Stop hunting for zebras in Texas: end the diagnostic culture of “rule-out”
BMJ 2014; 348 doi: https://doi.org/10.1136/bmj.g2625 (Published 07 April 2014) Cite this as: BMJ 2014;348:g2625
All rapid responses
Astutely deploying humour, humility and compassionate insight (into self and colleagues) Jha (1) makes the case for physicians to investigate patients in a rational, justifiable way based upon their clinical presentation. However, I am concerned by two instances quoted, whence the generalist physician may gain the impression that MRI is the first line imaging modality to “exclude the possibility of an underlying adrenal mass” and to investigate a “questionable adrenal cancer”.
For the purpose of risk stratification in the context of adrenal incidentaloma, adrenal-protocol CT imaging has a significantly greater evidence base than MRI, not to mention being significantly cheaper, with shorter scan times and is available “24-7” in all UK acute hospitals. It can readily be “bolted-on” in real time to a CT originally ordered for other indication without overly delaying/disrupting the scanning list, or having to bring the patient back for a second scan.
When investigating adrenal lesions, the critical issues to resolve are (a) “Is an adrenal abnormality really present?”; if so, (b) “Is it a small, benign-looking lesion (eg. adrenal adenoma or myelolipoma) that requires no further imaging, or something rather more worrying (eg. adrenocortical cancer, phaeochromocytoma, or metastasis) that requires resection, or interval imaging?” and (c) “Might it be autonomously secreting hormones?” (2).
On size criteria alone, any lesion >5cm raises concerns. Furthermore, unenhanced CT allows Hounsfield units (HU) to be estimated; a low density lesion (<10HU) is consistent with a lipid-rich adenoma, while higher-density lesions require further characterisation. Contrast-enhancement, with estimation of washout, provides additional risk stratification in intermediate cases (3,4). While specialised MRI sequences can also provide some of this information, it is only rarely a first line investigation.
References
1. Jha S. Stop hunting for zebras in Texas: end the diagnostic culture of “rule-out”. BMJ 2014;348:g2625
2. Young WF. The incidentally discovered adrenal mass. The New England Journal of Medicine. 2007;356(6):601–610.
3. Johnson PT, Horton KM, Fishman EK. Adrenal mass imaging with multidetector Ct: pathologic conditions, pearls, and pitfall. Radiographics. 2009;29(5):1333–1351.
4. Tenenbaum F, Lataud M, Groussin L. Update in adrenal imaging. Presse Med. 2014; 43:410-419.
Competing interests: No competing interests
The diagnostic culture of "rule out" is beautifully encapsulated in the 10-minute consultation(no irony intended) entitled "diagnosis and management of chronic heart failure"(BMJ 2014:348:33-35).
In that vignette the evaluation of the 80 year old woman who presents with a 2 months history of the association of pedal oedema and breathlessness going upstairs is supposed to include a differential diagnosis which includes pulmonary emboli, lung malignancy, chest infection, chronic obstructive pulmonary disease and sleeping in a chair all night. Surely how likely are any of those scenarios plausible in a patient in whom pedal oedema co-exists with breathlessness going upstairs? Even in chronic obstructive airways disease, effort dyspnoea would have antedated pedal oedema by several months, and the patient who has dependent oedema due to sleeping in a chair all night would be unlikely to have co-existing breathlessness going upstairs unless the latter had antedated the preference for sleeping in a chair all night. Surely, if a plausible provisional diagnosis has to be formulated within 10 minutes, the most probable underlying cause for the co-existence of pedal oedema and effort dyspnoea has to be heart failure and that provisional diagnosis has to be validated within the 10 minute time frame of that consultation. Validation should include,at the very least, evaluation jugular venous pressure as a parameter of volume overload. That evaluation takes at least 5 minutes because the patient has to be positioned properly, in a good light, with the neck muscles relaxed, and the doctor eliciting all the clinical signs for raised jugular venous pressure he ought to have been taught at medical school, if the medical school was not one which used the formula "JVP raised 2 cm"
Competing interests: No competing interests
Thank you Dr Jha for highlighting this issue in the current "market" of evidence-based rather than clinical-based medicine. For me, the most important question to ask is "What made you come to the hospital/clinic?" This will lead on to the various leading questions and your focused examination of the patient. It is important to understand that the chances of picking up multiple diagnoses in a patient presenting with a single problem is much less than finding the most common one associated with the clinical presentation.
We must realize the consequences of finding something we never intended to look for and its impact not only the system but also on the individual going through the particular test. The trauma of living in apprehension will continue beyond the recommended period, if one test leads to the other until somebody decides to put a stop. Yes, rare things can be common but common things are never rare. So let us all try and look for them!
Competing interests: No competing interests
Multimorbidity offers a zoo-ful of possibilities!
In his reflection, Jha offers a thoughtful summary of some of the problems with our current marketized, payment by activity, medical culture.
His piece hangs critically on the observation which Bayes initially identified and formalised, and now drives testing theory: the prior probability is a key determinant of a test's diagnostic accuracy (complementing the test's sensitivity and specificity). However good, in practice neither of the latter two factors completely makes up for it. The principle was later was much more pithily encapsulated in Willie Sutton's observation: 'I rob banks because that's where the money is.' [1]
Mukherjee has beautifully compressed the same concept into his first 'law of medicine:' 'A strong intuition is much more powerful than a weak test.'[2]
That said, Jha (and for that matter Dr Awad, too, in his rapid response) fails to make explicit the increasingly important point that in older people, especially if frail, there is a greater likelihood of underlying (often covert) multimorbidity. [3] Not only does this increase the chance of incidental findings, but they can also each mask or enhance the finding of other 'abnormalities.'
Gigerenzer has written powerfully of the importance of trusting gut feelings, rather than relying on algorithms, in making decisions in complex situations. [4]
A quote often attributed to Osler is “Listen to the patient, he is telling you the diagnosis.” We forget this distillation of wisdom at our peril.
1. http://quoteinvestigator.com/2013/02/10/where-money-is/ (accessed 25 Oct 2015)
2. The Laws of Medicine, TED Books Simon & Schuster, London 2015 1st ed. p 21
3. Mangin D, Heath I, Jamoulle M. Beyond diagnosis: rising to the multimorbidity challenge. BMJ : British Medical Journal. 2012 Jun;344. Available from: http://dx.doi.org/10.1136/bmj.e3526.
4. Risk Savvy: How To Make Good Decisions, Gigerenzer G. Allen Lane 1st ed 2014
Competing interests: I am a geriatrician