Endocrine Testing: Where a Single Number Misleads Most
Quick Answer — Endocrine testing fails more often from interpretation than from assay performance. Three patterns account for most of the damage: diagnosing hypothyroidism on a low T4, excluding hyperthyroidism on a normal T4, and diagnosing Cushing’s on a single cortisol. Each is preventable with a panel structure and an interpretation rule, and each is currently a common cause of lifelong unnecessary medication.
Quick Answer
Endocrine numbers are conditional. The condition is: which other test, which clinical signs, which drugs, and which species.
Thyroid: dogs
| T4 | TSH | Interpretation |
|---|---|---|
| Low | High | Supports hypothyroidism |
| Low | Normal | Euthyroid-sick, drug effect, or early disease — treat the other cause and retest |
| Normal | High | Unusual; consider interference, supplementation, non-thyroidal illness |
| Normal | Normal | Hypothyroidism unlikely |
The second row is where most diagnoses should not be made. A dog with a serious non-thyroidal illness, or on glucocorticoids or phenobarbital, can have a T4 in the hypothyroid range and recover normal function. Starting levothyroxine at that moment creates a patient who may never come off it.
Thyroid: cats
| T4 | Clinical picture | Interpretation |
|---|---|---|
| High | Compatible signs | Hyperthyroidism; treat |
| High-normal | Compatible signs | Suspicious; repeat in 2–4 weeks |
| Mid-normal | Compatible signs | Not excluded; repeat testing is the standard step |
| Low | Concurrent illness | Illness suppression possible; reassess after treatment |
Two feline-specific interactions matter: non-thyroidal illness suppresses T4 in exactly the population that develops hyperthyroidism, and treating hyperthyroidism can unmask azotaemia because increased filtration was masking reduced renal function. Both T4 and renal parameters belong in the same monitoring loop.
Adrenal: dogs
| Test | Use | Limitation |
|---|---|---|
| Resting cortisol | Rule-out of hypoadrenocorticism when clearly elevated | Cannot diagnose Cushing’s |
| ACTH stimulation | Cushing’s diagnosis and trilostane monitoring | Assay-specific cut-offs; ACTH preparation required |
| Low-dose dexamethasone suppression | Cushing’s diagnosis | Assay-specific cut-offs; time points matter |
Data pending. This section reports measured values from internal validation studies. They are published once the supporting study is complete; the current report is available from technical support on request.
The cut-off portability issue is not academic: importing thresholds from another assay produces systematic misclassification in both directions — treating normal dogs and missing affected ones.
What a defensible panel looks like
- Thyroid: T4 with TSH, in a dog not receiving steroids or anticonvulsants where possible, with a retest interval defined in advance.
- Feline thyroid: T4 with renal parameters, and a documented retest plan for the equivocal cat.
- Adrenal: resting cortisol only as a rule-out, with dynamic testing as the diagnostic step and the correct cut-offs for the assay in use.
- Monitoring: consistent sampling time relative to dosing, and the clinical response driving dose changes rather than the number alone.
Gaps in the current menu, stated plainly
Data pending. This section reports measured values from internal validation studies. They are published once the supporting study is complete; the current report is available from technical support on request.
A clinician who does not know the boundary will assume a thyroid panel covers diabetes monitoring, and will request the wrong test.
Six traps
- Diagnosing hypothyroidism on T4 alone. The most common endocrine error.
- Excluding feline hyperthyroidism on one normal T4. Repeat.
- Diagnosing Cushing’s on a resting cortisol. Dynamic testing is required.
- Importing cut-offs from another assay. Systematic misclassification.
- Ignoring drug effects. Steroids and phenobarbital lower T4.
- Changing sampling time across visits. Produces false trends in monitoring.
The Author’s Take
Position. The endocrine pages should be written to prevent prescriptions, not to encourage testing — and a vendor that publishes “do not diagnose on this alone” will sell more tests over five years, not fewer.
Reasoning. The commercial temptation is to make endocrine testing easy: a single abnormal number, a clear recommendation, a repeat prescription. That path produces over-diagnosis, and over-diagnosis eventually produces a backlash — clinicians who have been burned by a misdiagnosed dog become sceptical of the whole menu. The vendors who take the slower route, insisting on panels and retest intervals, end up with clinicians who trust the assays because they have seen them get it right in difficult cases. Endocrine testing is a judgement product; the vendor that supplies judgement is the one that stays on the shelf.
Disclosure. Written from published veterinary endocrinology guidance; no new experimental data. Values marked are placeholders for Migibio’s platform data.
Related pages
- Canine hypothyroidism · Canine hyperadrenocorticism / hypoadrenocorticism · Feline hyperthyroidism
- Canine total T4 · Feline total T4 · Canine cortisol
Sources
- Reference to be added.
- Reference to be added.
- CLSI EP28-A3c — reference interval methodology.
Frequently asked questions
- Can hypothyroidism be diagnosed from a low T4 alone?
- No. A low total T4 occurs in genuine hypothyroidism, in euthyroid-sick dogs, and in dogs receiving glucocorticoids or phenobarbital. Diagnosis requires compatible clinical signs plus a low T4 with an elevated TSH, or additional testing. Treating on T4 alone is the most common endocrine over-diagnosis in small animal practice.
- Why does a cat with hyperthyroidism sometimes have a normal T4?
- Early or mild disease can sit inside the reference interval, and concurrent non-thyroidal illness suppresses T4. The standard response is to repeat testing after an interval, or to pursue free T4 where available, rather than to exclude the diagnosis.
- Is a single cortisol measurement useful for Cushing's disease?
- Not for diagnosis. A single resting cortisol can neither confirm nor exclude hyperadrenocorticism; dynamic testing (ACTH stimulation or low-dose dexamethasone suppression) is required, and the cut-offs belong to the specific assay. The one exception is a clearly elevated value making hypoadrenocorticism very unlikely in an unstable patient.