SDMA changed feline nephrology. That is not an overstatement — before symmetric dimethylarginine became a standard panel item, the average cat presenting with CKD had already lost 40 to 50 percent of functional nephron mass. Creatinine does not rise until glomerular filtration rate drops below 25%. SDMA signals decline earlier, around 40% GFR loss. That gap matters enormously in clinical practice.
But 3 years of data from our partner clinics tell a more complicated story: SDMA alone misses roughly 22% of Stage 1 CKD cases when used without concurrent cystatin-C measurement. That is not a small rounding error. That is roughly one in five cats with early kidney disease leaving a wellness visit with no flag raised.
The Physiology Behind the Gap
SDMA is produced at a relatively constant rate through protein methylation and is excreted almost entirely by the kidneys. In theory, it should track GFR closely. In practice, three variables introduce noise:
- Muscle mass variability: SDMA production correlates with protein turnover. Cats that are cachectic, hyperthyroid, or have low lean body mass produce less SDMA per unit time. A cat losing nephron function while simultaneously losing muscle mass can maintain a normal SDMA reading longer than its GFR would predict.
- Protein intake: Cats on protein-restricted diets show systematically lower SDMA values. A cat recently switched to a renal diet by a previous clinician may appear to have better kidney function than it does.
- Analytical variability: SDMA assays from different manufacturers show coefficient of variation differences of 8 to 12% at the low end of the reference range. At SDMA values of 14 to 18 µg/dL — the zone where Stage 1 transitions to Stage 2 — that variability matters.
Cystatin-C is filtered by glomeruli, not secreted or reabsorbed in the tubules, and its production is not meaningfully affected by muscle mass or dietary protein intake. When SDMA is in a gray zone, cystatin-C provides an independent signal from a different mechanistic pathway.
What the Data Show
Across 847 cats tested at our partner clinics over 36 months, we identified 194 that were subsequently confirmed with IRIS Stage 1 or Stage 2 CKD within 12 months of their initial panel. In that group:
- SDMA alone flagged 152 of 194 (78.4%)
- Cystatin-C alone flagged 161 of 194 (82.9%)
- SDMA + cystatin-C together flagged 189 of 194 (97.4%)
The five cases that neither marker caught were all Stage 1 cats with concurrent hyperthyroidism — a known confound discussed below. The combined marker approach did not simply add sensitivity by combining two overlapping signals. The cats missed by SDMA were often caught by cystatin-C and vice versa, indicating genuinely complementary detection pathways.
The Concurrent Hyperthyroidism Problem
Feline hyperthyroidism elevates GFR through increased cardiac output and renal perfusion. A hyperthyroid cat with CKD will frequently have a GFR that appears normal or even elevated on standard markers, masking the underlying nephron loss. Treating hyperthyroidism then causes GFR to fall back toward the true baseline, which is when the CKD becomes apparent.
This is not rare. In our dataset, 41% of cats over age 10 presenting with confirmed hyperthyroidism had concurrent CKD that was detectable only after treatment. Identifying those cats before starting methimazole requires a broader biomarker panel — specifically, urinary biomarkers of tubular injury such as NAG (N-acetyl-beta-D-glucosaminidase) alongside SDMA and cystatin-C.
We flag this combination in our MetaDx feline renal panel specifically because the clinical consequences of missing it are significant. A cat started on methimazole without this information may develop acute CKD decompensation that was entirely predictable.
Practical Implications for Your Practice
The takeaway is not that SDMA is unreliable. It remains a genuinely useful marker and represents a substantial advance over creatinine-only monitoring. The issue is that treating it as a standalone screening tool for early CKD creates a false sense of certainty at exactly the stage where intervention has the most impact.
A tiered approach works better in practice:
- Annual wellness panels in cats over 7 should include SDMA, cystatin-C, urine specific gravity, and UPC ratio as a minimum set.
- Any SDMA in the 14-18 µg/dL range warrants immediate cystatin-C if not already run — do not wait for the 6-month recheck.
- Cats with concurrent hyperthyroidism need urinary tubular injury markers regardless of what SDMA shows.
- Trend analysis over 3+ data points is more informative than any single value. A cat with three consecutive SDMA readings of 14, 16, 18 µg/dL over 18 months is almost certainly in progressive CKD even if each individual reading falls within reference range.
Where Machine Learning Adds Value
The challenge with multi-marker renal assessment is that it requires integrating variables with non-linear relationships. The ratio of SDMA to cystatin-C, adjusted for muscle mass proxy markers like creatinine and albumin, behaves differently across breeds, age groups, and body condition scores. A Siamese at 12 years with a BCS of 3/9 needs different reference thresholds than a 7-year-old domestic shorthair at BCS 5/9.
Our model was trained on 40,000+ veterinary cases with known outcomes, allowing it to weight these interactions rather than apply static reference ranges. The result is a sensitivity improvement that static multi-marker panels cannot match — and a reduction in the kind of false negatives that lead to cats presenting with Stage 3 CKD at what should have been a routine annual visit.
SDMA was a breakthrough. But it was never meant to be the whole story. Use it as one input in a multi-marker assessment, and early detection rates improve substantially.