Diagnosing infectious disease
Historically, veterinarians have had four options for diagnosing most infectious agents.
Historically, veterinarians have had four options for diagnosing most infectious agents. One is to look under a microscope at a patient's blood or tissue sample for the organisms themselves, but this can be a highly insensitive method, says Edward Breitschwerdt, DVM, Dipl. ACVIM, professor of medicine and infectious diseases at North Carolina State University College of Veterinary Medicine and adjunct professor of medicine at Duke University Medical Center. A second option is to culture the organisms, growing them in special media to identify the specific infectious agent. However, most vector-borne organisms have been successfully grown only in specialized research laboratories, Breitschwerdt says.
The third option is an antibody test to indicate whether an animal was exposed to a particular organism (also known as an antibody titer). And the fourth is a polymerase chain reaction (PCR) test that detects an antigen or an infectious agent's DNA, both of which confirm active infection in most instances. The diagnostic advantage of PCR testing, which copies specific sections of an organism's DNA, over serology (antibody titers) is in diagnosing disease in the very early stages of infection, before the development of antibodies. PCR testing is generally considered as sensitive and specific as culture, is significantly faster and, for some organisms, is more reliable.
"Every diagnostic test has its limitations," Breitschwerdt says. "It's as important to know what a test will not tell you as what it will tell you."
Indications for serologic testing
The main use of antibody titers in companion-animal medicine is to help diagnose a number of different infectious diseases. These titers are useful for indicating exposure to a bacterial, viral or fungal agent, says Craig Datz, DVM, Dipl. ABVP, assistant teaching professor at the University of Missouri's College of Veterinary Medicine. Some of the common diseases that titers may help diagnose are distemper, ehrlichiosis, anaplasmosis, Rocky Mountain spotted fever, leptospirosis, borreliosis, blastomycosis, histoplasmosis, cryptococcosis, coccidioidomycosis, toxoplasmosis, feline heartworm disease and feline immunodeficiency virus. They also measure protection from infectious diseases, such as after vaccinations, and are sometimes used in healthy animals as screening tests to evaluate possible exposure to an infectious agent.
Veterinarians can run titer results in-house in some cases, but most submit titers to reference laboratories where results can take a few days, Datz says.
Interpretation of results is the most difficult and controversial aspect of using titers for diagnosing disease, Datz says. Different infecting organisms produce different antibody responses, and animals do not always develop a disease just because they have been exposed to an infectious agent. Also, a high antibody titer does not prove that disease is present or that treatment is needed.
Technicians obtain an antibody titer by doubling dilutions of the reactive blood (1:4, 1:8, 1:16, 1:32, etc.) until the test becomes nonreactive or negative. The titer that is reported to veterinarians is the last dilution that gave a reactive or positive antibody detection point. Although dilution ranges vary from laboratory to laboratory, the laboratories typically stop the titer at 1:256 and report the result as > 1:256 if the pet's blood is still positive. For example, if the titer was reported as 1:32, it means the serum was diluted serially out, and the 1:32 dilution gave the last positive response for presence of antibody, says Richard A. Hesse, MS, PhD, associate professor and director in the department of Diagnostic Virology at Kansas State University's College of Veterinary Medicine. If the titer is reported as ≤ 1:8, it means that the lab didn't test at a 1:4 or 1:2 dilution.
Sometimes the terms negative and positive are in themselves distracting. "Instead of reporting serology results as a positive or negative test for an antibody, we report serologic results as being reactive or not reactive," Breitschwerdt says. "This means the technician noted reactivity in the serum sent to the lab against the test antigen grown in the lab. This may or may not equate to the organism causing a disease in the patient, the patient being actively infected with that organism at the time of testing or the reactivity being specific for the infectious agent that is being targeted by the test."
The problem with using the term negative in reporting a titer is that the veterinarian's interpretation often is that the dog is not infected with the specific pathogen and, therefore, does not have the disease, Breitschwerdt says. However, with many diseases, the test may not detect a positive serologic response until the animal has been symptomatic for at least a week, so even a negative test, if done early in the course of infection, does not rule out the disease.
Conversely, the problem with using the term positive when reporting a titer is that veterinarians often interpret that positive report as the animal having the disease. "In reality, over half of the flea-infested cats in the Southeast United States have antibodies to Bartonella henselae, but many are not infected with the organism," Breitschwerdt says. As another example, dogs that are seroreactive to Ehrlichia canis antigens have been exposed, but in many cases their immune systems have cleared the infection.
Another challenge is that antibody titers often increase or decrease over time, and taking two or more samples at different time points may be necessary for diagnosis. A good practice is to draw serum samples when a sick animal is first presented, Datz says. Samples can be saved in a refrigerator for a few days or in a freezer for longer. If the veterinarian later suspects a specific infection, he or she can submit the stored sample.
For some diseases, the veterinarian should obtain titers on acute and convalescent samples. If the veterinarian stored the acute serum sample, he or she should submit it at the same time as the convalescent sample, Datz says. This reduces the variability in testing results that can occur when samples are run at different times. With these titers, a disease can be diagnosed based on comparing the results, looking for a large increase in antibodies as the disease progressed.
Using titers to detect acute disease
Serum samples are often nonreactive for antibodies during the acute stages of infection. For example, when a tick transmits Rickettsia rickettsii, the organism responsible for Rocky Mountain spotted fever, to a dog, it takes time before there will be a significant antibody titer, Breitschwerdt says. Most dogs with Rocky Mountain spotted fever don't have circulating antibodies to the organism when a veterinarian first examines them for the illness. If you take a sample right when the signs start, the animal has not had time to mount an antibody response, and serum antibody titer results are likely to be nonreactive.
"The more pathogenic the infectious agent, the more rapid the antibody response occurs, and the higher the antibody titer to that organism as compared with other organisms," Breitschwerdt says. For example, dogs infected with R. rickettsii will seroconvert in a matter of days, going from undetectable organism-specific antibodies to titers that exceed 1:8192. In contrast, dogs infected with Borrelia burgdorferi, the bacterium that causes Lyme disease, may not seroconvert until four to six weeks after tick transmission of the bacteria, and peak titers can occur months later.
The problem is that many serious infections are not confirmed because owners have to bring the patients back to get another sample, which then has to be sent to the laboratory, all necessitating an additional expense. If the pet is getting better, the owner might not be that interested in knowing that the infection was truly Rocky Mountain spotted fever.
Whenever possible, however, it's important to confirm acute infection to know what was actually treated. Confirmation can also have public health implications for the owners and their families as well. Some diseases, such as Rock Mountain spotted fever and leptospirosis, can be transmitted to people. If a veterinarian confirms a dog was infected with Rocky Mountain spotted fever or leptospirosis, he or she is obligated to inform clients that they and their children could be at risk for infection, Breitschwerdt says.
With leptospirosis, the route of infection is by direct or indirect exposure to urine from an infected animal. If leptospirosis is diagnosed in a dog, it's important for you and your staff to take care in handling the dog's body fluids, especially its urine. Wearing gloves is essential whenever urine and soiled bedding are handled. You should consider goggles for eye protection as well.
Using titers to detect chronic diseases
Using serology to diagnose chronic disease is even more complicated. One reason is because some animals that have been infected for months or years do not have antibody titers. For example, dogs infected with Babesia or Leishmania species can be overtly healthy, PCR positive, but lack detectable antibodies to the respective organism. For this reason, many governments require both a negative serology and negative PCR test before an animal can be imported.
Lack of antibodies often occurs with bartonellosis. Research over the last two or three years has shown that half of dogs infected with Bartonella species don't have circulating antibodies. In addition, PCR testing is often not sensitive enough to detect infection with a Bartonella species unless the patient's blood sample is enriched in a specially developed liquid culture medium (Bartonella alpha Proteobacteria Growth Medium, or BAPGM) before performing the PCR test, Breitschwerdt says.
Another example of the limitations of using titers as a diagnostic tool involves cases of Leishmania species infections in dogs imported to the United States. Pets brought here by military families returning from bases in the Mediterranean represent the most common cases. Established several years ago as a standard of care, these dogs are now tested to see whether they have been exposed to Leishmania infantum by using antibody titers, but many dogs don't make antibodies for years after they have been exposed to the organism, making this type of screening ineffective. The same appears to be true with babeisosis and bartonellosis.
Breitschwerdt and others have spent years researching and trying to educate veterinarians about finding these diseases in animals when antibodies are not present. "This is unsettling for veterinarians because, historically, the only real diagnostic tool we had for many infectious diseases was testing for antibodies," Breitschwerdt says. "Now, to find DNA evidence of infection in a diseased animal without the animal's immune system responding to the organism is a relatively new concept."
Without a definitive diagnosis, chronically infected patients may not be getting the appropriate treatment. Removing infectious diseases from the differential diagnosis list based on antibody titer alone can be a mistake. "You can't always rule out some of those organisms based solely on a negative antibody titer," Breitschwerdt says.
A note about PCR testing
The advent of DNA testing using PCR amplification has allowed veterinarians to confirm active infection, Breitschwerdt says. "PCR testing is analogous to obtaining a DNA fingerprint at a crime scene," he says. "It can confirm that the organism is present in a sample obtained from the patient on the same day the sample was obtained. However, PCR has its limitations, too, and is not 100 percent accurate for every patient or for every infectious disease."
In most acute illnesses, such as with anaplasomis, ehrlichiosis and leptospirosis, a PCR test is likely to detect the organism's DNA, but with many chronic diseases, such as babesiosis, bartonellosis and leishmaniasis, PCR can miss the organism in an infected animal. This is not to say that PCR is not a potentially useful test for these infections, only that each test has limitations.
"PCR can be the best diagnostic tool to detect a needle in haystack, but if you only have one needle and the haystack is too big, it can be hard to find, which equates to a negative PCR result in an infected patient," Breitschwerdt says. "Whenever possible, it is best to do both serology and PCR testing."
"Serology can tell you all kinds of information if you ask the right questions and have the right samples," Hesse says.
Veterinarians often ask which test they should run when they suspect an infection. The answer is more than one. Whenever possible, perform both a serologic test to determine whether there was exposure, and then do a culture, biopsy or PCR test to confirm that the patient is, in fact, infected with that organism. Obtaining samples before treatment greatly enhances a diagnostic laboratory's ability to assist the veterinarian with obtaining an accurate diagnosis.
Perhaps most important, before any testing is done, it's always necessary to perform a complete physical examination and get a detailed patient history, including vaccinations, travel and activities. The animal's history along with clinical signs will direct which tests are likely to be most beneficial. Results of serologic testing, a PCR test or any other diagnostic testing must be evaluated along with the patient's physical and historical findings to make an accurate diagnosis. "You need to look at the whole picture and not just the snapshot," Hesse says.
Ellen Jensen is a freelance writer and editor in Lawrence, Kan.