Using titers to diagnose disease: When is a positive a positive? (Proceedings)
For most diagnostic tests, the 'titer' is the minimum dilution of a substance that is required to yield a positive result.
For most diagnostic tests, the 'titer' is the minimum dilution of a substance that is required to yield a positive result. For example, for detection of anti-Leptospira sp. Antibodies, the titer is the dilution of serum which still causes microbes to crosslink. Unfortunately, the sensitivity and specificity of the available tests varies for each organism, for each laboratory, and even from day-to-day due to observer error. Regardless of the test being performed it is rare that a 'positive' result in and of itself is a guarantee of disease. This presentation will review the more common 'titer' tests available to us as clinicians, and discuss common methods of interpretation (and misinterpretation).
Antibodies are produced by B-cells and plasma cells. Initially IgM is rapidly produced to bind and neutralize antigens, fix complement, and opsonize microbes for phagocytosis. Once T-cells are recruited and recognize the activated B-cell, isotype switching to IgG or IgA occurs. Antibodies are then produced in large amounts and enter the systemic circulation. For unknown reasons some organisms are able to produce fulminant infections without induction of a significant antibody response; this likely occurs because they induce alternative immune responses that rely less on antibodies as effector molecules. After resolution of infection antibody concentrations may or may not decline over time, which may depend on whether the organism persists in the host.
The gold standard for diagnosis of leptospirosis is identification of organisms in urine or in tissues. Unfortunately these methods are very insensitive (organisms are rarely visible despite the presence of infection), invasive (renal biopsy), and not readily available (dark field microscopy). Therefore antibody agglutination titers are routinely used in conjunction with consistent clinical signs to diagnose cases. Agglutination tests detect antibodies by testing the ability of the patient's serum to cause intact Leptospira organisms to be crosslinked by circulating antibodies: when viewed with a microscope, crosslinking results in visible agglutination. Serum is tested against laboratory strains of multiple serovars in order to provide titers for each one. Unfortunately, the Leptospira spp. vaccines uses killed whole organisms to induce immunity, and as a result the antibodies that are produced are against multiple proteins, and cannot be distinguished from infection-induced titers by the agglutination test. Additionally infection results in cross-reacting antibodies against multiple serovars. Therefore, interpretation of Leptospira titers must be done with caution. In general:
1. The highest serovar titer is considered to be the infecting serovar. This rarely is clinically important, with the exception of those cases where hepatic failure is also present. Serovars icterohemmorhagica and canicola are most commonly associated with acute liver failure, whereas grippotyphosa has been associated with chronic hepatopathy.
2. The Leptospira spp. vaccines usually cause relatively low antibody titers against all serovars; the highest serovars may be the vaccinal ones, although this is not always true. Regardless, although titers may be high soon after vaccination (i.e. greater than or equal to 1:800), they frequently decrease to much lower concentrations.
3. A titer only indicates exposure or vaccination, and is not synonymous with disease. In the absence of renal dysfunction there is no indication that treatment is needed.
4. Infected dogs usually have high titers at the time of diagnosis. However in rare cases astute clinicians may test dogs early in their disease process, at which time the antibody titer may be low or negative. A convalescent titer may be required after a 10-14 day period to confirm a 4-fold increase in titer.
Feline enteric coronavirus is a common cause of transient diarrhea in kittens, although asymptomatic infection is likely just as common. In rare cases the coronavirus mutates and disseminates via the host's macrophage-phagocytic system, and infection results in the wet or dry forms of feline infectious peritonitis. At this time published data does not support the claim that any antibody test is capable of differentiating between the mutated and non-mutated viruses. As a result there is no such thing as a positive 'FIP' titer, only a positive coronavirus titer. There is no single mutation that causes FIP. Research has implicated mutations in two different viral proteins (3C and 7B) as being associated with the FIP viral phenotype. However mutations can occur in any site within these proteins, and thus attempting to detect antibodies against a single mutated form of these proteins is at this time not realistic. The only practical use to the coronavirus titer is as a screening test—negative patients are very unlikely to have FIP (although even this is not always true), whereas positive patients are of unknown status.
Histoplasmosis, blastomycosis, aspergillosis
Histoplasma capsulatum, Blastomyces dermatiditis, and Aspergillus sp. are widespread in the environment. Therefore it is reasonable to assume that many animals are exposed and develop antibody titers without developing disease; the specificity of these tests is thus relatively low. Conversely, many animals infected with these fungal agents never develop antibody titers that are measurable using existing assays, therefore the sensitivity is poor as well. Treatment for any of these diseases should never be instituted solely on the basis of titer results; in fact most internists no longer request H. capsulatum, B. dermatiditis antibody titers at all. Studies as to the usefulness of Aspergillus sp. titers are conflicting; a recent retrospective reported that this test may be more sensitive and specific than previously thought, although again internists will not treat solely based on the results of this test.
As with the previously discussed fungi, Coccidioides immitis is widespread in the appropriate climatic conditions and most animals and people are exposed at some time. Therefore low positive titers are common and do not merit therapy. However, unlike the other fungal organisms, active infection with C. immitis does usually cause readily detectable serum antibodies, which are usually detectable at a greater than 1:16 dilution (in other words, higher than animals that are exposed but not suffering from disease). Additionally the antibody titer usually increases or decreases with resolution or progression of disease. Therefore, this is perhaps the only antibody titer than can be used to monitor response to therapy, but titers will still rarely become completely 'negative' in animals that have had active disease.
The gold standard for diagnosis of Rocky Mountain Spotted Fever and the ehrlichioses is observation of the organism in blood smears. However this is a rare occurrence, and usually requires a trained eye. Rather than relying on this relatively insensitive test, antibody titers using the available IFA or ELISA tests is much more commonly used by both general practitioners and specialists. There is little cross-reactivity between rickettsial organisms, with the exception being closely-related species of Ehrlichia.
As with any antibody test, a positive result only indicates exposure rather automatically implying active disease. Particularly in endemic areas, positive test results must be accompanied by consistent clinical signs and clinicopathologic findings (RMSF: thrombocytopenia, petechiae/ecchymoses/vasculitis, vestibular disease, etc.; Ehrlichia spp.: polyarthritis, coagulopathy; thrombocytopenia, etc.). That being said, high titers in general do have a good correlation/specificity for active disease, particularly with RMSF. High titers with Ehrlichia canis are less specific, as dogs may harbor organisms for extended periods, and high titers that persist for months to years beyond resolution of infection are well-documented. Low titers or negative titers in the presence of consistent clinical signs may reflect early infection, particularly with RMSF where a lag time for rise in titer may be up to 3 weeks. In these cases a convalescent titer may be of benefit to document a 4-fold increase. Dogs with low-positive RMSF titers and no consistent clinical signs are occasionally encountered. This is likely due to non-pathogenic Rickettsiae that result in cross-reactive antibodies or because of historic exposure to the pathogenic RMSF organisms that did not result in clinical disease.
Bartonella spp. & Borrelia burgdorferi (Lyme Disease)
The following paragraphs on Bartonella spp. are directly quoted from: Pressler B. Bartonellosis. In: August JR, ed. Consultations in Feline Internal Medicine. 5th ed. St. Louis, MO: Elsevier Saunders, 2006;29-37.
The most widely available test for Bartonella spp. Detect the presence of anti-Bartonella antibodies. These tests include the immunofluorescent antibody test (IFA), which detects the presence of antibodies against whole organisms, and the Western blot, which detects antibodies against proteins from bacterial lysates. Most published studies on Bartonella spp. Infection and seroprevalence have used the IFA test, and the positive and negative predictive values of this test have been determined. Studies examining the accuracy of the Western blot are not available. The commercially available IFA and Western blot tests have not been compared directly to one another.
The high prevalence of seroreactivity to Bartonella spp. in cats makes antibody testing a poor method for diagnosing disease caused by this organism. Poor correlation exists between presence of anti-Bartonella spp. antibodies and illness. Other limitations to the antibody tests include inability to discriminate between different Bartonella spp. because of antibody cross-reactivity, and lack of correlation between titer magnitude and bacteremia, time since infection, or presence of clinical signs.
Antibody testing in cats may be useful in two situations. First, the high negative predictive value (>90%) of the IFA test allows rapid screening of cats. Although positive results do not always imply infection, absence of anti-Bartonella spp. antibodies is likely to mean that a patient is not infected. However, because a small number of infected cats remain seronegative this screening method does not reliably identify all infected cats. Second, a four-fold or greater increase in antibody titer may imply recent infection. This likewise should be interpreted with caution, because naturally and experimentally infected cats may have titers that fluctuate widely over time. A large increase in antibody titer should be accompanied by clinical signs to be supportive of Bartonella spp.-induced disease.
As with Bartonella spp. in cats, a large number of dogs in B. burgdorferi-endemic regions have been exposed to the Lyme organism at some point, and thus are antibody positive. Current evidence is that these dogs remain persistently positive because they never completely clear the organism. As a result a positive titer is never sufficient to diagnose Lyme disease, only exposure and likely infection. Only a small minority of dogs will go on to develop clinical signs—and even this statement is controversial. As with Bartonella spp., current recommendation is to use the presence of anti-B. burgdorferi antibodies as part of an overall diagnosis rather than sole confirmation of infection; clinical signs consistent with disease (polyarthropathy and/or protein-losing nephropathy) must be present.
As with Bartonella spp., the majority of household cats are asymptomatically, persistently infected with Toxoplasma gondii. Therefore, persistently positive IgG antibody titers, regardless of magnitude, are the rule rather than the exception. Consistent clinical signs must be present for a 'positive' antibody titer to potentially be significant. Even so, a single positive IgG titer is oftentimes not considered significant even with 'classic' toxoplasmosis signs. Instead a tentative diagnosis should rely on either 1) a four-fold increase in IgG titer, or 2) a positive IgM titer, as anti-T. gondii IgM antibodies usually decline to undetectable levels within approximately 2 months of infection. The other situation where a single antibody titer may be useful is when assaying anti-T. gondii antibodies in CSF; a higher antibody titer than that of the serum implies autonomous production within the CNS, and usually implies active infection.
Antigen titers are oftentimes the gold standards for determining infection. Unfortunately this is easier said than done when developing tests for infectious diseases—antibodies against an organism circulate throughout the entire body, whereas an organism may be confined to an inaccessible part of the host. For this reason the only infectious diseases in veterinary medicine that have established, widely available antigen titer tests are Cryptococcus neoformans and FeLV. Blastomyces dermatiditis and Histoplasma capsulatum have relatively newly available antigen tests available as well; although their sensitivity and specificity are still being established, they appear to be very reliable tests in dogs.
The C. neoformans latex agglutination test is positive when circulating organism capsular proteins cause latex particles which are coated with antibodies to agglutinate. This test is very sensitive and specific. Because this test directly measures the amount of circulating organism protein, decreases in titer correlate with clinical improvement and resolution of disease. There is no clear cut-off for a positive versus a negative result however, so therapy should be continued for some time after a 'negative' titer is obtained. Although most laboratories which offer this diagnostic test perform the latex agglutination method, there is some variation amongst them in regards to methodology. For this reason a single laboratory should be used to monitor response to therapy.
Blastomyces dermatiditis and Histoplasma capsulatum
Antigen tests are available for Blastomyces dermatiditis and Histoplasma capsulatum through MiraVista Diagnostics (Indianapolis, IN). The reported sensitivity and specificity of this antigen test for detection of dogs actively infected with Blastomyces dermatiditis has been reported to be greater than 85%, with the accuracy of the urine antigen test being superior to the serum test. Although a larger number of dogs must still be tested in order to establish that this test is accurate in dogs with a variety of other infectious agents, anecdotally (including in my experience), the urine antigen test is highly accurate, whereas results using serum are poorer than reported. Once this test is explored further, it may be that this will become the true gold standard for determining infection; however, at this time repeat attempts to isolate organisms by traditional means (cytology of skin lesions, transtracheal wash, etc.) should still be performed prior to treating solely based on antigen test results. In addition, the Blastomyces dermatiditis and Histoplasma capsulatum are known to cross-react using the various antigen tests, so overall clinical picture may be required to differentiate in the presence of a positive result. Finally, this test has not been evaluated in cats as of yet.