An example of how this might go wrong would be a patient admitted with back pain and difficulty passing urine. Without a proper history and thorough examination this patient could easily be labelled as having pyelonephritis with kidney pain. This would lead to the commencement of antibiotics to treat of severe UTI and possibly a referral to urology for inpatient care. However the patient doesn’t respond to these initial antibiotics within 48 hours so the team calls a microbiologist saying “their patient with pyelonephritis is not responding, what should they change the antibiotic to?” Unless the microbiologist questions the diagnosis, another 48 hours might elapse before anyone questions what is happening. Does this sound familiar?
The differential diagnosis for severe back pain may also include potentially life-threatening conditions such as a dissecting aneurysm and pancreatitis. Failing to consider these when admitting a patient could be disastrous. And what if the patient had severe back pain due to an epidural abscess? The pain may well be lumbar (i.e. near the renal angle) and if there was compression of the spinal cord they may well have difficulty passing urine due to compromise of the nerve supply to the bladder sphincter muscle.
The fact that a differential diagnosis includes other potential sources ensures doctors are not blinkered into following one course. Like a game of Cluedo differential diagnosis is a process of elimination, or reducing the probability, of life-threatening, common and uncommon conditions systematically. As differential diagnoses are ruled out a quick adjustment shows the most likely cause, in this case a spinal epidural abscess and allows correct urgent management to be implemented. Spinal epidural abscess is a surgical emergency and when present for more than 24
hours, the neurological damage may be irreversible. In this case, if the team had waited till the first diagnosis, pyelonephritis, was no longer probable, it
would have lead to long term irreversible damage in the patient.
My personal method for formulating a differential is to use the following list in order, and then to fit a system based approach afterwards:
- Life-threatening conditions
- Common conditions
- Uncommon conditions
as the story changes or as further results come back from investigating, the differential diagnosis can be modified until the final confirmed diagnosis is reached.
By using a differential diagnosis, patients are managed safely and more effectively and doctors become very good at diagnosing, which is what all good doctors do best.