception (copper or progestagenic)

When to insert, and when precautions are needed

Note the following general principles:

  • IUC should not be inserted where there is a possibility of existing pregnancy (previous sex without a reliable method of contraception being used correctly for every episode of intercourse)
  • However, the Cu-IUD can be used as PCC for up to 120 hours.

This means that Cu-IUD insertion timing is fairly simple: If a woman has previously been abstinent or correctly using contraception, it can be inserted at any time regardless of menstrual cycle phase and is effective immediately. (The one caveat on this is that, if switching from one type of IUC to another, advise the woman to use other contraception or abstain for the 7/7 before the switch, just in case there’s a problem with the insertion of the new coil.)

It can also be inserted immediately after the end of a pregnancy in the first or second trimester* (although expulsion rates are slightly higher, so discuss that with the woman), or at any time from 4 weeks post-delivery, as long as there is no significant risk of pregnancy. Although in the UK it isn’t licenced for insertion in the first four weeks after delivery, a number of countries insert it in the first 48 hours after delivery and the breadth of available experience of this as a successful strategy mean that the FRSH approves this 48-hour post-delivery window as a time during which IUC can reasonably be inserted off-licence. (Again, be aware that expulsion risk is higher.)

*For medical TOPs, this applies to the second visit, once it has been ascertained that the POC have been passed.

Where UPSI or contraceptive failure has occurred, timing of Cu-IUD insertion is more complicated as it can only be done within the window where it can potentially work as PCC. (This is the case even where oral PCC has been used; either insert within that window, or wait the 21 days to do a pregnancy test and exclude pregnancy, using an alternative contraceptive method if needed in the meantime.)

I find the easiest way to remember the PCC window for the Cu-IUD is that it’s within the 5 days (120 hours) after the earliest time at which both sperm and egg can be deduced to have been around. This means that either the earliest UPSI in the cycle must have been within the past 120 hours (therefore sperm not around before then) or the person must be in the first 19 days of their cycle, if a normal-length cycle (therefore, egg not around before the 5-day window assuming ovulation on Day 14).

For LNG-IUS insertion, things are a little more complicated as the ‘avoid insertion in case of possible pregnancy’ rule is still in place, but the IUS doesn’t have a reliable enough PCC effect to count on. (This seems to be because, although the IUS does work via anti-implantation effect, this works via progestagen-induced endometrial atrophy rather than endometrial inflammation and hence takes longer to kick in.) For IUS, the rules are:

  • During a menstrual cycle, the IUS is only licenced for insertion in the first seven days and is then considered effective immediately.
  • Off-licence, the FSRH approves insertion at any time in the cycle if it is reasonably certain the person is not pregnant. If inserted at any point from Day 8 onwards, extra precautions should be used for 7/7.
  • Within the first 7 days after the end of a 1st or 2nd-trimester pregnancy (second visit, if a medical TOP) although note slightly higher expulsion rates
  • >7 days after the end of a 1st or 2nd-trimester pregnancy if reasonably certain the woman is not pregnant, and with 7 days use of extra precautions.
  • Within 48 hours of delivery (possibly higher expulsion rates)
  • 4 weeks or more from delivery if reasonably certain the woman is not pregnant, and with 7/7 use of extra precautions unless inserted in the first 7/7 of cycle. (The extra precautions can be LAM if applicable and if correctly followed.)

When switching contraceptives, the general rule, from what I can see, seems to be that the IUS is considered immediately effective if switching from a method whose primary mechanism is anovulation (which, apparently, does not include the Cu-IUD, at least for these purposes). This means that you can switch with immediate effect if switching from:

  • Desogestrel POP
  • Implant, at any point up to the end of the three years
  • Injectable, at any point up to 14 weeks (98 days) from the administration of the injectable
  • CHC if in Weeks 2 or 3 of the pack or Day 1 of the HFI and the CHC has been used correctly and consistently for the seven days immediately proceeding the switch. (Note that, unlike with switches to other CHCs or POPs, we don’t count the second day of the HFI as being part of the grace period.)
  • LAM (check all criteria fulfilled)

However, use extra precautions for 7/7 if switching from:

  • Traditional POPs
  • CHC if in Days 2 – 7 of HFI or Week 1 of cycle
  • Cu-IUD; and advise extra precautions for 7/7 before the switch, as above (loss of post-coital effect from the outgoing Cu-IUD). (Logically, I would think that if switching in the first 7/7 of cycle you shouldn’t need the extra precautions afterwards, but this isn’t mentioned in the guidance.)
  • Barrier methods, unless in first 7/7 when counted under the ‘first week of the menstrual cycle’ rules above.

(For all of these except the Cu-IUD, the extra precautions can simply be continuation of the method for 7/7. Only thing is, I’m not sure whether this would be an issue with the diaphragm; could the strings damage it at all? Wouldn’t think so, and it’s not likely to come up often anyway.)


STI screening pre-insertion

Women who are at increased risk for STIs should be screened at least for chlamydia and probably for gonorrhoea as well. It’s normally better for the results to be back and treatment completed before IUC insertion, but if someone with asymptomatic chlamydia needs EC it’s OK to insert an IUD and start the antibiotics the same day. Giving prophylactic antibiotics to asymptomatic women where the test results aren’t known pre-insertion is probably not cost-effective due to the low rate of complications overall.

Women who are symptomatic should be tested, treated, and wait until symptoms resolve prior to having IUC. Remember to advise them to abstain until they and any current partners have finished treatment, or until 7/7 after having single-dose azithromycin as treatment.

If someone with IUC already in situ is diagnosed with pelvic infection, it isn’t necessary to remove the IUC, though it might lead to better short-term outcomes (this needs to be balanced against risk of pregnancy from sex in previous 7/7). The IUC probably should be removed in anyone not responding to antibiotics within 72 hours of treatment.

Other health considerations

Rates of side-effects are similar between copper and LNG IUCs.

There is evidence that both types provide protection against endometrial and cervical cancer. What little evidence we have on breast or ovarian cancer is conflicting.

LNG-IUS are associated with an increased rate of functional ovarian cysts; between 1 and 10%.

Period problems, including dysmenorrhoea, related to IUC (either type) tend to improve over time. Spotting and non-period-related types of pelvic pain don’t.

Recurrent BV is associated with use of the Cu-IUD. It isn’t clear whether it’s associated with LNG-IUS use or not. While there are theoretical reasons why IUC could make a difference to rates of vulvovaginal thrush, in practice it doesn’t seem to.

(Notes from FSRH guidance, downloaded from here.)

Posted in Contraception, Credits 2019 | Leave a comment

Multiple sclerosis

(Notes on the BMJ module regarding NICE guidelines)


There wasn’t much on this, but common presentations include:

  • Optic neuritis (loss of/reduction of vision in one eye, associated with aching)
  • Double vision
  • Ascending sensory disturbance (e.g. from feet to abdomen), with or without associated weakness.

Initial investigations

The point of investigating before neuro referral is to rule out other causes for the symptoms (MS diagnosis is complicated and can only be done by a consultant neurologist, which is fine by me). Most of these were the usual line-up: renal/liver/thyroid function, calcium, glucose, B12. However, the guidelines also advise doing:

  • Inflammatory profile (ESR, or PV in our case/CRP)
  • HIV serology


People with MS should be given a clear point of contact for concerns and should have yearly specialist follow-up. They should also be given a lot of information, which is obviously going to be hard to take in at the time, so it’s worth the GP checking that they have:

  • Information on how to contact the MS team with questions
  • Information on charities
  • Instruction to contact the DVLA
  • Advice on their rights involving employment/benefits.



…is an even worse idea than usual, since studies clearly show it to be linked to increased risk of relapses and progression. Ex-smokers seem to have the same risk as never-smokers, so stopping really can improve risk profile here.


Is beneficial, even for people with mobility problems/fatigue problems. If fatigue problems, consider referral for a supervised exercise programme. If mobility problems, refer to a rehabilitation team or physio with expertise in MS.

Don’t use something called fampridine; it’s not cost-effective. Fine by me, I’d never heard of it.

Some other things not to bother trying

  • Vitamin B12 unless diagnosed deficiency
  • Vitamin D unless in an at-risk group for other reasons as per usual Vitamin D guidelines
  • Omega 3 or Omega 6 fatty acids

According to the guidelines, there is no evidence for any of these.


Flu vaccination might sometimes trigger relapse, so consider carefully whether to use it or not and make an individualised decision weighing up the risks and benefits.

Remember that live vaccines might be contraindicated in DMTs.


Can present as intense stiffness, muscle jerking, or muscle spasms. Note that underlying conditions such as infection or constipation can make spasticity worse, so these should be identified and treated.

  1. 1st-line treatments: gabapentin, baclofen
  2. 2nd-line treatments: tizanidine, dantrolene
  3. 3rd-line: benzodiazepines such as clonazepam (which can also help with nocturnal spasticity).

Start at a low dose and titrate up gradually over weeks until either symptoms ease or the maximum tolerated dose is reached. Encourage patients to do their own auto-titration if possible. If there is partial relief at maximum tolerated dose, add another drug; if no relief, change to another drug. Note that too high a dose of the drugs can cause (reversible) muscle weakness.

If all that fails, refer to specialist services.

Sativex is not currently considered cost-effective for spasticity related to MS.


Treat SOS; exclude infection (especially UTI), start high-dose steroids (with bone protection if needed), and let the team know. Steroids might be oral or IV. Note that they can precipitate depression.

Urinary symptoms

Check a post-void scan for residual volume. If <100 ml, use anticholinergics. If >100 ml, try intermittent self-catheterisation.

Posted in Neurology | Leave a comment

Glucose control in Type II diabetes

(As per BMJ module ‘Clinical pointers: Blood glucose control in Type 2 diabetes’)

I’ve used sulfonylureas as second-line diabetic drugs for years on a ‘devil you know’ basis, but it now seems that’s out of date; one meta-analysis showed them to be associated with a higher risk of cardiovascular events (Diabet Med 2017; 24(12): 1412 – 18). (Another meta-analysis, in JAMA 2016; 316: 313 – 24, found all meds came out about the same in this aspect, so the jury’s out, but it does make me think perhaps I’m better getting with the programme and giving DPP-4s, at least until something dire inevitably comes out about those as well.)

NICE recommends a stepwise approach:

  1. HbA1c >48 on lifestyle measures: start metformin and aim for 48.
  2. HbA1c >58 on metformin + lifestyle measures: start a second drug and aim for 53. (I’m not sure how this is supposed to fit with aiming for 48 in people on non-hypoglycaemic drugs.)
  3. HbA1c >58 on two oral drugs + lifestyle measures: start a third drug, which can be insulin. Aim for 53.

Quick recap of the possible options for dual or triple therapy:

Second-line therapy with metformin:

  • Sulfonylurea
  • DPP-4 inhibitor
  • Pioglitazone
  • SLGT-2 inhibitor

Third-line therapy combos with metformin:

  • SU with any of the other three above
  • SLGT-2 with either SU (which obviously follows from the line just above) or pioglitazone
  • Insulin with metformin; review the need for other antihyperglycaemics when insulin is started.

To put that another way, the only two combos of the list of second-line drugs that can’t currently be given as part of oral therapy are DPP-4s with either pioglitazone or SLGT-2. That narrows things down; since I’ve now decided to give DPP-4s as my second-line drug with metformin unless there’s some good reason not to, my third-line, if not referring for insulin, will normally be a sulfonylurea. OK, I… I think I’ve got it.

But where do GLP-1 inhibitors fit in?

As combination therapy with metformin plus sulfonylurea in patients who fit anyof the following criteria:

  • BMI >35 (can be less in some ethnic minority groups) and this is causing specific medical or psychological problems
  • Weight loss would benefit other co-morbidities
  • Insulin therapy would have significant occupational implications

And what about people who can’t take metformin?

Firstly, if the problem is side-effects then consider trying a modified-release preparation before giving up completely. If that doesn’t work/isn’t an option:

Alternative first-line:

  • DPP-4i (target 48)
  • Pioglitazone (target 48)
  • Sulfonylurea (target 53)

Second-line: Any two of the above combined. (So, in other words, alternative first and second-line treatments are similar to the with-metformin second and oral third-line treatments except that SLGT-2 inhibs are off the table and DPP-4is apparently can be used with pioglitazone if metformin isn’t around, all of which is sounding more complicated than figuring out who can and can’t sit with whom at extended feuding family Christmas dinner.)

Third-line: Insulin.


Posted in Diabetes, Medication | Leave a comment

Assessment of purpuric/petechial rashes

This was a BMJ module, based on this article.

Steps in assessing purpuric/petechial rashes:

  1. Assess for acute illness and consider the possibility of meningococcal sepsis. If suspected, administer parenteral antibiotics and admit urgently.
  2. If meningococcus is not the suspected diagnosis, and if the patient is a child or young person (‘young person’ is left undefined here; from what I could find from the source, it seems to mean someone under 24), then send them for same-day specialist assessment. If patient is an adult, arrange FBC within 48 hrs, or sooner if indicated, ensure you have emergency contact details, and advise them to avoid aspirin/NSAIDs while awaiting results.
  3. The above should exclude anything serious, so if the FBC is normal then manage according to the most likely cause.

Causes of petechiae/purpura in adults

(Note that there are slightly different lists of causes in neonates or children. I haven’t bothered transcribing those, on the basis that, as per above, any children with petechial rashes are going to be referred directly to paediatrics anyway, and thus if I get a diagnostic problem in this area it’ll be with an adult.)

  • ITP
  • Primary bone marrow failure (leukaemia, myelodysplasia)
  • Secondary bone marrow failure (metastases)
  • Drug-induced
  • Nutritional deficiency (didn’t list causes; B12 deficiency is the one I could think of)
  • Degenerative causes such as senile purpura
  • Mild coagulopathies (which can show up only in adulthood; if very mild, they might only show up in the elderly)
  • Acquired haemophilia (which I’d never heard of, but it’s on the list so I guess it’s a thing).
Posted in Dermatology, Don't miss, Haematology | Leave a comment