Managing Incidental Findings

Introduction

An incidental finding is an unexpected abnormality discovered on a test performed for an unrelated reason - a lung nodule seen on a CT done for trauma, an adrenal lesion on an abdominal scan for renal colic, a mildly raised calcium on a routine blood panel. As imaging resolution and the volume of testing have both increased, incidental findings have become a routine part of general practice work, and managing them well is now an explicitly examined skill.¹

The central tension is between under-reaction (missing a finding that represents early treatable disease) and over-reaction (subjecting a patient to anxiety, further invasive investigation, and sometimes treatment for an abnormality that would never have caused them harm). Both are real risks, and good management requires judgement, not a reflex response in either direction.

Overdiagnosis and overtreatment

Overdiagnosis is the identification of an abnormality that meets the technical definition of disease but would never have caused symptoms or shortened life if left undetected. It is distinct from a false positive (a test result that is simply wrong): an overdiagnosed finding is real, but clinically insignificant to that individual.

The rate of incidental findings rises with the sensitivity of the test: a CT chest detects small pulmonary nodules in a large minority of scans in older or ever-smoking populations, the great majority of which are, and remain, benign. The problem is not unique to imaging - screening blood panels, genetic testing, and even careful physical examination all generate findings whose clinical significance is genuinely uncertain, and overdiagnosis has been described across cancer screening, thyroid disease and several other domains as testing has become more widely available and more sensitive.³

A structured approach

There is no single algorithm that covers every possible incidental finding, because the right response depends on the specific tissue, the specific abnormality, and features unique to that finding type - what matters is applying a consistent reasoning process, then looking up or seeking specialist input on the specific risk-stratification criteria that apply.

1. Establish what the finding actually is - read the full report, not just the flagged abnormal line, since context (size, characteristics, comparison with prior imaging if available) usually determines risk far more than the finding's mere presence
2. Assess malignant/clinical potential using recognised risk features for that finding type (size, growth rate, morphology, patient risk factors)
3. Consider the patient's overall context: age, comorbidity, life expectancy, and whether they would be fit for and want further investigation or treatment if something serious were found
4. Decide: reassure and discharge, arrange interval surveillance, or refer for specialist assessment - and document the reasoning explicitly
5. Communicate clearly, calibrating language to actual risk rather than defaulting to alarming or falsely reassuring extremes
6. Clarify ownership of follow-up - who is responsible for actioning any surveillance imaging or repeat blood test, and by when

Worked examples

Pulmonary nodules

Solid pulmonary nodules found incidentally are risk-stratified by size, morphology (spiculated margins raise concern) and patient risk factors (smoking history, age, prior cancer). Guidelines (British Thoracic Society) define size thresholds below which no follow-up is needed, and above which surveillance CT or PET-CT/biopsy is indicated - most incidental nodules under 5 mm need no follow-up at all, while larger or growing nodules warrant a defined surveillance pathway.²

Adrenal incidentalomas

Found on a substantial minority of abdominal CT scans done for other reasons. Management requires two parallel questions: is it hormonally active (screen with overnight dexamethasone suppression test, plasma metanephrines, and if hypertensive, aldosterone:renin ratio), and does its imaging appearance suggest malignancy (size over 4 cm, irregular margins, high unenhanced CT attenuation raise concern). Most are non-functioning benign adenomas needing only imaging follow-up or, if small and reassuring, no follow-up at all.

Renal and hepatic cysts

Simple renal and hepatic cysts are extremely common, increase with age, and are almost always benign incidental findings requiring no further action if they meet simple imaging criteria (thin wall, no septations, no solid component, fluid attenuation). This is a classic exam point precisely because the correct answer is reassurance, not referral - candidates who reflexively refer every incidental finding are marked down here.

Thyroid nodules

Common, and the great majority are benign. Ultrasound characteristics (using a risk stratification system such as U-classification) guide whether fine needle aspiration is warranted; a nodule under a defined size threshold with reassuring sonographic features may simply be monitored.

Incidental blood test abnormalities

A mildly raised ALP, a borderline-low platelet count, or a mild leukocytosis found on a panel ordered for another reason should prompt a check of the clinical context and, usually, a repeat test before a cascade of further investigation - a single mildly abnormal value out of context is often a normal population variant or a transient effect (e.g. recent exercise, minor infection), not a new diagnosis.

Incidental findings on brain imaging

As MRI brain imaging is increasingly used for headache, dizziness and cognitive assessment, incidental findings are correspondingly common: small, non-specific white matter hyperintensities (common with age and usually representing chronic small vessel change rather than acute pathology), small arachnoid cysts, and pituitary microincidentalomas are all frequently seen and, in isolation without a matching clinical picture, rarely require action beyond noting them in the record. A pituitary incidentaloma warrants basic hormonal screening and size-based follow-up, since a minority are functioning or large enough to threaten the optic chiasm, but the great majority found incidentally are small, non-functioning and can simply be monitored.

Pancreatic cystic lesions

Increasingly picked up on abdominal CT or MRI done for unrelated reasons, pancreatic cystic lesions range from entirely benign (simple cysts, pseudocysts) to premalignant (intraductal papillary mucinous neoplasms, mucinous cystic neoplasms). Because the malignant potential of some subtypes is genuine, these findings are generally not appropriate for reassurance alone and warrant specialist pancreaticobiliary or radiology input to characterise the lesion and agree a surveillance or intervention plan - an example where the correct default response is different from the reassurance-first approach that fits simple renal or hepatic cysts.

Communicating incidental findings

How a finding is communicated has a measurable effect on anxiety independent of its actual risk. An ambiguous, poorly explained result ("there's a shadow on your scan, we're not sure what it is") generates disproportionate distress compared with the same finding explained with context and a clear plan.

• State clearly what was found, in plain language, avoiding alarming medical terms without explanation
• Give the actual risk in context - "this is a very common finding and in the vast majority of people it is completely harmless" - rather than leaving the patient to imagine the worst
• Explain the plan and the reasoning behind it, whether that is reassurance, a repeat test, or referral
• Elicit and address the patient's own concerns explicitly (a classic ICE application)
• Give a clear, specific safety net even where the plan is reassurance and discharge

Avoiding findings being lost

A recurring system failure is an incidental finding that is noted in a report but never actioned, because responsibility for follow-up was unclear between the requesting clinician, the reporting radiologist/pathologist, and the patient's GP. Good practice includes:

• Explicitly stating in the referral or request who is responsible for actioning the result
• Using structured surveillance pathways (e.g. a pulmonary nodule clinic) with their own recall system rather than relying on an individual clinician's memory
• Coding and flagging the finding in the patient record so it surfaces at future encounters
• Ensuring the patient themselves understands the plan, since they are often the most reliable safety net for their own follow-up

Red flags

Shared decision-making around surveillance

Where a finding sits in genuinely uncertain territory - not clearly benign enough to discharge, not clearly concerning enough to demand urgent action - a surveillance plan (repeat imaging at a defined interval) is often the right compromise, but it should be agreed with the patient rather than simply scheduled and communicated as a fait accompli.

• Explain why surveillance, rather than immediate further investigation or reassurance, is the recommended approach
• Give a clear timeframe for the next test and what will happen depending on the result
• Check the patient's own tolerance for uncertainty - some patients prefer more frequent reassurance imaging even where guidelines would support a longer interval, and this preference should be discussed rather than overridden
• Ensure the patient knows what to do if new symptoms develop before the scheduled surveillance date, rather than assuming they will wait passively

Summary principle

The overarching skill is proportionality: match the intensity of the response to the actual risk the finding carries, informed by recognised risk-stratification criteria for that finding type, the patient's own context and preferences, and clear communication - rather than defaulting either to reflexive reassurance or reflexive full work-up.

A useful discipline before ordering any test is to ask what will be done with each possible result, including the unexpected ones. If a scan is being requested to answer one question, but an abnormality elsewhere in the imaged field would generate a cascade of further investigation that the patient would not want and could not benefit from, that is worth anticipating - and occasionally worth discussing with the patient - before the test is ordered rather than after the report arrives. This is particularly relevant in frail patients with limited life expectancy, where the discovery of an indolent abnormality may generate substantial anxiety and investigation for no realistic gain.