1.3 Special Tests, Investigations & Diagnostic Accuracy

Investigations Follow the Differential, Not the Other Way Round

Outcome B2.5 asks candidates to suggest appropriate investigations and interpret these to diagnose oral conditions. The examinable principle is that an investigation is only worth doing if its result will change management. A test ordered without a question in mind produces findings that must then be explained, and in radiography it also fails the IR(ME)R justification requirement.

Sensibility Testing

The tests in daily use assess nerve response, not blood supply, so the accurate term is sensibility testing, not vitality testing.

Three rules make sensibility testing usable in an SBA stem:

1. Always test controls — the contralateral tooth and an adjacent tooth — before interpreting the tooth in question.
2. A response is not a diagnosis. EPT tells you a nerve conducted; it says nothing about whether the pulp is healthy, inflamed or partly necrotic.
3. A multi-rooted tooth can have partial necrosis and still respond, which is why a persistently symptomatic molar with a normal response still deserves radiographic and percussion assessment.

Percussion, palpation of the buccal sulcus, mobility, transillumination for cracks and selective biting on a Tooth Slooth complete the chairside set.

Radiographic and Other Imaging Investigations

• Periapical (paralleling technique) — the standard for apical pathology, root morphology, working length and periodontal bone levels at a single site.
• Bitewing — interproximal caries and crestal bone; the interval is set by caries risk, roughly six-monthly to annually in high-risk patients and around two-yearly in low-risk adults.
• Panoramic — a screening view of both jaws for pathology, third molars, and trauma; poor for fine detail and not a substitute for a periapical.
• Occlusal — sialoliths in the submandibular duct, unerupted anterior teeth, palatal expansion of a lesion.
• Cone beam CT (CBCT) — only where conventional imaging cannot answer the question: third molar to canal relationship, localisation of impacted teeth, complex endodontic anatomy, assessment of a bony lesion. The dose is substantially higher than an intra-oral film and the whole imaged volume must be reported.
• Ultrasound for salivary gland and neck lumps; MRI for soft tissue and perineural spread; CT for bone and staging.

Blood and Microbiological Investigations

Diagnostic Accuracy: Outcome B2.4

Outcome B2.4 asks for an understanding of sensitivity and specificity, and it is examined with real numbers.

• Sensitivity = TP / (TP + FN) — the proportion of diseased people the test detects.
• Specificity = TN / (TN + FP) — the proportion of healthy people the test correctly clears.
• Positive predictive value (PPV) = TP / (TP + FP) — of those who test positive, how many truly have disease.
• Negative predictive value (NPV) = TN / (TN + FN).

Sensitivity and specificity are properties of the test and do not change with prevalence. Predictive values do. Screen a low-prevalence population with even a highly specific test and most positives will be false, so the PPV falls. This is precisely why population screening for oral cancer with an adjunctive device performs poorly compared with targeted examination of high-risk patients.

Two mnemonics summarise the clinical use:

• SnNout — a highly Snsitive test that is Negative rules the disease out.
• SpPin — a highly Specific test that is Positive rules the disease in.

Worked example

A salivary biomarker for oral cancer has sensitivity 90% and specificity 95%. Applied to 10,000 people in whom the prevalence is 0.1% (10 cases): true positives = 9; false positives = 5% of 9,990 = about 500. PPV = 9 / 509, under 2%. The same test used in a clinic where 20% of attenders have a suspicious lesion produces a PPV above 80%. The test did not change — the population did.

Clinical Photography (Outcome C6.5)

Outcome C6.5 asks specifically for an understanding of the role and limitations of clinical photography in diagnosis, longitudinal care and referral.

Where it helps:

• Longitudinal monitoring of a mucosal lesion, tooth wear or gingival recession, where a standardised image taken at intervals is far more reliable than a written description.
• Referral — an image attached to a referral lets the receiving clinician triage urgency, and is particularly valuable for mucosal lesions.
• Consent and communication — showing patients what you can see supports shared decision-making and behaviour change.
• Records — an image is part of the clinical record and part of the medico-legal evidence base.

Its limitations:

• Colour rendition varies with lighting, flash, white balance and camera; a lesion can look substantially redder or paler between two images of the same mouth, so colour change must not be judged from photographs alone.
• Angle, magnification and retraction are rarely reproducible without a standardised protocol, so apparent change in size may be artefact.
• A photograph cannot assess induration, texture or fixation, all of which are palpation findings, and it cannot replace biopsy.
• Data protection applies. Clinical images are identifiable personal data: obtain and record specific consent, store them within the clinical record rather than on a personal device, and obtain separate consent for any teaching or publication use.