Diabetic Nephropathy Diagnosis
Diabetic Nephropathy Diagnosis
Diagnosing kidney disease in diabetes is usually a matter of two tests repeated over months rather than of any procedure. What requires judgement is the minority of cases: the ones where something does not fit the expected pattern, and where a kidney biopsy is the only way to find a second, treatable disease sitting on top of the diabetes. This page sets out how the routine tests are performed and read, and how that minority is recognised.
The screening tests, and why they are repeated
Annual screening in diabetes is two measurements, both cheap and both done on a single visit. The first is a spot urine albumin-to-creatinine ratio, taken on a random sample rather than a timed collection, because correcting albumin against creatinine on the same sample removes the effect of how concentrated the urine is. The second is a filtration rate calculated from blood creatinine.
Repetition is built into the test rather than added to it. Albumin excretion varies substantially between days, and it rises temporarily with exercise, fever, a urinary infection, heart failure, menstruation and high glucose — so a single abnormal result carries little information. Persistence is established when two of three samples are abnormal across three to six months.
| Test | The condition for it to mean something |
|---|---|
| Urine albumin-to-creatinine ratio | Collected when well — not during a fever, a urinary infection, heavy exercise or a heart failure exacerbation — and repeated to confirm persistence |
| Filtration rate from creatinine | Compared against earlier values; a single value without a comparator cannot show a slope |
| Blood pressure | Measured with a correct cuff after sitting quietly, and ideally supported by home readings |
| Urine microscopy | Read by someone looking for cells and casts, not only by an automated strip |
How often screening happens depends on the type of diabetes and is set out on the causes page: in type 1 diabetes the clock is meaningful and screening usually begins five years after diagnosis, while in type 2 diabetes it begins at diagnosis, because the disease has often been present silently for years before it was found.
When the picture is diagnostic on its own
In most people with diabetes and kidney disease, no biopsy is needed, and performing one would add risk without changing anything. The diagnosis is clinical, resting on a pattern rather than on tissue.
| Feature | What typical looks like |
|---|---|
| Duration of diabetes | Long — usually more than ten years in type 1, and long or unknown in type 2 |
| How the albumin rose | Gradually, from normal to modest to heavy, across years |
| Urine sediment | Bland — few cells, no red cell casts |
| Eye findings | Diabetic retinopathy present, often at a similar stage of severity |
| Rate of filtration decline | Slow and steady, measurable year on year rather than week on week |
| Other complications of diabetes | Often present — neuropathy, or vascular disease elsewhere |
Where all six rows hold, the diagnosis is diabetic kidney disease and the treatment is the one set out on the treatment page. Where any of them fails, the next section applies.
The four features that mean this may not be diabetic kidney disease
A separate kidney disease occurring in a person who has diabetes is common enough to have a name — non-diabetic kidney disease, sometimes found alongside diabetic changes in the same biopsy. It matters because its treatment is often immune-directed and has nothing to do with glucose, and because assuming the diabetes explains everything is the standard way it gets missed.
| Feature | Why it does not fit | What is done |
|---|---|---|
| A short duration of diabetes | Diabetic kidney disease takes years to develop; heavy protein loss within a year or two of diagnosis is not it | Biopsy, with a full serological screen for a primary glomerular disease |
| An active urine sediment | Red cells, especially dysmorphic ones, and red cell casts indicate inflammation in the filter, which diabetic disease does not produce | Biopsy; the finding usually reclassifies the disease entirely |
| A rapid fall in filtration | Diabetic disease declines steadily; a drop over weeks suggests a crescentic or vasculitic process | Urgent assessment and biopsy, because the window for treatment is short |
| Absent retinopathy in type 1 diabetes | The two complications share a mechanism and usually appear together; their dissociation is informative | Biopsy, particularly where protein loss is heavy |
| Nephrotic-range protein appearing abruptly | The typical course is a slow climb rather than a step change | Biopsy, with attention to membranous and other primary glomerular diseases |
A biopsy is a procedure with real risk, and the decision is made by comparing that risk against the alternative — treating for years on the basis of an assumption. Where one of these features is present, the assumption is the riskier option. Our nephrology department performs kidney biopsy and reads the tissue with light microscopy, immunofluorescence and electron microscopy, which is what any of these differential diagnoses require.
How the two numbers are read together
The two measurements are plotted against each other rather than read in isolation. Filtration gives the G grade, albumin gives the A grade, and the combination carries the risk: a moderate albumin loss matters more at a reduced filtration rate than at a normal one, and a normal filtration rate does not neutralise a heavy albumin loss.
What clinicians act on is the change rather than the value. A filtration rate that has fallen across four readings over two years is a slope, and a treatment decision made from a slope is a different thing from one made from a single number. The same applies in reverse: an albumin result that has come down and stayed down over a year is evidence that the regimen is working, even if the absolute value is still above the normal range.
One caution keeps the reading honest. The filtration estimate is calculated from creatinine, which is produced by muscle — so it behaves differently in people with very large or very small muscle mass, and it drifts with diet and hydration. Where precision matters, a second marker or a measured clearance is used. The direction of travel survives these imprecisions; a single value does not.
What the eye examination adds
A dilated retinal examination is the one test on this page that is not about the kidney, and it is among the most informative. The small blood vessels of the retina are damaged by the same process that damages the filtering units, so the two complications run together — and the eye can be looked at directly, which the filter cannot.
Its value is asymmetric, and the asymmetry is useful. Retinopathy present alongside proteinuria supports a diabetic cause and reduces the need for a biopsy. Retinopathy absent in a person with type 1 diabetes and heavy proteinuria does the opposite: it makes a non-diabetic cause likely enough to justify the procedure. In type 2 diabetes the test is weaker in both directions, because retinopathy is common and its absence is less informative.
It is also a five-minute examination performed in an outpatient clinic, with no preparation and no recovery period. Among the tests on this page it is the least burdensome and one of the most decisive, which is a combination worth knowing about.