Brady (2020): radiological diagnosis, biopsy and the Bolam standard

Brady v Southend University Hospital NHS Foundation Trust

[2020] EWHC 158 (QB)
High Court, Queen’s Bench Division
Andrew Lewis QC, sitting as a Deputy High Court Judge
Judgment delivered: 31 January 2020

Background

Following a laparoscopic appendicectomy in May 2013, during which acute appendicitis and turbid pelvic fluid were identified, the claimant developed abdominal pain and raised inflammatory markers.

An ultrasound examination demonstrated an ill-defined epigastric mass. A CT performed on 5 August 2013 showed a right upper-quadrant mass containing fat, increased density and localised vascularity. The reporting consultant radiologist considered the appearances most consistent with omental infarction rather than neoplasm. The claimant’s symptoms improved and she was managed conservatively.

Persistent pain and a palpable abdominal mass led to further admission in September 2013. A second CT showed that the mass had enlarged and was associated with gastric outlet obstruction, encasement of the transverse colon and diffuse abdominal nodularity. The reporting radiologist was uncertain of the diagnosis, recommended urgent further evaluation and discussed possible diagnoses and biopsy with the treating surgeon.

Following specialist review of the imaging at another teaching hospital, the working diagnosis remained omental infarction. As the claimant improved with antibiotics, biopsy was deferred and conservative management continued.

In February 2014, the claimant presented with a left psoas abscess. Surgical drainage and microbiological analysis confirmed Actinomyces. She required further procedures and was left with extensive abdominal scarring and continuing symptoms.

The claim

The claimant alleged that:

the abdominal mass seen on the 2013 CT examinations represented actinomycosis rather than omental infarction.
the first CT had been negligently interpreted.
the second CT report inadequately communicated that omental infarction had become less likely.
infection and malignancy should have been stated as differential diagnoses.
an urgent biopsy should have been performed in August or September 2013.
an earlier biopsy would have diagnosed actinomycosis and prevented the subsequent psoas abscess and surgery.

The defendant maintained that the radiological appearances reasonably supported omental infarction and that conservative management, including obtaining a specialist second opinion before undertaking an invasive biopsy, fell within responsible clinical practice.

The legal approach

The court distinguished between two questions:

What did the imaging actually show?
This was a question of fact for the court, determined on the balance of probabilities with the assistance of expert evidence.
Was the interpretation or management negligent?
This required assessment of whether the clinician acted in accordance with a responsible body of professional opinion under Bolam, subject to the Bolitho requirement that the supporting opinion be logical and defensible.

Applying Penny v East Kent Health Authority, the court held that Bolam remains applicable to the exercise of professional skill and judgment in diagnostic interpretation. The existence of conflicting expert opinions does not itself resolve the issue. The court must examine what was visible, what a reasonably careful practitioner should have identified and whether the resulting interpretation was professionally defensible.

For recommendations concerning differential diagnoses, biopsy and further management, the conventional Bolam and Bolitho principles applied directly because those decisions involved clinical judgment and the balancing of investigative risks and benefits.

What was the abdominal mass?

With the benefit of hindsight, the court found on the balance of probabilities that the 2013 mass was probably an infective process rather than an omental infarction.

Relevant factors included:

the earlier appendicitis and turbid pelvic fluid provided a plausible source of abdominal actinomycosis.
the mass enlarged between the August and September CT examinations.
diffuse peritoneal nodularity was compatible with infection.
the mass resolved following antibiotic treatment.
the timing was consistent with abdominal actinomycosis.
it was less likely that the claimant had developed two unrelated rare conditions, namely spontaneous omental infarction and actinomycosis.

That factual finding did not determine breach of duty. The issue remained whether the clinicians’ contemporaneous interpretation and management were reasonably competent on the information then available.

The first CT examination

The court found that the first CT report was not negligent.

The mass was situated in the greater omentum and contained fat and increased density. Its location and appearance were compatible with omental infarction. Localised hypervascularity, ill-defined margins and the absence of visible pus did not exclude that diagnosis.

The report described omental infarction as the “most likely” diagnosis rather than presenting it as certain. This appropriately conveyed the reporting radiologist’s level of confidence.

The court also held that:

malignancy was sufficiently remote that it did not have to be expressly included in the conclusion.
infection was exceptionally unusual in that location and did not have the appearance of an abscess.
radiologists are not required to list every theoretically possible diagnosis.
indiscriminate differential lists may reduce rather than improve the clinical value of a report.
the claimant’s clinical improvement supported conservative management.

Even if infection or malignancy ought to have been mentioned, the court considered that this would probably not have resulted in immediate biopsy. A watch-and-wait approach would still have been reasonable given the imaging appearances, clinical improvement and risks of invasive investigation.

The second CT examination

By September 2013, the clinical and imaging position had changed. The mass had enlarged and was associated with obstruction, colonic encasement and abdominal nodularity. These findings required reconsideration of the original working diagnosis.

The second report was described as suboptimal because it did not set out the relevant differential diagnoses. Nevertheless, it was not negligent because:

it accurately described the important imaging abnormalities.
it identified the need for urgent further evaluation.
the radiologist communicated her diagnostic uncertainty to the treating surgeon.
infection, infarction and malignancy were discussed.
biopsy was recommended as a means of obtaining a tissue diagnosis.
the combined written and verbal communication was clinically safe and did not mislead the treating team.

The case demonstrates that the adequacy of radiological communication may be assessed by considering both the written report and any documented direct discussion with the responsible clinician.

The decision not to biopsy

The court accepted that many surgeons would have performed a biopsy after the second CT. The legal question, however, was not whether biopsy was a reasonable option, but whether it was mandatory for any reasonably competent surgeon.

The decision to defer biopsy was held to be defensible because:

omental infarction remained a possible diagnosis.
biopsy carried recognised risks, including bleeding, infection and complications from anaesthesia.
the surgeon sought specialist upper gastrointestinal and radiological review.
the external review supported the diagnosis of omental infarction.
conservative management was appropriate if that diagnosis was correct.
the claimant’s symptoms improved during admission.

Obtaining and acting on a specialist second opinion was itself a reasonable response to diagnostic uncertainty. The existence of another responsible course, including biopsy or laparoscopy, did not make conservative management negligent.

Causation and outcome

The court found that an earlier biopsy would probably have identified the infective process. Appropriate prolonged antibiotic treatment would then probably have prevented the psoas abscess and subsequent surgery.

Causation was therefore established on the assumed basis that there had been a breach. However, the claim failed because none of the alleged breaches of duty was proved.

The claim was dismissed. Damages had been agreed at £56,000, subject to liability.

Significance for radiological practice

Hindsight must be separated from negligence. A later diagnosis does not retrospectively render an earlier, reasonably supported interpretation negligent.
Diagnostic confidence should be explicit. Terms such as “most likely” may appropriately communicate uncertainty when supported by the imaging.
Differential diagnoses should be clinically useful. A radiologist need not include every remote possibility, but should identify material alternatives where they may affect management.
Changing appearances require reassessment. Enlargement, obstruction, nodularity or failure of an expected process to resolve should prompt reconsideration of the working diagnosis.
Recommendations must follow the level of uncertainty. Where imaging cannot provide a sufficiently confident diagnosis, appropriate further investigation should be advised.
Direct communication should be documented. A contemporaneous discussion may be relevant to whether uncertainty, urgency and recommendations were adequately conveyed.
A revealing test is not automatically a required test. The fact that biopsy would have produced the correct diagnosis does not establish that failure to perform it was negligent.
Expert evidence should separate observation from opinion. Experts should identify what was visible on the images, what a competent radiologist should have recognised and whether the actual interpretation remained defensible at the time.

Practical framework for reviewing an alleged radiological error

When assessing a diagnostic-negligence allegation, consider:

What findings were objectively present on the original examination?
What clinical information was available to the reporting radiologist?
Was the stated diagnosis reasonably supported by the imaging appearances?
Was the degree of certainty accurately communicated?
Were any omitted differential diagnoses sufficiently material to affect management?
Did the report recommend appropriate further investigation?
Were important findings or recommendations communicated directly and documented?
Would a different report probably have changed the clinical course?
Was the proposed alternative investigation one that competent clinicians were required to undertake at that time?
Is the criticism based on contemporaneous evidence or knowledge acquired only in hindsight?

Educational case commentary only. It is not legal advice and is not a substitute for the full judgment. The publication date above is the date of this commentary, judgment was delivered on 31 January 2020.