Laboratory tests for Feline Infectious Peritonitis

Introduction – What is FIP?

Feline infectious peritonitis –FIP– is a severe disease of domestic and wild felids that occurs worldwide. The etiological agent is the feline coronavirus (FCoV) which mutates from the enteric, almost harmless biotype (feline enteric coronavirus, FECV) to the highly virulent, systemic biotype (feline infectious peritonitis virus, FIPV) 1. FCoV is a large, enveloped, positive-sense, single-stranded RNA virus which is commonly found in cats, with a seroprevalence above 90% in multicat households 2. This article offers a review of the etiopathogenesis of the virus and the diagnostic options for FIP, and although outside the scope of this paper, it is worth noting that whilst the condition has historically always been considered inevitably fatal, innovative therapeutic approaches (unlicensed in most countries) have recently shown good efficacy in treating the disease 3. 

Etiopathogenesis of FIP

Viral transmission is primarily fecal-oral, with other routes, such as salivary or transplacental, only rarely described 1. Litter boxes represent the main source of infection, where FCoV can survive in fecal matter up to 7 weeks 4. Kittens typically become infected when the maternal antibodies start to wane, usually around 5-6 weeks of age 5. FCoV then reaches the columnar epithelial cells of the small intestine where it replicates and can cause very mild (or occasionally more severe) gastrointestinal signs 6. Even in healthy cats the virus will replicate within monocytes, and can, therefore, be found in blood for a short period of time 7. 

Three major patterns of viral shedding in feces have been identified. A small percentage of cats (3-9%) appear to be resistant to infection and either never or only briefly shed the virus; 10-15% will shed long term or persistently, while the majority (70-80%) appear to intermittently eliminate the virus. This last pattern is probably a consequence of continuous re-infection and/or PCR testing limitations 1,8. The fecal excretion in young cats is very high, especially in multicat households. The higher the viral load, the greater the levels of virus replication and consequently the rate of mutation 8. Several genetically related but distinct viral populations will develop (quasispecies), and one will switch its cellular tropism to acquire the ability to both efficiently replicate within and activate the monocytes/macrophages, and spread systemically 1,8.

Additionally, the type of host-immune response, alongside additional factors (e.g., stress), may play a role in both the pathogenesis and the type of disease 8. In fact, while a cell mediated response seems to confer resistance to development of the disease, the “wet” form of FIP, characterized by cavitary effusions, depends on a massive B-lymphocyte mediated immune response. The non-effusive (dry) form seems to be the consequence of a partially effective cell mediated response, which confines the lesions to a limited number of organs 9. It is common to see an overlap between the two forms, with non-effusive cases developing effusions in the terminal stages, or effusive forms showing granulomatous lesions at necropsy 6. 

While it is widely accepted that the immune response may influence the course of infection, the precise mutation thought to be responsible for the shift from FECV to FIPV biotype has not yet been identified. This limits the possibility to diagnose FIP through identification of the mutated strain, since results of serology or PCR will be positive in cats infected by either biotype. Therefore, diagnosis must be based on several other clinical and laboratory findings which may either provide very specific results, or increase the diagnostic likelihood of FIP 1,6,8.

Signalment and clinical signs

Cats with FIP are usually young (especially < 2 years) and males seem to be more susceptible. However older animals (> 10 years of age) are sometimes affected, and cases in adult cats have recently increased, especially with the new FCoV 23 variant 8,10. There is often a history of a recent stressful event such as adoption or neutering 11. Individuals from multicat environment are at higher risk of developing FIP; although a large study noted that the majority of diseased cats were from one or two cat households, it was suggested that affected cats had been previously exposed to the virus 1,11. 

Clinical signs common to the two forms of the disease are lethargy, inappetence, weight loss/stunted growth, fever (waxing and waning, 39.5-40°C), lymphadenopathy and jaundice (Figure 1) 11,12. Effusive (wet) FIP is characterized by diffuse vasculitis and serositis, leading to the development of one or more cavitary effusions (abdominal, pleural, pericardial and rarely scrotal), with ascites and abdominal distension commonly reported (Figures 2 and 3) 13. Signs of non-effusive (dry) FIP depend on localization of the granulomatous lesions; these often affect the central nervous system (commonly manifesting with seizures, abnormal behavior, ataxia, nystagmus, hyperesthesia or sometimes paralysis and depression), the eyes (often with uveitis and/or chorioretinitis) (Figure 4) and/or abdominal organs such as the lymph nodes, kidney, liver, spleen and/or gastrointestinal tract 1,8. Occasionally, non-effusive FIP can be localized, with palpable large abdominal masses that can resemble a tumor; these can be caused by mesenteric lymph node enlargement or solitary intestinal lesions, especially of the colon or ileocecocolic junction (Figure 5) 13,14. 

Diagnosis of FIP 

Blood testing

Hematological changes indicative of an inflammatory process are commonly seen with FIP, but are nonspecific. The most frequent abnormalities are non-regenerative normocytic, normochromic anemia, lymphopenia, and neutrophilia with or without left shift. Microcytosis is also often observed, with or without anemia 11,13. 

Several biochemical abnormalities may be detected, which can be of some diagnostic accuracy. The protein profile usually shows hyperglobulinemia, with or without hyperproteinemia, low albumin concentration and low albumin-to-globulin ratio (A:G). FIP is considered very likely with a A:G < 0.4 and unlikely with a A:G > 0.8. Nevertheless, these values should be evaluated along with the clinical picture and other laboratory findings 8,13. 

Serum protein electrophoresis (SPE) will typically show decreased albumin, increased α2 fraction and a polyclonal gammopathy, although this latter can be less prominent if the SPE is performed early in the course of the disease (Figure 6) 15. 

Hyperbilirubinemia (in the absence of hemolysis, hepatic parenchymal alterations or cholestasis) is also commonly found, especially in the effusive form, and is probably a direct consequence of erythrocyte destruction within the lesions 8. Other biochemical alterations (e.g., increased liver enzyme levels) may be observed, depending on the localization and severity of the lesions 6. Most of the feline acute phase proteins (APPs), namely serum amyloid A, haptoglobin and α1-acid glycoprotein (AGP), increase greatly with FIP, but AGP is the most specific; in fact, marked increases in AGP support the diagnosis and can differentiate FIP from other inflammatory disorders 16,17. Nevertheless, again this has to be evaluated together with other alterations supportive of the disease.

As stated above, a positive result on serology will only demonstrate exposure to FCoV. Relatively higher antibody titers may be found in cats with FIP, but this can also occur in healthy cats in FCoV endemic catteries, due to continuous reinfection; at the same time a good percentage of FIP cats can be seronegative, due to antibodies forming complexes with circulating antigens 6. Similarly, viral RNA can be found in the blood of healthy animals. Additionally, the low viral load which is characteristic of FCoV viremia means that there is low analytical sensitivity on reverse transcription polymerase chain reaction (RT-PCR) testing. For these reasons, both serological and RT-PCR tests on blood should not be used as a diagnostic test for FIP 6,13.

Effusion tests 

FIP effusions have several peculiar features, and some tests will show optimal diagnostic accuracy, making sampling and analysis of effusions, when present, a necessary step. Macroscopically, the effusion is typically yellowish, sticky and can contain fibrin clots (Figure 7). Total protein content is high (> 3.5 g/dL) and cell count is typically low (< 5000 cell/µL), although this can be variable. Electrophoresis of the effusion will resemble that of SPE, and the A:G ratio is usually similarly low (< 0.4) 1,6,8. Cytology shows mostly non-degenerated neutrophils, macrophages and few lymphocytes on a granular eosinophilic proteinaceous background (Figure 8). Despite this non-specific pattern, cytology should always be performed to exclude the presence of a septic inflammatory process or neoplastic cells, which are other common causes of exudate that, if found, make FIP less likely 13.

Rivalta’s test is a cheap, point-of-care test performed by adding a drop of effusion to an acidic solution; a positive result is indicated if the effusion clots and retains its shape. This test has a high negative predictive value (i.e., a negative result makes FIP unlikely 18); however, a positive result alone cannot confirm FIP, since false positives are possible with other types of exudate (e.g., bacterial peritonitis, lymphoma). Again, using cytology together with this test will help identifying the disease 13,18,19. A similar test, the delta total nucleated cell count (DTNC), has been described utilizing a commercial hematology analyzer 20. Again, it relies on clumping of cells after addition of an acidic reagent, and involves using the analyzer to measure the leucocytes in two distinct channels. The DTNC – the ratio between the two obtained counts – is raised in FIP effusions, and has good diagnostic accuracy.

As with serum, the diagnostic accuracy of measuring antibody titers in effusions is low. FIP effusions can show negative serological results while having positive RT-PCR, sometimes even with an inverse correlation (i.e., negative serology despite a high viral load) 19. On the other hand, direct tests that demonstrate the antigen are very useful on effusions. RT-PCR shows good-to-very-good sensitivity and specificity, although false positives are sometimes observed 21. This can be secondary to circulating FCoV, even in small amounts, leaking from blood into the effusion due to inflammation of other origins. In general, a positive RT-PCR result coupled with cytological and biochemical alterations consistent with FIP is highly indicative of the disease 8.

Various immunocytochemistry (ICC) methods can be performed; these can demonstrate FCoV antigen inside macrophages via immunostaining, but the sensitivity is generally low to moderate, meaning that false negative results can easily occur; however, the specificity is high but not optimal, making it a good confirmatory test in the presence of other alterations supportive of the diagnosis 22. As with RT-PCR, false positive results can occur following viremia and leakage of viral RNA into the effusion of non-FIP cats, or from technical issues such as non-specific antibody binding 6,22.

Tests on other fluids

Neurological signs are more likely to occur in the non-effusive form of FIP, and cerebrospinal fluid (CSF) sampling may be appropriate in such cases. Clinicopathological alterations are non-specific but can be indicative of inflammation, such as increased protein content and mixed pleocytosis, usually pyogranulomatous. Nevertheless, cytology can also be unremarkable 8,19. Direct tests have proven utility, but with some limitations; as with effusions, ICC on CSF has shown high sensitivity, but the specificity is too low to make this a valuable confirmatory test 1,19.

RT-PCR has shown variably low to moderate sensitivity depending on the study, and a very good specificity (up to 100%), meaning that false positive results are very unlikely to occur. Additionally, the sensitivity of this test has been reported to increase drastically when only cats with neurological signs are evaluated, making RT-PCR on CSF a very useful test in these cases 8,19. Nevertheless, there is the rare chance that false positive results could result if the blood-brain barrier is damaged, causing leakage of circulating FCoV – demonstrating once again the importance of evaluating laboratory results together with other diagnostic methodologies supportive of the disease 19.

Aqueous humor (AH) can be collected in cats with ocular involvement (pyogranulomatous and granulomatous uveitis and/or chorioretinitis), with or without concomitant neurological signs 13. Clinicopathological changes have not been extensively described. Cytology can show neutrophilic, pyogranulomatous inflammation, and can be useful if neoplastic cells are present (e.g., lymphoma) 19. Total protein concentration is increased, especially in non-effusive FIP, and measurement can be useful (unpublished data).

Few studies have evaluated ICC for AH, and such studies show only moderate sensitivity and specificity, with false positive results recorded, so this option cannot be used as a confirmatory test. Further evaluation is needed to assess the diagnostic power of this test, since it could be a useful tool where other tests are not available (e.g., effusion analysis in non-effusive cases) 13. In addition, there are very few studies evaluating the use of RT-PCR on AH; reports show optimal specificity in the face of a very low sensitivity, making it a good tool to confirm the disease but not to exclude it, and again false positives are possible, when the blood-ocular barrier is damaged by other pathological processes (19; unpublished data).

Tissue testing

Currently, a definitive diagnosis of FIP is achieved by detection of typical histopathological changes in tissues, together with intralesional detection of FCoV using immunohistochemistry (IHC) (Figure 9) 19. Nevertheless, these tests have some limitations. Histological lesions indicative of FIP (i.e., pyogranuloma on serosal surfaces, granulomas, lymphoplasmacytic infiltrates, vasculitis) are not uniformly distributed, and can be missed on biopsy sampling. Viral antigen can also be variably distributed within lesions, thus multiple sections should be obtained in case of an unexpected negative IHC result. Moreover, the histological patterns typical of FIP can occasionally be observed with other conditions, making the use of IHC mandatory in these cases 23.

It is also noteworthy that laparoscopy or laparotomy to collect biopsy samples may be risky if the patient’s clinical condition is poor. Given this limitation, a few studies have evaluated the diagnostic accuracy of less invasive approaches. Cytological examination of affected organs has not been extensively studied, and specific changes have not been reported 8,24. ICC on fine-needle aspirates (FNAs) of liver and kidneys have low sensitivity, although a recent study evaluating FNAs on mesenteric lymph nodes showed reasonable sensitivity but a non-optimal specificity, with a false positive result probably related to the lymph nodes acting as a site of persistence of FCoV in non-FIP cats. Therefore, this test cannot exclude FIP, but can be a useful confirmatory test in conjunction with other consistent alterations 19,25.

When comparing results from IHC and RT-PCR on tissue biopsies, the former is more accurate, as RT-PCR has a lower specificity due to the well-recognized presence of systemic FCoV in non-FIP cats 8,23. On the other hand, RT-PCR on FNAs obtained from mesenteric nodes demonstrated a sensitivity and specificity of 90% and 96% respectively, with false negatives recorded in neurologic cases only. Therefore, this method can be a useful non-invasive addition to the diagnostic panel for FIP 24.

Detection of Spike gene mutations 

Although extensively studied for years, none of the mutations identified in sequences isolated from FIP cats have been shown to be specific for the disease 8. Mutations of the Spike (S) gene, which is responsible for host-receptor recognition and viral-cell-membrane fusion, have been evaluated in several studies, but the diagnostic power of this test is limited and is affected by the sequencing technique used. For example, a high number of false negative results will be recorded when some techniques (e.g., allelic discrimination) are used, since high viral loads are needed to obtain a result, and a failed sequencing is recorded as a negative result 6. On the other hands, when pyrosequencing is used, the specificity of the test is either low or no better than other techniques 21. Therefore, despite being likely involved in the pathogenesis of FIP, the variable detection of S gene mutations suggests that multiple mutations are probably involved, and its identification adds little or no information compared to conventional RT-PCR 8.

Conclusion

Feline infectious peritonitis is a worldwide disease, yet definitive diagnosis of the condition is elusive; antibody titers demonstrate only that a cat has been exposed to FCoV, and some FIP-affected cats can actually have negative titers. Hematology and biochemistry results can be helpful but not pathognomic for the disease, and testing on effusates can offer both false positive and false negative results. A combined approach using advanced diagnostic technologies is, therefore, recommended, and the importance of evaluating diverse laboratory results to reach a reasoned conclusion cannot be over-emphasized.