To the Editor: I was surprised by Wong’s letter on the role of history-taking and examination in the diagnostic process.1 I would suggest that Wong, as a surgical registrar, receives the majority of his abdominal pain referrals from the medical staff of the emergency department. Although he advocates the liberal use of abdominal computed tomography (CT) scanning, I believe he ignores the fact that another medical practitioner has already taken a history and performed an examination that has suggested a surgical cause of pain for which a surgical opinion is then requested. Wong would thus remain unaware of other cases in which patients present with abdominal pain but the case is ruled non-surgical on the basis of history, examination and limited investigation not involving abdominal CT scanning.

History, examination and even appropriately targeted investigations remain imperfect diagnostic tools, but I agree with Schattner2 that history-taking and examination are very important adjuncts in the diagnostic process.

To the Editor: Like Schattner, I am appalled by the attitude to diagnosis displayed by Wong regarding the use of computed tomography (CT) scanning in preference to initial history-taking and physical examination in abdominal pain.1 Unfortunately, this approach is becoming increasingly more prevalent among junior staff (and even among some senior staff).

Wong poses the question, “[W]hy do some clinicians continue to routinely promulgate the sacred and arcane ritual of taking a history and doing an examination, which, as diagnostic tools, are clearly second-rate?” There are several reasons why I continue to promulgate the classical paradigm.

Firstly, I would remind him of Bayes’ theorem: post-test probability equals pre-test probability multiplied by the likelihood ratio of the test. Put simply, this means that, for a test that is not 100% accurate (ie, effectively, all imaging tests), you cannot interpret the meaning of the result without having some idea of the pre-test probability of a diagnosis. And how can you satisfactorily arrive at a pre-test probability without having clinically assessed the patient? In addition, the radiologist is able to interpret the images more accurately when there are clinical details provided.2

Secondly, is Wong seriously suggesting that all patients with abdominal pain, including young adults and children, undergo CT scanning without any kind of clinical filtering or assessment? This is wrong and potentially negligent. The radiation dose received by the patient from an abdominal CT scan is a serious consideration. Assuming a total effective body dose of 10 mSv, there is an excess risk of a radiation-induced fatal cancer of about 1 in 2000.3 Apart from the risk to the individual, the number of iatrogenic cancers potentially induced in the community by indiscriminate use of CT would be a major concern.4

Thirdly, the implication of Wong’s letter is that clinical assessment and imaging are somehow in competition with each other, whereas nothing could be further from the truth. Of course, modern imaging has contributed to making diagnosis far more accurate than in the time of Hippocrates, but a complementary approach is far more rewarding for patients and doctors.

Lastly, in patients with abdominal pain, there are many occasions when no imaging is required and others when ultrasonography is more appropriate than CT, because it avoids ionising radiation in young patients and is more accurate for diagnosing gynaecological causes of pain.5

To the Editor: It is clear Dr Wong1 has a practice rather different from mine. He is not used to the truly undifferentiated patients that present in their thousands to emergency departments and general practices every day. There, the art of history and examination is truly alive.

No one questions the value of complex imaging. It has its place after a detailed history has been taken and focused examination and relevant investigations have been carried out, leading to a risk assessment and management plan. One does not order computed tomography (CT) scans willy-nilly. For example, the Canadian CT Head Rule2 for patients with minor head injury sets out which patients should have a head CT scan, based on a simple set of historical and examination findings. Moreover, CT scans are wasted on conditions for which CT imaging is inappropriate — it is rare that I order a CT scan for a child with abdominal pain.

When I ask surgical registrars for their opinion, I am actually asking for their consultant’s opinion. Nothing guides like an experienced hand, whether it be feeling a belly or writing a CT request form. On many occasions, I have concluded that all the imaging performed on a patient with abdominal pain did not contribute to the diagnosis and the patient simply needed a laparotomy. At my insistence, the consultant is called, appropriate treatment commences, and the patient boards the experience express on the track to recovery. As Shem quips, in his satirical book on medical training and hospital life — nothing heals like cold steel.3

CT is not the be-all and end-all of medicine. Hopefully, by the end of his training, Wong will have developed the hand of experience and be able to continue the art of medicine through the ages. In the words of William Osler:

To the Editor: Apropos the letter by Wong entitled “Clinical paradigms revisited” in the Christmas issue,1 declaring fossilisation of the very pillars on which medicine stands, we would like to express a contrary opinion.

To be adept physicians, clinicians must hone their skills at taking a lucid and informative history and conducting a thorough physical examination. It would be a crying shame if young doctors, having slaved for 5 or more years to obtain a medical degree, had to rely solely on expensive investigations when they have the God-given tools of the five senses. To confirm a clinical diagnosis and assess the extent of disease, doctors should order specific and appropriate investigations, rather than ordering tests that may be irrelevant and financially bleeding the patient. The issues of cost, radiation hazard, availability of trained personnel, and need for expensive equipment have been trivialised.

In a country like India, where the majority of the population cannot afford even minimal hospital fees, to even contemplate using a computed tomography scan as a first-line diagnostic tool for something as basic as abdominal pain is absurd.

Moreover, the use of advanced technology does not guarantee a correct diagnosis. A recent case of aortic dissection was misdiagnosed as acute coronary syndrome on the basis of electrocardiography.2 If due emphasis had been given to pulse and blood pressures in both limbs, this mistake could have been avoided. In another case, involving recurrent loss of consciousness, investigations were non-contributory, but a history of substance misuse at home pointed to the correct diagnosis.3 In another study, clinical judgement regarding the severity of pneumonia was found to be a more reliable predictor than a standardised scoring system based on clinical signs and laboratory findings.4

Doctors ought to be able to make a clinical judgement in the first instance, rather than resorting blindly to expensive investigatory tools. We do not deny the usefulness of modern technological devices for confirming or ruling out clinical possibilities, but they must be used judiciously. Such investigations cannot take precedence over physicians’ reliance on their clinical skills, lest we become helpless without technology.

To the Editor: Schattner1,2 and Wong3 raise issues that examine what has been the core of medical practice since antiquity. Grasping antiquity for its own sake is problematic, at best, and possibly heralds the extinction of long held practices, at worst. As technology improves, we are witnesses to improved imaging modalities that provide higher diagnostic yields, with improved sensitivity and specificity, at increasingly reduced costs. Refusal to even acknowledge the possibility that the history and examination may be terminal is not prudent. Instead, we need to examine carefully our mantra(s) with respect for the temporal nature of medicine.

History and examination evolved in their current form because previous generations could not see inside the body, or examine physiological and pathological processes in real time. Our predecessors amassed a series of verbal cues and physical rules that generally conformed to the presentation of a particular disease. The future of medicine heralds dramatic departure from the world view that preceded computed tomography and magnetic resonance imaging.

Wong raises an important issue with regard to diagnosing emergency abdominal conditions in busy hospital practice. He does not discount a role for the history or physical examination. He does, however, challenge their pre-eminence in “conditions that require emergency surgical treatment”. Is it really in the best interests of patients and the health care system for the emergency department intern/resident, then the registrar/consultant, then the surgical fellow, to all take the history and perform a physical examination? In essence, doesn’t Wong’s “scan first approach” reflect a prudent reliance on, and respect for, the information already gathered?

Schattner4 states that “all imaging studies combined (computed tomography, magnetic resonance imaging, ultrasound, and echocardiography) were decisive in only 10.5% of cases” whereas “the patient’s history and the evolution of the condition proved to be the decisive diagnostic method in 23% of cases”. Doesn’t this show that Wong’s approach provides a heuristic that increases the diagnostic yield, reduces delays and guesswork, and streamlines the processing of patients presenting with acute abdominal pain — or is it acceptable to miss the significant percentage of diagnoses that are decided by imaging alone?!