Every doctor is familiar with the necessary debate in modern medicine over our heavy reliance on scanners. Imaging has become the diagnostic workhorse across most specialties, leading purists to rightly point out the erosion of traditional clinical skills amongst new generations of doctors.
They often romanticise an era when a patient’s history and physical examination held immense diagnostic power, and it is easy to understand the appeal of this nostalgia: It recalls a time when physicians were viewed as almost superhuman by patients and society, capable of waltzing into a cubicle and uncovering complex pathology with nothing more than a hand placed on an abdomen or a listen to the chest. Sending a patient to the radiology department for the answer doesn’t quite have the same gravitas for the doctor.
Others may cry out at the risks of ionising radiation, the strain on the healthcare system, and the psychologic burden of the incidental finding. But for all attempts to reduce our proclivity to lean on imaging, demand continues to climb. The Royal College of Radiologists’ 2024 workforce census found that CT and MRI examinations in England rose by 8% in 2023-24, while the UK’s consultant radiology workforce grew by 4.7%.
So is overreliance on the scanner really the problem?
Article Key Points
Imaging can prevent unnecessary surgery when clinical probability is high but not definitive.
Acute abdomen decisions often hinge on small residual uncertainty; 5% error may still mean avoidable operations.
Negative appendicectomy risk = 1 in 20 when acting on score-based probability alone.
Low-dose CT abdomen can rapidly rule out appendicitis, potentially reducing surgical morbidity.
Hx/physical exam remain essential for imaging selection; imaging should refine, not replace, clinical assessment.
Dive Deeper
Which appendicitis scores best reduce negative appendicectomy?
How accurate is low-dose CT for appendicitis exclusion?
What are CT risks in younger acute abdomen patients?
When we begin to stack the above factors against the monumental risks associated with going under the scalpel (such as intraoperative risks, hospital-acquired infections, anaesthetic complications, and severe psychological distress), the case for withholding a scan begins to shrink dramatically. In this piece, I make the case that where surgery is the alternative, the greater risk lies in scanning too little.
The Prospect of Surgery
The prospect of surgery is perhaps one of the most anxiety-inducing moments a patient can face.
From the anaesthetist’s cocktail of paralytics and hypnotics awaiting them in the induction room to the trauma of the surgery itself and the gruelling uphill battle of postoperative recovery, it is an immense psychological and physical toll to ask from any human being. As clinicians, it is our duty to make every conceivable effort to minimise the necessity of surgical intervention.
Historically, however, clinical medicine was forced to operate within a necessary margin of error.
Before high-fidelity imaging became so readily accessible, we simply could not afford the luxury of absolute certainty. We took histories to the best of our ability, palpated abdomens, scrutinised blood results, and made educated leaps of faith. But clinical intuition, even at its best, is never an exact science — it deals with probabilities rather than certainties.
This uncertainty steered countless patients to undergo surgeries they didn’t actually need. Laparoscopies that showed a healthy appendix, or exploratory laparotomies that opened the abdomen only to close it again, were seen as acceptable trade-offs to prevent fatal complications like sepsis and perforation.
The Reign of Radiology
Cross-sectional imaging has completely rewritten this contract. Think about this for a moment: We now possess the ability to readily visualise almost every recess and corner of the human anatomy noninvasively, in three dimensions, and at a good resolution. Oh, and all while the patient lies comfortably awake on the table with no need for analgesia or sedation.
When the case is laid out this way, is it any wonder that the on-call radiology registrar’s phone is never silent?
But with those surgical risks that we covered in mind, it might be time to flip the original question on its head: What if the problem isn’t that we rely on the scanner too much, but that we use it too little?
The Cost of a 5% Margin of Error
Consider the classic surgical presentation of an acute abdomen. Imagine we use a highly validated clinical scoring system, such as the Alvarado score, for a patient presenting with right iliac fossa pain. The score in our hypothetical patient suggests a 95% likelihood of acute appendicitis. Most doctors would tell you that this is a strong mandate for a trip to the operating theatre.
But we need to touch on the inverse of that statistic: a 5% margin of error.
This means that 1 in 20 patients presenting with these exact clinical variables as our patient will be subjected to an entirely unnecessary surgery. We must ask ourselves a difficult, almost paradigm-shifting question: Can a 5% negative appendicectomy rate ever truly be acceptable in an era where a CT scan can definitively rule it out within minutes?
When we hold the power to eliminate that 1-in-20 risk for surgical morbidity with something as simple as a scan, opting for surgery based on probability alone begins to feel more like an antiquated gamble. So, it seems that the balance of risk has fundamentally shifted in modern medicine. Although we should try to minimise radiation exposure, particularly in more radiosensitive younger patients, the potential harm of an unnecessary operation may outweigh the small long-term risk associated with a clinically justified, modern low-dose CT abdomen.
Needless to say, this is not a call to abandon the clinical examination, nor is it an endorsement of defensive, thoughtless medicine. The history and physical exam remain the crucial foundation that guides the need for imaging and tells the reporting radiologist where to look. But when it comes to the final decision to open the human body, imaging has fundamentally changed how we approach risk for our patients.
Arya Anthony Kamyab is a speciality trainee doctor in clinical radiology working in Oxford University Hospitals. You can follow him onSubstack and on Instagram@aryak.writes.
Every doctor is familiar with the necessary debate in modern medicine over our heavy reliance on scanners. Imaging has become the diagnostic workhorse across most specialties, leading purists to rightly point out the erosion of traditional clinical skills amongst new generations of doctors.
They often romanticise an era when a patient’s history and physical examination held immense diagnostic power, and it is easy to understand the appeal of this nostalgia: It recalls a time when physicians were viewed as almost superhuman by patients and society, capable of waltzing into a cubicle and uncovering complex pathology with nothing more than a hand placed on an abdomen or a listen to the chest. Sending a patient to the radiology department for the answer doesn’t quite have the same gravitas for the doctor.
Others may cry out at the risks of ionising radiation, the strain on the healthcare system, and the psychologic burden of the incidental finding. But for all attempts to reduce our proclivity to lean on imaging, demand continues to climb. The Royal College of Radiologists’ 2024 workforce census found that CT and MRI examinations in England rose by 8% in 2023-24, while the UK’s consultant radiology workforce grew by 4.7%.
So is overreliance on the scanner really the problem?
When we begin to stack the above factors against the monumental risks associated with going under the scalpel (such as intraoperative risks, hospital-acquired infections, anaesthetic complications, and severe psychological distress), the case for withholding a scan begins to shrink dramatically. In this piece, I make the case that where surgery is the alternative, the greater risk lies in scanning too little.
The Prospect of Surgery
The prospect of surgery is perhaps one of the most anxiety-inducing moments a patient can face.
From the anaesthetist’s cocktail of paralytics and hypnotics awaiting them in the induction room to the trauma of the surgery itself and the gruelling uphill battle of postoperative recovery, it is an immense psychological and physical toll to ask from any human being. As clinicians, it is our duty to make every conceivable effort to minimise the necessity of surgical intervention.
Historically, however, clinical medicine was forced to operate within a necessary margin of error.
Before high-fidelity imaging became so readily accessible, we simply could not afford the luxury of absolute certainty. We took histories to the best of our ability, palpated abdomens, scrutinised blood results, and made educated leaps of faith. But clinical intuition, even at its best, is never an exact science — it deals with probabilities rather than certainties.
This uncertainty steered countless patients to undergo surgeries they didn’t actually need. Laparoscopies that showed a healthy appendix, or exploratory laparotomies that opened the abdomen only to close it again, were seen as acceptable trade-offs to prevent fatal complications like sepsis and perforation.
The Reign of Radiology
Cross-sectional imaging has completely rewritten this contract. Think about this for a moment: We now possess the ability to readily visualise almost every recess and corner of the human anatomy noninvasively, in three dimensions, and at a good resolution. Oh, and all while the patient lies comfortably awake on the table with no need for analgesia or sedation.
When the case is laid out this way, is it any wonder that the on-call radiology registrar’s phone is never silent?
But with those surgical risks that we covered in mind, it might be time to flip the original question on its head: What if the problem isn’t that we rely on the scanner too much, but that we use it too little?
The Cost of a 5% Margin of Error
Consider the classic surgical presentation of an acute abdomen. Imagine we use a highly validated clinical scoring system, such as the Alvarado score, for a patient presenting with right iliac fossa pain. The score in our hypothetical patient suggests a 95% likelihood of acute appendicitis. Most doctors would tell you that this is a strong mandate for a trip to the operating theatre.
But we need to touch on the inverse of that statistic: a 5% margin of error.
This means that 1 in 20 patients presenting with these exact clinical variables as our patient will be subjected to an entirely unnecessary surgery. We must ask ourselves a difficult, almost paradigm-shifting question: Can a 5% negative appendicectomy rate ever truly be acceptable in an era where a CT scan can definitively rule it out within minutes?
When we hold the power to eliminate that 1-in-20 risk for surgical morbidity with something as simple as a scan, opting for surgery based on probability alone begins to feel more like an antiquated gamble. So, it seems that the balance of risk has fundamentally shifted in modern medicine. Although we should try to minimise radiation exposure, particularly in more radiosensitive younger patients, the potential harm of an unnecessary operation may outweigh the small long-term risk associated with a clinically justified, modern low-dose CT abdomen.
Needless to say, this is not a call to abandon the clinical examination, nor is it an endorsement of defensive, thoughtless medicine. The history and physical exam remain the crucial foundation that guides the need for imaging and tells the reporting radiologist where to look. But when it comes to the final decision to open the human body, imaging has fundamentally changed how we approach risk for our patients.
Arya Anthony Kamyab is a speciality trainee doctor in clinical radiology working in Oxford University Hospitals. You can follow him on Substack and on Instagram @aryak.writes.