Intensive care medicine resists the way most exams are revised, because its patients rarely have one thing wrong with them and its questions are rarely about a diagnosis. A patient arrives with several failing organ systems, a set of numbers, and a trajectory, and the question is what you support, what you treat, what you monitor, and when the answer becomes that you should stop. Candidates who revise ICU as a catalogue of conditions find the exam slippery. Candidates who revise it as organ failure and its support find that most questions resolve to the same small set of decisions.
Key takeaways
- Organise your revision by failing organ system rather than by disease, because that is how the patients present.
- Separate three questions in every case: what supports the organ, what treats the cause, and what monitors the response.
- Intensive care is a loop rather than an intervention, so reassessment is examinable and candidates forget it.
- Ethics, ceilings of treatment and rehabilitation are core content, not soft additions.
- Practise mixed, multi-organ cases, because single-system questions do not reproduce the specialty.
Start with the failing organ
For any critically ill patient, before you consider the diagnosis, ask which organ systems are failing and how badly.
Respiratory failure, and is it hypoxaemic, hypercapnic, or both, because that determines the support. Cardiovascular failure, and is the problem preload, pump, afterload, or rhythm, because the vasoactive strategy follows from it. Renal failure, and is it pre-renal, intrinsic or obstructed, and does it need replacement. Neurological failure, and is the problem primary or secondary, and is the intracranial pressure the issue. Hepatic, haematological, metabolic.
This is the organising principle of the specialty. A patient with severe pneumonia and a patient with pancreatitis may need very similar support, because the organ failures overlap even though the diagnoses do not, and the exam frequently tests exactly that recognition.
Then separate support from treatment
Every intensive care question contains at least two threads, and conflating them is the commonest error.
Support buys time. Ventilation, vasopressors, inotropes, renal replacement, fluid, sedation. It does not fix the underlying problem and it is not intended to. Its purpose is to keep the patient alive and their organs perfused while the cause is addressed or resolves.
Treatment addresses the cause. Antimicrobials with source control. Surgery. The antidote. Immunosuppression. The specific therapy.
The question is usually asking about one of these, and the answer list will contain plausible options from both. A patient with septic shock needs fluid and vasopressors and needs antibiotics and source control, and which of those is the next step depends on what the question asked and on how sick they are right now.
If you cannot say whether a given option is support or treatment, you have not yet understood the question.
Then the loop
Here is the feature of intensive care that written exams test more than candidates expect and that clinicians frequently under-revise.
Intensive care is not a series of interventions. It is a loop: intervene, monitor, reassess, adjust. The correct answer to a great many questions is not the next drug but the next measurement, or the recognition that the response to the last intervention tells you that your model of the patient was wrong.
So for every intervention you learn, learn its monitoring. What tells you the ventilation strategy is working, and what tells you it is causing harm. What tells you the vasopressor is adequate, and what tells you the patient is under-filled rather than under-supported. What the trajectory of a lactate or a urine output actually means.
Questions that ask what to do next, in a patient who has already been treated, are asking whether you can close the loop.
The domains candidates neglect
Three areas are reliably examined and reliably under-revised, and they share a characteristic: they do not feel like intensive care medicine to trainees who have been focused on the physiology.
Ethics and ceilings of treatment. When to escalate, when not to, who decides, and how. The recognition that continuing organ support in a patient who cannot benefit from it is a clinical error rather than a kindness. Best interests decisions, capacity, advance decisions, and the position on withdrawing versus withholding.
End-of-life care in the unit. The practicalities of withdrawal, symptom control during it, and communication with families. This is examinable and it is not a matter of opinion.
Rehabilitation and recovery. What happens after the acute phase: the sequelae of critical illness, nutrition, delirium prevention, early mobilisation, and the long-term consequences that follow patients out of the unit. Candidates who revise only the acute phase are missing an examined domain.
Practise multi-organ cases
A structural point about how to work.
Single-system questions, where a patient has one problem and you decide what to do about it, are useful early and do not resemble the specialty. Real intensive care patients have a failing lung, a failing kidney, a coagulopathy and a new fever, and the interactions between those are where the difficulty lives: the ventilation strategy that compromises the circulation, the fluid that helps the kidney and harms the lung, the drug that cannot be given because of the liver.
So as your preparation matures, work in mixed, multi-organ cases, and practise holding several failing systems in mind at once while deciding which one is currently the most dangerous. That prioritisation, across systems, is what the specialty is, and it is a skill you can only build by practising it.
Where iatroX fits
iatroX's FFICM bank is built around organ failure and its management rather than around isolated disease recall, with explanations grounded in current guidance so the reasoning behind a support strategy or a monitoring decision sits with the question. Missed questions can be opened in the Socratic Tutor, which asks you to reason before it explains and names whether your error was in the support, the treatment, or the reassessment, which are three different failures with three different remedies. The adaptive engine returns the same principle across a different organ system, which tests whether the reasoning transferred. Try it with free sample questions at iatroX. For the acute prioritisation reasoning this shares with acute medicine, see prioritisation and deterioration.
Frequently asked questions
How should I organise FFICM revision? By failing organ system rather than by disease. Intensive care patients present with organ failure, and different diagnoses frequently require identical support, which is what the exam is testing when it presents two very different conditions with overlapping management.
What is the difference between support and treatment in an ICU question? Support buys time by maintaining organ function, such as ventilation, vasopressors or renal replacement. Treatment addresses the cause, such as antimicrobials with source control or surgery. Answer lists usually contain plausible options from both, and the question is asking about one.
Why do I get questions wrong about what to do next? Frequently because you offer another intervention when the question wants a measurement or a reassessment. Intensive care is a loop, and questions about a patient who has already been treated are usually asking whether you can interpret the response and adjust.
Are ethics and end-of-life care really examined? Yes, and as knowledge rather than as opinion. Ceilings of treatment, capacity and best interests, the equivalence of withholding and withdrawing, and the practicalities of end-of-life care in the unit are all testable and all under-revised.