EMREE practice questions

Correct answer: C. Replace magnesium intravenously while continuing potassium

Hypomagnesaemia causes refractory hypokalaemia because low intracellular magnesium allows potassium to leak through renal outer medullary potassium channels, so the kidney keeps wasting it. Potassium will not correct until magnesium is replaced. Loop diuretics and long-term proton pump inhibitors are common causes, and both lower magnesium. Magnesium is given intravenously when the deficit is severe or the QT is prolonged, and potassium is continued. Fast peripheral potassium infusion risks arrhythmia and vein injury, and potassium-sparing diuretics do not correct the magnesium deficit.

• Why A is wrong. A rate of 40 mmol/hour is unsafe through a peripheral line and does not fix the cause.
• Why B is wrong. More potassium will not work while magnesium is low, and the QT interval makes delay unsafe.
• Why D is wrong. Spironolactone may help later but it does not treat the acute deficit or the long QT, and stopping potassium is unsafe.

Topic: Renal: hypokalaemia that does not respond to potassium replacement. Reference: Huang CL, Kuo E. Mechanism of hypokalemia in magnesium deficiency. J Am Soc Nephrol 2007;18:2649-52

Correct answer: B. Oxygen, analgesia, antibiotics and incentive spirometry, with early specialist review for exchange transfusion

A new pulmonary infiltrate with fever or respiratory symptoms in sickle cell disease is acute chest syndrome, a leading cause of death in these patients. Management is supportive and active: oxygen to correct hypoxia, careful analgesia without over-sedation, broad-spectrum antibiotics covering atypical organisms, incentive spirometry, cautious fluids to avoid pulmonary oedema, and early haematology and critical care input. Simple or exchange transfusion is used for worsening hypoxia, and haemoglobin is usually not raised much above about 100 g/L because high viscosity can worsen sickling. Embolism is not the first explanation, and oral antibiotics with early discharge under-treat a hypoxic patient.

• Why A is wrong. Embolism is possible in some patients, but empirical anticoagulation is not the standard first treatment of a new infiltrate with fever.
• Why C is wrong. Aiming for a haemoglobin well above about 100 g/L increases blood viscosity and can worsen sickling.
• Why D is wrong. He is hypoxic with an evolving infiltrate, so oral treatment and early discharge is unsafe.

Topic: Haematology: acute chest syndrome in sickle cell disease. Reference: Howard J, Hart N, Roberts-Harewood M, et al. Guideline on the management of acute chest syndrome in sickle cell disease. Br J Haematol 2015;169:492-505

Correct answer: A. Admit for intravenous isotonic fluids, monitoring and thromboprophylaxis

This is severe ovarian hyperstimulation syndrome with haemoconcentration and oliguria from third-space fluid loss. Treatment is supportive: admission, careful intravenous fluid replacement, monitoring of fluid balance, electrolytes and haematocrit, drainage of tense ascites if needed and thromboprophylaxis because of the thrombosis risk. Diuretics worsen intravascular depletion, and surgery is reserved for complications such as torsion or rupture.

• Why B is wrong. Laparotomy is not indicated without evidence of torsion or rupture.
• Why C is wrong. Diuretics deplete an already haemoconcentrated intravascular volume and can precipitate renal failure.
• Why D is wrong. Fluid restriction worsens haemoconcentration and discharge is unsafe with oliguria.

Topic: Infertility: hyperstimulation syndrome after IVF. Reference: RCOG Green-top Guideline 5 The management of ovarian hyperstimulation syndrome

Correct answer: D. Penicillin desensitisation in a monitored setting, then benzathine penicillin G

Penicillin is the only recommended treatment that reliably treats the fetus and prevents congenital syphilis. A pregnant woman with a penicillin allergy should be desensitised in a monitored setting and then treated with benzathine penicillin G. Doxycycline is avoided in pregnancy, and azithromycin has documented treponemal resistance and unreliable placental transfer.

• Why A is wrong. Azithromycin resistance is well described and it does not reliably treat the fetus.
• Why B is wrong. Untreated syphilis at this titre risks stillbirth and congenital infection, so waiting is unsafe.
• Why C is wrong. Doxycycline is contraindicated in pregnancy because of fetal bone and tooth effects and is not adequate for fetal treatment.

Topic: Antenatal care: syphilis in penicillin allergy. Reference: CDC Sexually Transmitted Infections Treatment Guidelines 2021; WHO guideline on syphilis screening and treatment for pregnant women (2017); BASHH UK national guideline on the management of syphilis (2015)

Correct answer: B. 2 in 3

Each child of two carriers has a 1 in 4 chance of being unaffected non-carrier, a 1 in 2 chance of being a carrier and a 1 in 4 chance of being affected. Since she is known to be unaffected, the affected possibility is excluded, and the remaining outcomes are the 1 in 4 non-carrier and 2 in 4 carrier, which gives a probability of 2 in 3 of being a carrier. The error of answering 1 in 2 ignores the information that she is unaffected.

• Why A is wrong. One in two is the figure at conception, before it is known that she is unaffected.
• Why C is wrong. One in four is the chance of being affected, and not of being a carrier.
• Why D is wrong. One in three is the chance of being a non-carrier among unaffected siblings.

Topic: Genetics: carrier risk in an unaffected sibling. Reference: Emery's Elements of Medical Genetics, autosomal recessive inheritance and risk calculation; Cystic Fibrosis Trust information on genetics and inheritance

Correct answer: C. Echocardiography to measure the aortic root

The skeletal features, upward lens dislocation and a family history of sudden death suggest Marfan syndrome, caused by mutations in the fibrillin-1 gene (FBN1). The most dangerous complication is progressive aortic root dilatation leading to dissection, and so echocardiography is urgent to measure the root, with beta-blocker or ARB therapy and surveillance. Homocystinuria also causes tall stature and lens dislocation, but the lens moves downward, and it is associated with thrombosis and learning difficulty. A karyotype and abdominal ultrasound do not address the lethal complication.

• Why A is wrong. Homocystinuria shows downward lens dislocation, and the cardiac risk is the more urgent question.
• Why B is wrong. A karyotype does not diagnose Marfan syndrome.
• Why D is wrong. An abdominal ultrasound does not assess the main life-threatening problem.

Topic: Genetics: Marfan syndrome. Reference: Revised Ghent nosology for Marfan syndrome (Loeys BL et al. J Med Genet 2010); American Heart Association and ACC guideline for diseases of the aorta

Correct answer: D. Same-day referral to an ophthalmologist for dilated fundus examination

Flashes, new floaters and a spreading visual field defect in a myopic eye are the classic symptoms of retinal detachment, which is an ophthalmic emergency. Prompt repair, by laser, vitrectomy or a scleral buckle, can save central vision, especially before the macula detaches. Visual acuity may still be good at this stage, which is the window to act. Delay risks permanent loss of vision, and steroids and brain imaging have no role.

• Why A is wrong. A delay of weeks allows the macula to detach and vision to be lost for good.
• Why B is wrong. Steroids and neuroimaging do not address a retinal tear or detachment.
• Why C is wrong. Topical steroids do not reattach the retina.

Topic: Retinal detachment. Reference: Royal College of Ophthalmologists, retinal detachment guidance; Kanski's Clinical Ophthalmology, retinal detachment

Correct answer: B. Stop the injection, support the airway and give an intravenous bolus of 20 percent lipid emulsion

Perioral numbness, seizure and cardiovascular collapse minutes after injection of a long-acting local anaesthetic indicate local anaesthetic systemic toxicity. The Association of Anaesthetists guidance is to stop injecting, call for help, give oxygen and airway support, control seizures with a benzodiazepine, and give a bolus of 20 percent lipid emulsion followed by an infusion. Adrenaline is used in small doses of 1 microgram per kilogram or less, because large doses may worsen outcome. Lidocaine is itself a local anaesthetic and calcium is not part of the treatment.

• Why A is wrong. Lidocaine is a sodium channel blocker like bupivacaine and can worsen the toxicity.
• Why C is wrong. Standard doses of adrenaline can impair lipid rescue and are not advised, and small doses are used if needed.
• Why D is wrong. Calcium does not reverse sodium channel blockade by local anaesthetic.

Topic: Local anaesthetic systemic toxicity. Reference: Association of Anaesthetists guideline on the management of severe local anaesthetic toxicity; Miller's Anesthesia, local anaesthetics

Correct answer: D. Night-time neutral wrist splints and reassurance that symptoms often settle after delivery

Carpal tunnel syndrome is common in late pregnancy because of fluid retention, and symptoms usually resolve within weeks of delivery. Neutral wrist splints worn at night are the first-line treatment, with a local steroid injection reserved for troublesome symptoms. Surgery and nerve studies are for persistent symptoms, motor weakness or atypical features, which are absent here.

• Why A is wrong. Surgery is for severe or persistent disease with motor loss and is not needed here.
• Why B is wrong. Nerve studies are not needed when the diagnosis is clinical and there is no weakness.
• Why C is wrong. Systemic steroids add risk and are not first-line in pregnancy.

Topic: Common presentation: carpal tunnel syndrome in pregnancy. Reference: NICE Clinical Knowledge Summaries: Carpal tunnel syndrome

Correct answer: D. Withhold PDE5 inhibitors while he uses nitrates and review alternatives

Phosphodiesterase type 5 inhibitors potentiate the vasodilator effect of nitrates and can cause profound, potentially fatal hypotension, so concurrent use is contraindicated for all agents in the class, including tadalafil, whose effect lasts up to 48 hours. A man with angina who may need glyceryl trinitrate urgently cannot reliably avoid it. Alternatives include vacuum devices, intracavernosal alprostadil and review of his cardiac status with his cardiologist.

• Why A is wrong. A lower dose does not remove the risk of severe hypotension.
• Why B is wrong. Delayed nitrate use is unsafe because he may need it for an acute attack.
• Why C is wrong. Tadalafil is also contraindicated with nitrates and has a longer duration of action.

Topic: Erectile dysfunction in a man who uses nitrates. Reference: EAU Guidelines on Sexual and Reproductive Health

Correct answer: A. Explain this is a genuine functional disorder and arrange physiotherapy

Functional neurological disorder is a genuine condition in which symptoms arise from altered nervous system function and not from structural damage. It is diagnosed on positive clinical signs, such as the Hoover sign, as well as a normal investigation result. Management begins with a clear, non-judgemental explanation, followed by physiotherapy-led rehabilitation, with psychological therapy for associated distress and treatment of comorbid depression or anxiety. Dismissal worsens outcomes, repeated investigation reinforces illness beliefs, and antipsychotics have no role.

• Why B is wrong. Telling her the symptoms are imagined is inaccurate and damages trust and outcomes.
• Why C is wrong. Further investigations without new findings reinforce illness beliefs and add risk.
• Why D is wrong. Antipsychotics are not effective for functional motor symptoms and carry adverse effects.

Topic: Functional neurological disorder: communicating the diagnosis. Reference: Stone J et al. Functional neurological disorder: defying dualism. World Psychiatry 2024; NICE QS198 Suspected neurological conditions: functional neurological disorders in adults

Correct answer: A. Intramuscular lorazepam, with monitoring of respiratory rate and oxygen saturation

When oral treatment is refused and rapid tranquillisation is needed, NICE NG10 recommends intramuscular lorazepam alone, or intramuscular haloperidol combined with promethazine. It advises lorazepam when the person has not taken antipsychotic medication before, and advises avoiding haloperidol with promethazine and using lorazepam when no ECG has been carried out. The first prescription is a single dose that is reviewed before any repeat, and physical monitoring follows. Intramuscular diazepam and chlorpromazine are not among the NG10 options: diazepam is painful and erratically absorbed, and chlorpromazine causes hypotension and lowers the seizure threshold.

• Why B is wrong. Intramuscular chlorpromazine is not an NG10 option, and it causes hypotension and lowers the seizure threshold.
• Why C is wrong. Intramuscular diazepam is not an NG10 option, it is painful and erratically absorbed, and NG10 advises a single dose and review rather than repeats every 15 minutes.
• Why D is wrong. NG10 advises avoiding intramuscular haloperidol when no ECG has been done, and an antipsychotic-naive man is at high risk of acute dystonia.

Topic: Psychiatric emergencies: rapid tranquillisation in an antipsychotic-naive patient. Reference: NICE NG10 Violence and aggression: short-term management in mental health, health and community settings

Correct answer: B. Healthy worker effect because fit people are selected into employment

Comparing workers with the general population often shows lower mortality because people who are ill or disabled are less likely to be employed, and those who become ill leave work. This selection bias is called the healthy worker effect, and it can hide a true hazard. Using an internal comparison group of less exposed workers reduces it. The general population includes people too ill to work, so the SMR is biased downward.

• Why A is wrong. Recall bias concerns self-reported exposure in case-control designs, and this cohort did not rely on recall.
• Why C is wrong. Lead-time bias concerns screening and survival time, and this study has no screening element.
• Why D is wrong. Older workers would have higher, not lower, mortality, and the SMR is already age standardised.

Topic: Bias: healthy worker effect in an occupational cohort. Reference: Gordis Epidemiology: Selection bias and the healthy worker effect

Correct answer: D. Treatment lowers the 5-year stroke risk from 4 in 100 to 2 in 100, a 2 percentage point reduction

The relative risk reduction is (4 - 2) / 4 = 50 percent, but the absolute risk reduction is 4 - 2 = 2 percentage points, and the number needed to treat is 1 / 0.02 = 50. Presenting the absolute change, or natural frequencies such as 4 in 100 to 2 in 100, helps patients understand their own benefit, because relative reductions can make small absolute effects sound large. Not everyone who takes the drug benefits.

• Why A is wrong. The risk has halved in relative terms, but the absolute reduction is 2 percentage points.
• Why B is wrong. The number needed to treat is 50, which is a different measure from the relative risk reduction of 50 percent.
• Why C is wrong. Most treated patients would not have had a stroke in any case, so half of all patients do not benefit.

Topic: Relative risk reduction and absolute risk reduction: communicating results. Reference: Gigerenzer G et al. Helping doctors and patients make sense of health statistics. Psychol Sci Public Interest 2007