How the SAA Blood Test Detects Infection in Horses

Serum amyloid A, commonly called SAA, is the most sensitive blood marker of inflammation available in equine medicine. In healthy horses, SAA circulates at very low or undetectable levels, but when the body encounters infection, tissue injury, or significant stress, concentrations can spike dramatically within hours. That rapid rise and equally rapid fall once the problem resolves is what makes it so useful: a single blood draw can tell a veterinarian whether something inflammatory is happening right now, and repeated measurements can track whether treatment is working.

What SAA Actually Measures

SAA is a protein produced mainly by the liver as part of what immunologists call the acute phase response. When tissues are damaged or invaded by bacteria, the immune system sends out signaling molecules that prompt the liver to flood the bloodstream with acute phase proteins. In horses, SAA is the dominant one. Its concentration in the blood of a healthy horse is near zero, but during active inflammation it can climb to hundreds or even thousands of milligrams per liter within 6 to 12 hours.¹

What sets SAA apart from older markers like fibrinogen or white blood cell counts is speed. Fibrinogen takes days to climb and days to come back down. White cell counts can be hard to interpret because they shift for many reasons. SAA rises fast and falls fast. Once the inflammatory trigger is removed, concentrations drop back toward baseline within a day or two because the protein has a short half-life in the bloodstream.² That tight time course makes it possible to see, almost in real time, whether a horse is getting better or worse.

Why Veterinarians Prefer SAA Over Traditional Markers

For decades, equine vets relied on fibrinogen levels, total white blood cell counts, and neutrophil counts to gauge infection and inflammation. These markers still have their place, but they lag behind the clinical picture. A study comparing SAA to these traditional markers in sick foals illustrated the gap clearly: by the time foals had recovered and were ready for discharge, their SAA concentrations had returned to normal, while fibrinogen and white cell counts were still elevated.³ In practical terms, this means SAA gives the veterinarian a clearer real-time picture of whether the infection is truly resolving, rather than showing a residual echo of inflammation that has already passed.

A similar pattern shows up in adult horses with bacterial pneumonia. In one study tracking hospitalized horses on antimicrobial treatment, SAA concentrations dropped significantly over the course of therapy and correlated with clinical improvement, while fibrinogen, neutrophil counts, and white blood cell counts did not track the improvement nearly as reliably.⁴ If a horse on antibiotics has a falling SAA, the treatment is likely working. If SAA plateaus or climbs, the vet knows to reassess the drug choice or look for complications.

Post-Surgical Monitoring

Surgery always triggers some degree of inflammation, and a moderate SAA spike in the first few days after an operation is expected. The question vets care about is whether that spike resolves on schedule or keeps climbing, which could signal a post-operative infection. After routine castration, for instance, SAA levels typically peak around day three and return to pre-operative levels by about day eight if healing proceeds normally. In horses that developed complications, SAA remained elevated past that window.⁵

Perioperative antibiotics can blunt the post-surgical SAA response. Horses castrated standing that received penicillin alongside anti-inflammatory drugs had significantly lower SAA by day eight compared to those given anti-inflammatories alone.⁶ This is a useful baseline expectation: if your horse was given perioperative antibiotics and SAA is still climbing at a week out, something beyond normal surgical inflammation may be going on.

Emergency abdominal surgery for colic presents a trickier picture. The inflammation from both the colic itself and the operation can be intense, and that high background noise makes it harder for SAA to pinpoint an early surgical-site infection. Research has found that SAA at discharge was associated with surgical-site infection, suggesting it may be more useful in the later post-operative window than in the first couple of days when everything is inflamed regardless.⁷

Joint Infections and Synovial Fluid

Septic arthritis is an emergency in horses. Bacteria inside a joint can destroy cartilage fast, and early diagnosis is critical. SAA has proven valuable here, not only in blood but also when measured directly in the joint fluid. In experimental septic arthritis, both serum and synovial fluid SAA spiked sharply after bacterial inoculation, peaking around days three and four respectively. As the infection was eradicated, both dropped, though they remained above baseline for about nine to ten days. Crucially, SAA did not rise in joints that had sterile inflammation (non-septic synovitis) or in control joints, making it a useful tool for distinguishing true joint infection from simple irritation.⁸

There is a wrinkle, though. SAA can leak from the bloodstream into joints even when the joint itself is not infected. Horses with systemic inflammation (say, from an infection elsewhere in the body) can show elevated SAA in the synovial fluid of unaffected joints. The same study group found that synovial fluid SAA rose to measurable levels in non-affected joints during systemic inflammation and even in the opposite joint of horses with septic arthritis on one side.⁹ This means a single joint fluid SAA value should be interpreted alongside the blood SAA and the clinical picture, not in isolation.

For horses with acute limb wounds that may involve a joint or tendon sheath, repeated SAA measurements have been recommended as part of post-operative management. Serial testing tracked the course of synovial inflammation reliably and helped veterinarians adjust treatment in response to what was actually happening inside the joint.¹⁰

SAA in Foals

Neonatal foals are especially vulnerable to sepsis, and figuring out which sick newborn is septic and which has a non-infectious problem can be challenging. SAA has been investigated as a sepsis marker in hospitalized neonatal foals, with mixed results. The test is quite specific: at an optimal cutoff of roughly 1,050 mg/L, about 91% of non-septic foals were correctly identified as non-septic. But sensitivity was low, around 30%, meaning the majority of truly septic foals had SAA concentrations below that threshold.¹¹ In practical terms, a very high SAA in a newborn foal strongly suggests infection, but a normal SAA does not rule it out. It is a useful “rule-in” test but a poor “rule-out” test in neonates.

Reproductive and Peripartum Uses

In healthy broodmares, SAA stays low during late pregnancy and then briefly spikes within the first 36 hours after foaling before returning to baseline by about 60 hours post partum.¹² That brief post-foaling bump is normal and expected. What is not normal is a rising SAA during late gestation, which can signal ascending placentitis, a bacterial infection of the placenta that threatens both mare and foal.

In experimental placentitis models, SAA climbed within days of the infection being introduced. Mares whose SAA rose despite treatment had a much higher incidence of abortion compared to mares whose SAA stayed at baseline with therapy.¹³ Other acute phase markers like haptoglobin also climbed in infected mares, while peripheral white blood cell counts and fibrinogen often stayed normal, once again highlighting SAA’s advantage in detecting inflammation that traditional bloodwork can miss.¹⁴

Exercise, Transport, and False Alarms

One question horse owners commonly ask is whether exercise alone can raise SAA and produce a misleading result. The answer depends on the horse’s fitness level and the circumstances. In trained racehorses and experienced endurance horses, routine training did not change SAA concentrations. However, horses just beginning endurance conditioning did show a post-exercise increase.¹⁵ The likely explanation is that strenuous exercise in an unfit horse causes more tissue microtrauma and gut-barrier disruption than the same effort in a conditioned horse.

The combination of trailer transport and exercise amplifies this effect. One study found that horses who were trailered and then exercised had significantly higher SAA and markers of gut-barrier leakage compared to controls, suggesting that the stress of transport primes the body for a bigger inflammatory response to subsequent exertion.¹⁶ For practical purposes, if you are having your horse’s SAA tested for a medical reason, ideally draw the blood before or well after strenuous exercise and transport rather than immediately after a long haul followed by a workout.

Point-of-Care Testing and Its Limitations

One of the reasons SAA has become so popular in equine practice is the availability of stall-side point-of-care (POC) analyzers. These devices give a result within minutes using a small blood sample, which means the vet can make treatment decisions on the spot rather than waiting for a lab turnaround. The technology is genuinely useful but comes with caveats worth understanding.

Validation studies show that POC devices are linear (give proportional readings) up to fairly high SAA concentrations, roughly 1,000 to over 10,000 mg/L depending on the device and study. However, precision is lower than laboratory-based methods. One comparison found that the POC device had intra-assay variation of about 13 to 18% and inter-assay variation reaching up to 45% at very high concentrations, while laboratory turbidimetric assays generally stayed below 12%.¹⁷ A second comparison confirmed this pattern and additionally flagged a “hook effect,” where SAA concentrations above approximately 3,000 mg/L could paradoxically produce a lower reading on the POC device, potentially underestimating severe inflammation.¹⁸

Interfering substances also matter. Bilirubin and hemoglobin (from a hemolyzed or icteric sample) can push POC readings downward, while lipemia (a fatty sample, common after a meal) can push them upward.¹⁹ And whole blood gives significantly lower SAA results than serum or plasma, so the sample type matters.²⁰ None of this means POC devices are unreliable for clinical decision-making. It means the numbers are best interpreted as trends rather than precise absolutes: a reading of 500 one day followed by 50 the next day reliably tells you the horse is improving, even if the exact value on each individual reading carries some uncertainty.

Sample Handling

If a blood sample cannot be run immediately, how it is stored matters. SAA in serum appears stable for at least 17 days when stored at refrigerator temperature or room temperature, based on one evaluation.²¹ However, a separate study looking at broader analyte stability in equine serum stored at room temperature found that SAA could decrease over time, with changes of potential clinical significance.²² The safest practice is to separate serum promptly and refrigerate or freeze it if there will be any delay, and to let the laboratory know the sample conditions.

Where SAA Does Not Help

SAA is not a universal detector for every inflammatory condition in the horse. Its strength is systemic inflammation, meaning it responds to problems that trigger a whole-body acute phase response. Localized inflammatory conditions that stay compartmentalized may not move the needle on blood SAA at all. A study of horses with ulcerative keratitis (corneal ulcers) and anterior uveitis (moon blindness flares) found that neither condition was associated with elevated blood SAA or fibrinogen.²³ If your horse has an eye problem and the SAA is normal, that does not mean nothing is wrong. It means the eye’s inflammatory process is staying local and not generating a measurable systemic signal.

Dental Extractions and Oral Disease

One newer area of interest is using SAA to monitor horses undergoing dental work. Routine tooth extraction in a healthy mouth causes only a mild and brief SAA response. But horses with pre-existing periodontal disease showed a markedly different pattern. At 24 hours after extraction, horses with periodontal disease had a median SAA around 135 mg/L, while those without periodontal disease had a median of about 28 mg/L. By 48 hours, the gap widened further, with periodontal-disease horses reaching a median near 264 mg/L compared to near-zero in healthy mouths.²⁴ This finding has practical implications for equine dentistry: a rising SAA after a seemingly routine extraction could indicate underlying periodontal disease or a developing complication that warrants closer attention.

Serial Testing Versus Single Snapshots

Across nearly every clinical application, one theme emerges: a single SAA measurement is less informative than a series of measurements taken over time. A one-time reading tells you whether systemic inflammation exists at that moment, but it cannot tell you whether things are getting better or worse. The post-surgical studies on castration and colic, the joint infection work, and the pneumonia treatment monitoring all point to the same conclusion. Trends matter more than any individual number. A horse with SAA of 800 mg/L that was at 2,000 the day before is on a very different trajectory than one with SAA of 800 that was at 200 yesterday. The septic arthritis researchers made this point explicitly, concluding that serial measurements rather than a single value are needed to determine whether joint infection has been cleared.²⁵

This is worth discussing with your vet if your horse is being monitored for any inflammatory condition. A follow-up SAA draw 24 to 48 hours after the first one often tells you far more than either reading alone. The speed at which SAA responds, its main clinical advantage, also means a snapshot can be misleading if you catch it on the way up or down without knowing the direction.