Limitations of MAT for the Serodiagnosis of Canine Leptospirosis
ACVIM 2008
Katharine F. Lunn, BVMS, MS, PhD, MRCVS, DACVIM
Fort Collins, CO, USA


Leptospirosis is a zoonotic bacterial disease of worldwide importance in both human and veterinary medicine.1 Disease is caused by serovars of the pathogenic species Leptospira interrogans sensu latu.2 Infection is maintained in nature by mammalian reservoir hosts that are endemically infected with host-adapted strains.3 Reservoir hosts do not typically develop clinical disease, but serve as a source of infection for incidental hosts, such as humans, through excretion of organisms in the urine. Examples of reservoir hosts include the vole and raccoon (both carrying the serovar Grippotyphosa), skunk and opossum (Grippotyphosa and Pomona), the rat (Bratislava and Icterohaemorrhagiae) and the mouse (Autumnalis).2

The Organism

Leptospires are motile obligate aerobic spirochaetes that have a bent or hooked appearance in liquid culture media. Prior to 1989, two species of the Leptospira genus were recognized: Leptospira interrogans (pathogenic strains) and L. biflexa (saprophytic strains).4 Within these species, numerous serovars (> 200 for L. interrogans) have been phenotypically defined, based on antigen expression detected by agglutination. Antigenically related serovars have traditionally been grouped into serogroups. Serogroups are useful for studying the epidemiology of leptospirosis, but they do not provide taxonomic information.4 More recently, the genus has been classified genotypically, resulting in the identification of several genomospecies.4,5 Examples include L. interrogans (L. interrogans sensu stricto), L. noguchii, L. meyeri, L. borgpetersenii, and L. kirschneri. These genomospecies include serovars of the "old" L. interrogans (L. interrogans sensu lato) and L. biflexa. This new classification system leads to confusion, because: 1) the genomospecies do not correspond to the previous phenotypic classification (L. interrogans and L. biflexa), 2) pathogenic and non-pathogenic serovars occur within same genomospecies, 3) the same serogroup can be associated with several genomospecies, and (iv) some serovars have multiple strains and genetic heterogeneity.4,5 Nonetheless, clinicians and epidemiologists continue to work with the serovar/serogroup system, and we will be using this approach in this discussion.

How is Leptospirosis Diagnosed?

As for any infectious disease, we can attempt to diagnose leptospirosis by finding the organism itself, or by detecting an immune response to the organism. However, as with other infectious diseases, neither finding the organism nor detecting an antibody response necessarily equate to clinical disease. For example, asymptomatic dogs may shed leptospires in the urine, and dogs may have circulating antibodies to leptospires as a result of vaccination or exposure. Thus the results of diagnostic testing should always be interpreted in light of the clinical signs and vaccination history. One of the difficulties associated with the evaluation of diagnostic tests for leptospirosis is the fact that there is no readily accessible "gold standard" test for ante mortem diagnosis in the dog. For most patients, the diagnosis is based on a combination of clinical signs, clinicopathological findings, and MAT titers, sometimes combined with PCR or other methods of organism identification. In this discussion we will focus on an antibody detection test, the MAT.

What is the MAT?

The MAT is the microscopic agglutination test, which is still considered to be the reference method for diagnosis of leptospirosis.4 In this test, dilutions of a patient's serum are incubated with suspensions of live leptospires. The suspensions are then examined by darkfield microscopy for evidence of agglutination. The end-point is the highest dilution of the patient's serum at which 50% agglutination occurs. The MAT is available at many diagnostic laboratories, and each laboratory will maintain cultures of serovars that are locally common or that are expected to cause disease in the animal species of interest. Most laboratories offer the MAT for serovars Canicola, Icterohaemorrhagiae, Pomona, Grippotyphosa, Hardjo and Bratislava. Autumnalis and other serovars may be available upon request. Although the laboratories will report results for the serovars tested, the MAT is a serogroup-specific assay.4

What Are the Advantages and Disadvantages of the MAT?

The advantages are that the test is widely available, it has been used for many years, and there is a wealth of data available from use of the MAT in humans and animals. Disadvantages include the safety concerns associated with laboratory maintenance of live cultures of leptospires, the subjective nature of the interpretation of the end-point of the test, the risk of cross-contamination of live cultures, and the significant occurrence of cross-reactions between different serogroups. The MAT may also give either false positive or false negative results in the diagnosis of leptospirosis, as discussed below.

How is the MAT Used in the Diagnosis of Canine Leptospirosis?

In a patient with consistent clinical signs and clinicopathological abnormalities, a single MAT titer of 1:800 or greater is often considered to be diagnostic for leptospirosis.1,2 If leptospirosis is suspected and the initial titer is not confirmatory, a convalescent titer should be performed. A four-fold (or greater) rise in titer between paired sera is regarded as confirming the diagnosis.4 The timing of the convalescent titer likely depends on the interval between the onset of clinical signs and presentation of the patient.4 We have observed seroconversion within eight days in patients that were clinically ill, but seronegative, at the time of the first MAT. Therefore it may not be necessary to wait for the typically recommended 2-4 weeks.

How Sensitive is a Single MAT Titer for the Diagnosis of Canine Leptospirosis?

We have evaluated the sensitivity of the first MAT titer submitted from dogs in which leptospirosis was subsequently confirmed (by organism detection by PCR, or by demonstrating a four-fold, or greater, rise in titer on a subsequent MAT). Serum samples were submitted to several laboratories for MAT testing with a typical range of serogroups. Sensitivity of an initial titer of > 1:400 (for any serogroup) ranged from 50% to 67%. Sensitivity of an initial titer of > 1:800 (for any serogroup) ranged from 22% to 67%. These results are not surprising, and confirm that a single negative or low MAT titer cannot be used to rule out leptospirosis in the early stages of disease.

Can Mat Results be Used to Determine the Infecting Serovar?

No, the available evidence suggests that the MAT cannot be reliably used to predict the infecting serovar in a patient with leptospirosis. In a large study of culture-proven leptospirosis in human patients, it was found that the predominant serogroup (based on MAT results) predicted less than 50% of the serovars isolated.6

In both human and veterinary medicine it is well-recognized that patients with leptospirosis have high MAT titers to more than one serogroup. These cross-reactions are attributed to the presence of several common antigens among leptospires. In veterinary medicine it is commonly assumed that the serovar with the highest MAT titer represents the infecting serogroup.1,2 We believe that this may be incorrect. In our studies we have found that when a single serum sample is obtained from a canine patient with leptospirosis, and then divided and submitted to several diagnostic laboratories, the identity of the serogroup with the highest titer varies between laboratories.7 The reason for this variation is not known, but it implies that the results of a single MAT from a single laboratory may not accurately predict the infecting serogroup in a patient with leptospirosis.

In the human literature, "paradoxical" immune responses are well-documented in leptospirosis, in which the highest MAT titer is detected to a serogroup unrelated to the infecting strain.4,6 These reactions are particularly noted in the acute stages of leptospirosis, whereas the results of convalescent MAT titers appear to more accurately reflect the infecting serogroup. This is attributed to the presence of common antigens between different leptospiral serovars, the fact that the MAT detects both IgG and IgM antibodies, and the different rates of decline of cross-reacting titers.4 We have followed MAT titers over time in several dogs with confirmed leptospirosis and have found that the identity of the serogroup with the highest MAT titer can change over time.7

How Specific is an Elevated MAT Titer For the Diagnosis of Leptospirosis?

We have evaluated the specificity of MAT titers in dogs with clinical signs and laboratory abnormalities consistent with leptospirosis in which an alternate diagnosis was confirmed. Serum samples were submitted to several laboratories, and specificity ranged from 69% to 100%, depending on the cut-off titers used, and the laboratory to which the samples were submitted. Seroprevalence studies in healthy dogs also demonstrate that the detection of antibodies by the MAT does not necessarily correlate with clinical disease. In Michigan, almost 25% of healthy dogs tested had evidence of exposure to leptospires.8 We have also surveyed healthy unvaccinated dogs in Wisconsin, and found a seroprevalence rate of 11%.9 Finally, when tested in specific pathogen free dogs, the MAT appears to have an extremely low false-positive rate.

How Does Vaccination Affect MAT Titers?

Vaccination of dogs with commercially available leptospirosis vaccines will produce detectable MAT titers; however they appear to persist for a relatively short time. MAT titers were studied in puppies and adult dogs after giving a commercial Pomona/Grippotyphosa subunit vaccine.10 Titers to Grippotyphosa were < 1:400, most titers to Pomona were < 1:800, and titers to Autumnalis were as high as 1:12,800. All Pomona and Grippotyphosa titers were < 1:100 by 16 weeks post-vaccination, but low Autumnalis titers were still present at this time.10 We have also observed high MAT titers to Autumnalis and Bratislava after vaccination with a subunit vaccine against the Canicola, Icterohaemorrhagiae, Pomona, and Grippotyphosa serovars. After using inactivated whole-cell bacterin vaccines against Canicola and Icterohaemorrhagiae, high MAT titers are detected for up to 10-12 weeks post-vaccination, and are usually decreased to <1:100 by 16 weeks after vaccination. Some dogs have been noted to develop titers as high as 1:3200, and these titers may persist for 6 months or longer after vaccination.2,11

What is the Significance of Titers to Autumnalis?

This is unclear. As described above, more than one investigator has noted high titers to Autumnalis after vaccination against other serovars. In our studies we have also found that the magnitude of MAT titers to Autumnalis varies greatly between laboratories, and titers to this serovar can be very high in dogs that do not have leptospirosis.

Is the MAT a Useful Test?

Yes, most definitely. Although the results of the MAT may not be reliably used to determine the identity of the infecting serogroup, the results of the test can be used to confirm a diagnosis of leptospirosis. The MAT is widely available and relatively inexpensive. In the presence of appropriate clinical signs and clinicopathological findings, a single high MAT titer, or four-fold rise in titer in paired samples, provides strong evidence in support of the diagnosis.


1.  Langston CE, et al. Vet Clin North Am Small Anim Pract 2003;33(4):791.

2.  Greene CE, et al. In: Infectious Diseases of the Dog and Cat, Greene (Ed), 3rd Ed, W. B. Saunders, Philadelphia, 2006; 402.

3.  Heath SE, Johnson R. JAVMA 1994;205(11):1518.

4.  Levett PN. Clin Microb Rev 2001;14(2):296.

5.  Bharti AR, et al. Lancet Infect Dis 2003;3:757.

6.  Levett PN. Clin Infect Dis 2003;36:447.

7.  Miller MD, et al. J Vet Intern Med 2007;21(3):624.

8.  Stokes JE, et al. JAVMA 2007;230(11):1657.

9.  Lunn KF, Sawchuk SA. Proceedings of the 82nd Annual CRWAD Meeting, 2001.

10. Barr SC, et al. AJVR 2005;66(10):1780.

11. Bolin CA. Sem Vet Med Surg 1996;11(3):166.

Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Katharine Lunn, BVMS, MS, PhD, MRCVS, DACVIM
Laporte, CO