Article Text

Download PDFPDF

Update on feline infectious peritonitis
  1. Emi Barker and
  2. Séverine Tasker


Background: Feline coronavirus (FCoV) infection in cats is common, usually only causing mild intestinal signs, such as diarrhoea. It is highly infectious and found worldwide. A sequela of FCoV infection, feline infectious peritonitis (FIP), is a common cause of death in young cats, occurring in up to 10 per cent of cats infected with FCoV. Although suspicion of FIP is frequent in sick, particularly young, cats, obtaining a definitive diagnosis using non- or minimally-invasive approaches is difficult.

Aim of the article: This article provides an update regarding diagnosing cases of FIP and guidance on current treatment recommendations.

View Full Text

Statistics from

Embedded Image

Emi Barker qualified from the University of Bristol in 2003. She was awarded a PhD from the same university in June 2011 for research into haemotropic mycoplasmas. She became a diplomate of the European College of Veterinary Medicine in 2016. She is currently a referral clinician in small animal medicine at Langford Vets, Bristol.

Embedded Image

Séverine Tasker qualified from the University of Bristol in 1994. She is an RCVS-recognised Specialist in feline medicine and a European Veterinary Specialist in small animal internal medicine. She is honorary professor of feline medicine at Bristol Veterinary School and currently chief medical officer for the Linnaeus Group, Solihull.

Key learning outcomes

After reading this article, you should:

  • Understand the relationship between feline coronavirus and feline infectious peritonitis (FIP), and how this impacts transmission and diagnosis;

  • Be able to list commonly identified findings on history, physical examination and routine clinicopathological tests that raise concern for FIP in a sick cat;

  • Know how to diagnose probable FIP in a suspected case, and how to definitively diagnose FIP when this is necessary;

  • Be aware of the current treatment options for a cat diagnosed with FIP, including recent advances in this field;

  • Be able to discuss possible methods to reduce the risk of FIP within a multicat household.


Coronaviruses are relatively large, enveloped, positive-sense, single-stranded RNA viruses (Figs 1, 2). They exhibit a high rate of mutation during replication and therefore exist as clusters of genetically diverse populations. Cats have been found to be infected with feline coronavirus (FCoV) worldwide, with the exception of cats on a small number of isolated islands.

Fig 1: Schematic diagram of the feline coronavirus genome (approximately 29,250 nucleotides in length excluding the polyA tail at the 3’ end) with component genes and nucleotide scale. Note that over two thirds of the genome comprises the gene that encodes the non-structural polyprotein 1
Fig 2: Schematic drawing of a feline coronavirus virion with relative position of structural proteins and genomic single-stranded RNA (ssRNA) indicated. The spike glycoprotein trimers project from the surface of the virus, resulting in the ‘crown’-like appearance when viewed under transmission electron microscopy

Two serotypes of FCoV are recognised: Type 1, which represents the vast majority of field strains, and Type 2. The latter arises following recombination events between Type 1 FCoV and canine coronavirus. The two serotypes are distinguished primarily by differences in their transmembrane spike (S) glycoprotein. The S glycoprotein (Fig 2) mediates binding to and entry into host cells.

Infection with FCoV is very common – 35 per cent of the owned domestic cat population have detectable antibodies to FCoV, indicating exposure (combined data from the eight serological studies listed in Drechsler and others 2011). In single-cat households (combined data), seroprevalence reduces to 21 per cent, but correspondingly in multicat households it can be over 90 per cent (Addie and others 2000). Most infections are transient (although reinfection is common) with only a small percentage becoming persistent ‘carriers’ or ‘chronically shedding’ cats (Kipar and Meli 2014).

Transmission and pathogenesis

Transmission is primarily faeco-oral, with litter boxes representing the principal source of infection among cats within a household. In breeding catteries, kittens commonly become infected at a young age, mostly at five- to six-weeks-old (Addie and Jarrett 1992), as maternally derived antibodies have started to wane. Nose-to-nose contact is considered an uncommon route, and transplacental transmission is considered rare. Experimentally, infection has been transmitted by parenteral injection of virus derived from cats with feline infectious peritonitis (FIP).

Small intestinal villi enterocytes are the primary point of host cell entry and replication. In most cases, FCoV infection is subclinical or results in only mild gastrointestinal signs (eg, diarrhoea, vomiting). However, occasionally more severe gastrointestinal disease is seen. Subclinical FCoV infection was previously believed to be confined to the intestinal tract, but we now know that healthy FCoV-infected cats develop a detectable low-level viraemia during acute infection (Kipar and others 2010). In a small percentage of cases FCoV infection results in FIP, which typically occurs sporadically. Occasional outbreaks of FIP in multicat households or shelters affect a larger percentage of cats (Barker and others 2013).

In FIP, virus-laden monocytes attach to the walls of small veins and release inflammatory cytokines that damage the endothelial basal lamina (Kipar and Meli 2014). This results in extravasation of monocytes (which mature into tissue macrophages) and proteinaceous fluid. In effusive (ie, ‘wet’) FIP, this extravasation of proteinaceous fluid is evident as fluid accumulations within body cavities. In non-effusive (ie, ‘dry’) FIP, the extravasated macrophages recruit other inflammatory cells and result in perivascular granulomata, which may appear grossly as a mass lesion (Figs 3, 4a). The role of other cytokines, including interferon (IFN)-γ and tumour necrosis factor (TNF)-α, in the pathogenesis of FIP is not completely understood, but is thought to be significant (Kipar and Meli 2014). The mesenteric lymph nodes (MLNs) may represent an important site in which the host immune response to FCoV plays a role in the outcome of infection, as MLNs are presumed to be the first site of FCoV replication outside the intestinal tract and before monocyte/macrophage infection occurs (Malbon and others 2019).

Fig 3: Numerous granulomas (black arrows) within a sectioned kidney from a two-year-old male neutered cat with feline infectious peritonitis. The cat was presented with an acute history of lethargy and inappetence. Physical examination revealed a thin body condition, pale and icteric mucous membranes, subtle unilateral anterior uveitis and bilateral renomegaly. Blood analysis documented severe anaemia (haematocrit 7.8%; reference interval [RI] 27–47%), hyperglobulinaemia (83 g/l; RI 21–51 g/l) and jaundice (31 µmol/l; RI <10 µmol/l). On postmortem examination a moderate volume of ascites was present, alongside a small volume of pleural and pericardial effusions. Both kidneys were enlarged with margins distorted by vascularised mass lesions. Pyogranulomatous lesions were found on histopathology of samples collected throughout the body (including the iris), with presence of feline coronavirus confirmed with RT-PCR and immunohistopathology
Fig 4: Ultrasonographic images of two cats with feline infectious peritonitis. (a) This cat had a large mid-abdominal mass on examination, which was identified as being a mesenteric lymph node using ultrasound (yellow calliper). (b) This cat also had enlarged abdominal lymph nodes (yellow calliper); however, a small volume of ascites was also visible (white arrow). Ultrasound enabled guided sampling of both the enlarged lymph nodes and the ascites

Viral factors are important in the pathogenesis of FIP. As mentioned earlier, the S glycoprotein of FCoV mediates host cell entry, with mutations in the S gene influencing cell tropism (Kipar and Meli 2014). Mutations at different sites within the S gene have been detected with increased frequency in FIP tissue-derived FCoVs, as compared to faecally shed FCoV from clinically ‘healthy’ cats (Chang and others 2012, Licitra and others 2013). This has led to suggestions that some of these mutations could be a useful target in differentiating cats with FIP from cats without. Unfortunately, a recent large-scale study suggested that one of these sets of mutations, involving the fusion peptide, was more indicative of systemic FCoV infection, occurring in FCoV viraemic cats with and without FIP with equal frequency, rather than FIP per se (Barker and others 2017). Other viral factors mediating effective and sustained replication in monocytes, and activation of infected monocytes, are also likely to be important for the development of FIP following systemic FCoV infection. Very recently there has also been suggestion that specific viral mutations could be associated with tissue tropism (Andre and others 2019).

Host factors contributing to the immune response, such as genetic background (eg, breed-, line- or individual-specific genetics) and maturity (eg, age, history of prior exposure to infectious agents), likely play an important role in FIP development. Host factors are inextricably linked with environmental factors, such as stress (eg, cat-to-cat interactions, novel experiences such as rehoming, vaccination, surgery, or resource accessibility) and overcrowding, which themselves may lead to increased environmental viral burden, increased viral replication within cats and support FIP development.

Clinical signs associated with feline infectious peritonitis

The variability in the extent and distribution of both vasculitis and perivascular granulomata underlies one of the difficulties in diagnosing FIP. Clinical signs of FIP can change over time, warranting repeated clinical examinations to detect newly apparent pathology. Non-specific clinical signs (Table 1) – often waxing and waning – are attributable to the systemic inflammatory response and frequently occur in cats both with and without detectable effusions, with FIP a significant differential for pyrexia of unknown origin (Spencer and others 2017). However, it should be noted that the absence of these signs does not rule out FIP.

Table 1: Commonly encountered features of signalment, history and physical examination seen in cats with feline infectious peritonitis

Although effusive FIP is regarded as being three to four times more common than non-effusive FIP (Kipar and Meli 2014, Riemer and others 2016), and the distinction between the two forms is important for diagnostic purposes, there is considerable overlap between them. Cases with effusive FIP often have pyogranulomatous lesions visible at postmortem examination, while many cats with non-effusive FIP go on to develop effusions. Effusive FIP is often acute in nature, progressing within a few days or weeks, whereas non-effusive FIP tends to be more chronic, progressing over a few weeks to months. In effusive FIP, effusions may form in one or more body cavity, with abdominal effusion leading to a clinical presentation of ascites and abdominal distension being the most common manifestation. Cats with pleural effusion often develop dyspnoea, whereas cats with pericardial effusion rarely show signs of cardiac tamponade. Occurring only rarely, scrotal effusion leads to scrotal enlargement in entire males. Non-effusive FIP is often more difficult to diagnose, particularly in the earlier stages of disease, as vague non-specific clinical signs may be all that can be seen. More specific signs depend on the organs affected by the granulomatous lesions, often the central nervous system (CNS), eyes, or abdominal organs (eg, liver, MLNs, kidney, gastrointestinal tract); however, any tissue can be affected and primary involvement of the lungs or skin have been described.

In sick cats, careful neurological and ocular examination may reveal changes that support a diagnosis of FIP, as well as indicating a potential source of samples for testing. Neurological signs associated with focal, multifocal or diffuse changes in the CNS may be seen in up to 30 per cent of cats with FIP, and for some these are the only signs noted (Fig 5); this makes FIP a common differential for neurological disease, particularly in young cats. Commonly reported signs in FIP with neurological involvement include:

Fig 5: (a) Six-month-old male entire ragdoll, presented with a 48-hour history of altered behaviour and a 24-hour history of hindlimb paresis, reduced appetite, nystagmus and mild head tilt. Neurolocalisation was most consistent with central vestibular syndrome, most likely a result of cerebellar disease. Routine haematology and serum biochemistry were unremarkable. (b) MRI (T2-weighted midline sagittal view) revealed severe hydrocephalus with dilation of the entire ventricular system of the brain and secondary herniation of the cerebellum, the latter likely accounting for the majority of clinical signs. Ultimately, the presence of feline coronavirus was confirmed with RT-PCR of cerebrospinal fluid, alongside histopathology and immunostaining of meningeal tissue samples, confirming a diagnosis of feline infectious peritonitis
  • Ataxia (with varying degrees of tetra- or paraparesis);

  • Hyperaesthesia;

  • Head tilt;

  • Nystagmus;

  • Seizures;

  • Behavioural change;

  • Mental state change;

  • Cranial nerve deficits; and

  • Postural reaction deficits.

However, differentiating subtle neurological signs from those exhibited by systemically unwell cats may not be possible. Similarly, FIP is a major differential for uveitis (Fig 6) with anterior and posterior uveitis commonly identified in cats with both effusive and non-effusive FIP, particularly when examined by an experienced clinician.

Fig 6: Feline infectious peritonitis is a major differential for uveitis, manifestations of which include: changes in iris colour, thickness and texture; dyscoria (abnormal shape of the pupil); anisocoria (unequal pupil sizes); sudden loss of vision; hyphema; keratic precipitates (‘mutton fat’ deposits on the ventral corneal endothelium); chorioretinitis; retinal detachment; and aqueous/vitreous flare. These changes may be subtle and unilateral or bilateral and severe. (a) Mild iridial changes, aqueous flare and keratic precipitates (white arrow) of the right eye in a eight-month-old chinchilla persian with pyrexia of unknown origin. (b) Bilateral severe fibrinous, flocculent, aqueous flare limiting examination of the interior of the eye, where assessment of the retina is possible changes may be present there too. (c) Severe bilateral chorioretinitis including haemorrhage and granulomata (pink and red arrows); aqueous flare and retinal oedema results in the image appearing to be out of focus. Pictures: (a) Caroline Smith, University of Bristol, Langford Vets, (b and c) Vim Kumaratunga, University of Bristol, Langford Vets

Where FIP manifests in the intestinal tract and/or regional lymph nodes (sometimes called ‘focal FIP’, although the disease is still systemic) it can present as a palpable abdominal mass (Fig 4) that must be differentiated from neoplasia, toxoplasmosis or other granulomatous disease (eg, mycobacterial infection) (Kipar and others 1999, Pedersen 2009). Where the lesion involves the intestinal wall, clinical signs may include vomiting, diarrhoea or constipation, or signs referable to an obstructive or protein-losing enteropathy may be seen.


A diagnostic flow diagram is shown in Fig 7.

Fig 7: Flow diagram showing the diagnostic approach to cats with suspected feline infectious peritonitis. CBC Complete blood count, CSF Cerebrospinal fluid, FCoV Feline coronavirus, FIP Feline infectious peritonitis, FNA Fine-needle aspiration

A high index of suspicion can be obtained from a combination of signalment, history, and physical examination (Table 1). However, none are pathognomonic for FIP and other common differential diagnoses (Table 2) should be considered when performing further investigation.

Table 2: Common differential diagnoses for feline infectious peritonitis

Blood tests

Routine clinicopathological tests (Table 3) may indicate the presence of a chronic systemic inflammatory response, further supporting a clinical suspicion of FIP; however, no clinicopathological changes are diagnostic for FIP and, in some cats, routine blood analysis can be unremarkable. This is compounded by both vets (and owners) suspecting FIP earlier in the course of the disease process, so reducing the negative predictive power of some findings (eg, absence of hyperglobulinaemia) (Stranieri and others 2017).

Table 3: Commonly encountered changes on routine clinicopathological analysis seen in cats with feline infectious peritonitis

Further support for FIP may be gained from the measurement of inflammatory markers. Serum protein electrophoresis is a crude way of determining the presence and nature of an inflammatory response, particularly where there is a hyperglobulinaemia. The most frequently encountered change in cats with FIP is a polyclonal gammopathy, indicating a non-clonal increase in antibodies; however, a small number of cats present with a monoclonal gammopathy (Taylor and others 2010), while others show increases in the α2-globulin fraction (reflecting an increase in acute-phase proteins [APPs]) (Stranieri and others 2017). APPs are made in the liver in response to cytokines released from activated macrophages and monocytes. Marked increases (> 1.5 mg/ml) in serum α1-acid glycoprotein (AGP) can support a diagnosis of FIP (Duthie and others 1997, Paltrinieri and others 2007, Hazuchova and others 2017). Other APPs, serum amyloid A and haptoglobin, have been assessed in the diagnosis of FIP but were both less sensitive and specific than AGP (Duthie and others 1997, Hazuchova and others 2017). Overall, increased AGP (or other APPs) in serum, despite supporting a diagnosis of FIP, is not confirmatory and may be limited by cost, availability and turnaround time.

Clinicians vary as to whether they perform FCoV serology or not in suspected cases. Although a positive result indicates exposure to FCoV, many clinically healthy cats have positive, often high, antibody titres, while a small proportion of cats with both effusive and non-effusive FIP are seronegative. Diagnosis of FIP should never be made based upon positive serology alone. Faecal RT-qPCR has replaced serology in monitoring the effect of control measures in the management of FCoV infection within a breeding cattery.


Imaging (Figs 4, 5) can be useful, in that it can often identify areas of pathology (eg, mass lesion, effusion) that may prove useful to sample as well as guiding sample acquisition (eg, ultrasound-guided needle biopsy). However, imaging alone cannot be used to make a diagnosis of FIP.

To provide a definitive diagnosis of FIP, cytological or histopathological changes consistent with FIP (ie, pyogranulomatous inflammation) should be identified and subsequently co-localised with FCoV antigen, using immunostaining for viral antigen. More recently RT-PCRs have also been used to support a diagnosis of FIP (Box 1).

Box 1:

use and abuse of RT-PCR in the diagnosis of Feline infectious peritonitis

Feline coronavirus (FCoV) RT-PCR, when designed appropriately, is a very sensitive and specific assay for the detection of FCoV within samples, and is generally more sensitive than immunostaining for FCoV antigen in tissues (Barker and others 2017). However, it cannot co-localise virus to cytological/histological lesions, merely to the samples in which those changes are present. Following intestinal infection with FCoV, most cats develop a viraemic, disseminating virus throughout the body. Therefore, cats without feline infectious peritonitis (FIP) can have detectable virus within blood, effusions and tissues – albeit at a lower frequency and viral copy number. Due to low circulating levels of viraemia, use of RT-PCR of whole blood in cats with suspected FIP is not recommended (Emmler and others 2020).

In a recent large study, all but one cat of 57 (98%) with FIP had at least one tissue positive for FCoV by quantitative RT-PCR, as compared to 12 of 45 cats without FIP (Barker and others 2017); viral copy numbers were also significantly higher in the positive samples from cats with FIP than those without FIP. Further investigation of the single cat with FIP and a negative RT-PCR result revealed the FCoV present to have multiple mutations in the sequence normally detected by the RT-PCR assay, resulting in its failure. None of the cats without FIP (including those with a positive RT-PCR result) had histopathological evidence of granulomatous disease or positive immunostaining.

RT-PCR has been applied to cytological samples. Most (72–100%) effusions from cats with FIP are RT-PCR positive, compared to only two false-positives out of 76 samples from cats without FIP across three studies (Barker and others 2017, Felten and others 2017c, Stranieri and others 2018). Most (18 of 20; 90%) mesenteric lymph node aspirates from cats with non-effusive FIP are RT-PCR positive (Dunbar and others 2018); however, one false-positive result (out of 20 cats) did occur in a cat seropositive for FCoV. Detection of FCoV in cerebrospinal fluid (CSF) by RT-PCR from cats with FIP is variable, ranging from 21% to 86% (Foley and others 1998, Doenges and others 2016, Barker and others 2017, Emmler and others 2020), whereas RT-PCR was negative in all control cats. Detection of FCoV in aqueous humour from cats with FIP was poor (25%), and no control cats were tested (Emmler and others 2020).

Additional analysis has been applied to RT-PCR-positive samples to determine whether the FCoV present carries genetic mutations that have been said to be associated with FIP. The use of Spike gene mutation analysis has been most frequently studied for this purpose, albeit using different techniques, different sample types and with different conclusions. Where a highly sensitive method (pyrosequencing) was employed to evaluate the Spike gene, mutations were detected in FCoV-positive tissue from 15 of 17 (88%) samples from cats without FIP as compared to 202 of 206 (98%) samples from cats with FIP (Barker and others 2017). Other techniques (eg, allelic discrimination) that require a relatively high viral copy number in the sample to generate a result (often not present in cats without FIP), and consider a result where sequencing has failed to be negative, will increase the test specificity by a modest amount by reducing, but not eliminating, the number of false-positives; however, the detection of true-positives results in cats with FIP (ie, the test sensitivity) is more markedly reduced (Emmler and others 2020, Felten and others 2017a).

In conclusion, although a positive RT-PCR result on fluid, effusions, aspirates and tissue can provide strong support for a diagnosis of FIP (particularly for CSF), both false positives and false negatives occur such that RT-PCR should not be solely relied upon to make a diagnosis. Furthermore, Spike gene analysis is either of little benefit over RT-PCR at removing false-positives (ie, when pyrosequencing is used), or markedly increases the number of false-negatives (ie, when allelic discrimination is used) and may inadvertently cast doubt on a diagnosis of FIP in a cat with FIP, potentially delaying treatment.

Note that RT-PCR of faeces is not a test for FIP, it is a test for FCoV shedding (which can be intermittent). Most cats that have a positive faecal FCoV result will not go on to develop FIP, and only two in every three cats with FIP are shedding FCoV at time of euthanasia (Barker and others 2017). It is only of use in special circumstances (eg, attempting to identify shedders within a multicat household).

Analysis of effusions

In effusive FIP, sampling the effusion is the single most useful diagnostic step in confirming a diagnosis. For this reason, where effusions are not evident on initial evaluation, repeated ultrasonography to identify any small volume effusion is recommended (Fig 4) and may facilitate sampling of small pockets of fluid. FIP effusions (Fig 8) are usually clear, poorly cellular (total nucleated cell count <5 x 109/l), yellow, viscous, protein-rich (with a total protein concentration of >35 g/l), have a low albumin to globulin ratio, and have a positive Rivalta test (Box 2). However, in some cats the effusions might be cloudy, of slightly lower protein levels (eg, in cats that were not originally markedly hyperproteinaemic, or following repeated abdominocentesis), or contain much higher cell counts (up to 20 x 109/l). Cytological examination usually reveals pyogranulomatous inflammation with macrophages, non-degenerate neutrophils and few lymphocytes. Effusion AGP concentrations may also be useful in supporting a diagnosis of FIP, potentially affording greater sensitivity and specificity than serum measurements (Duthie and others 1997, Hazuchova and others 2017), but are not confirmatory. Positive FCoV antigen immunostaining is strongly supportive of a diagnosis of FIP. However, false-negatives occur in 5 to 43 per cent of cats with FIP (Hartmann and others 2003, Paltrinieri and others 1999), particularly in low cellularity samples, and false-positives have been reported in up to 30 per cent of cases (Hartmann and others 2003, Litster and others 2013, Felten and others 2017b), including cats with neoplasia or cardiac disease. False-positive results may be dependent on technique, methodology or laboratory used; therefore, checking the specificity and use of internal controls within the laboratory used is recommended when interpreting results.

Fig 8: Although effusions from cats with feline infectious peritonitis are typically clear, reflecting a low cellularity (< 5 x 109/l total nucleated cell count), viscous with a tendency to froth when agitated, reflecting a high concentration of predominantly inflammatory proteins (> 35 g/l), and with a slightly yellow tinge, reflecting patient jaundice, their gross appearance can be variable
Box 2:

Rivalta test

The Rivalta test is a simple, inexpensive, point-of-care test to differentiate a transudate from an inflammatory effusion. It does not replace more advanced analysis in-house or at external laboratories, but it can facilitate decision-making, particularly in financially constrained settings and where tests to quantify effusion protein concentrations are not available.

The Rivalta test is more useful at ruling out feline infectious peritonitis than ruling it in: over 90% of FIP effusions are Rivalta test positive (Hartmann and others 2003, Fischer and others 2012); positive results are also seen with bacterial peritonitis and neoplastic exudates; and low protein, non-inflammatory effusion (such as those seen with cardiac failure or hypoproteinaemia) are typically negative.

Method: Mix 10 ml distilled water with 2 to 3 drops of white vinegar in a test tube or universal container; one drop of effusion is added carefully to the top

Negative result = dispersion of the drop of effusion

Positive result = the drop of effusion retains its shape and floats slowly to the bottom of the tube or sits on the surface of the water

Analysis in the trickier cases

For cats with suspected non-effusive FIP and accessible mass lesions (eg, mesenteric lymphadenomegaly) fine-needle cytology may be considered. While in cats where CNS (Fig 5) or ocular signs (Fig 6) predominate, more specialist techniques to obtain samples of cerebrospinal fluid (CSF) or aqueous humour for analysis are discussed in the literature but rarely performed in first-opinion practice. Cytology typically reveals non-septic pyogranulomatous to granulomatous inflammation; however, this is only documented in 42 to 82 per cent of cats with FIP, and up to 30 per cent of samples from cats without FIP (Giordano and others 2005, Gruendl and others 2017, Felten and others 2018). Positive immunostaining for FCoV antigen can provide further support for FIP. However, as with cytological analysis alone, false-negatives occur in more than 15 per cent of CSF samples (Gruendl and others 2017), more than 35 per cent of aqueous humour samples (Felten and others 2018), and 11 to 53 per cent of tissue aspirates (Giordano and others 2005, Felten and others 2019) from cats with FIP, with false-positives reported in around 20 per cent of samples from cats without FIP, including those with neoplasia or vascular disease.

Until recently, histopathology alone (ie, in the absence of immunostaining) was the reference standard for the diagnosis of FIP; however, histopathology can be equivocal or misleading in some cases (Giordano and others 2005, Pedersen 2009, Giuliano and others 2020), particularly where needle-core samples are collected blind. Many now consider the demonstration of FCoV antigen within granuloma-associated macrophages by immunostaining as the reference standard, but it is subject to the similar limitations to histology, albeit with 100 per cent specificity. In a recent large study, only 62 per cent of tissue samples from cats with FIP revealed FCoV-positive lesions (Barker and others 2017). However, it should be noted that all the samples in this study were collected postmortem, visibly normal tissues were frequently sampled in addition to grossly abnormal tissues, and at least one tissue sample per cat was diagnostic for FIP. Wherever possible, grossly abnormal tissue should be sampled to maximise the likelihood of achieving a diagnosis.

Often, and especially in non-effusive FIP, collection of biopsies from tissues with gross lesions is necessary to achieve a definitive diagnosis. In the absence of a definitive diagnosis, or pending confirmatory tests, available results form the basis of discussion as to whether further, invasive investigation is likely to change treatment options and whether to start treatment. This can been frustrated by the geographical restriction (outside the UK) of some tests (eg, AGP, immunocytochemistry, immunohistochemistry, and RT-PCR) that would otherwise be strongly supportive of a diagnosis of FIP. If euthanasia is performed without a definitive diagnosis, postmortem examination is strongly recommended to assess whether gross findings (with histopathology if funds allow) are consistent with a diagnosis of FIP.

Treatment and prognosis

Potential alternative diagnoses, such as toxoplasmosis and mycobacterial infection, should be ruled out and a definitive diagnosis of FIP made before considering treatment. However, the reality is that treatment is often started when as close to a definitive diagnosis of FIP as possible has been achieved, taking into account the overall clinical picture alongside owner preferences and finances. A lack of definitive diagnosis makes it impossible to know whether a treatment response indicates efficacy against FIP, or a missed alternative diagnosis. Treatments administered may also interfere with the sensitivity and specificity of future diagnostic test results. A paucity of placebo- or ‘current best-treatment’-controlled clinical trials of cats with definitively confirmed FIP limits treatment recommendations. Currently, no licensed drug is available that has proved effective in curing FIP.

Prognosis for cats with effusive disease is grave, with death or euthanasia within days to occasionally weeks in most cases. The prognosis for cats with non-effusive disease is also poor, with death or euthanasia within weeks to months in most cases. However, it is not necessary to euthanase immediately if the cat still has a reasonable quality of life. It is possible to maintain palliative treatment for as long as weight and activity are maintained. Rarely, some individuals have survived for months to sometimes years, often with supportive treatment, but it is unclear as to whether the treatment administered influenced survival.

Supportive care

Treatment is currently limited to supportive care. Cats, once anorexic, can quickly become dehydrated; therefore, simple fluid therapy, correction of electrolyte disturbances, and encouraging them to eat can be extremely useful at improving their quality of life. The value of removing fluid effusions in cats with FIP has been debated. Thoracocentesis is indicated where effusion has resulted in dyspnoea. Abdominocentesis is controversial and may be detrimental due to exacerbation of dehydration, although some authors have described fluid drainage followed by intracavitary corticosteroid administration.

Immune-mediated treatment

Given that FIP has a significant immune-mediated component, treatment to either suppress or modify the immune response can be considered. Corticosteroids are the most frequently used medication, and some cats receive benefit from them, particularly in terms of quality of life. However, there are no controlled studies to prove beneficial effect, only anecdotal reports, and they appear to do very little for the viral infection itself. Doses are empirical (prednisolone 2 to 4 mg/kg/day orally) and can be tapered slowly to response.

Medical treatment

Many different drugs have been considered for the treatment of FIP. Cyclophosphamide, ciclosporin A and anti-TNF-α antibodies have been anecdotally used to prolong survival, but no controlled studies have been performed. Pentoxifylline has also been anecdotally used to manage the vasculitis; however, in a placebo-controlled trial of the related drug, propentofylline, no benefit was found (Fischer and others 2011). Interferons are also commonly used, on the basis of positive anecdotal reports; however, a placebo-controlled clinical trial failed to demonstrate a clinically relevant benefit (Ritz and others 2007). Polyprenyl immunostimulant has limited data to support its use and also has significant limitations (including lack of control treatment group and limited diagnostic criteria for FIP) (Legendre and others 2017); however, it is possible that it may improve survival times in the milder forms of non-effusive FIP without detrimental impact on the patient. Herbal medication has also been suggested for cats with FIP, often with no scientific data to support its use.

Future treatments

Recently described promising new, but as yet unlicensed, drugs comprise viral protease inhibitors and nucleoside analogs. FCoVs produce large viral proteins (eg, the gene encoding polyprotein 1 forms a large component of the FCoV genome; Fig 1) that are cleaved into smaller functioning units by proteases. Therefore, inhibitors of these proteases affect viral production. The protease inhibitor GC376 has produced remarkable responses in both experimentally induced and naturally occurring FIP, with six of eight cats with experimentally induced FIP alive at eight months and 19 of 20 cats with naturally occurring FIP showing a positive response, which was sustained in seven cats (Kim and others 2015, Pedersen and others 2018). The nucleoside analog GS-441524 acts as an alternative substrate and RNA-chain terminator of the viral RNA polymerase, thus interfering with FCoV replication. It too has produced remarkable responses in both experimentally induced and naturally occurring FIP, with all 10 cats with experimentally induced FIP alive at eight months and 26 of 31 cats with naturally occurring FIP having a positive response, which was sustained in 25 cats (Murphy and others 2018, Pedersen and others 2019). Unfortunately, both the protease inhibitor GC376 and the nucleoside analog GS-441524 appear to poorly penetrate the blood-brain and blood-eye barriers, likely accounting for increased likelihood of relapses involving the nervous system or lack of initial response to treatment in study cats presenting with neurological or ocular signs of FIP. The use of higher than previously reported doses of these agents, along with extended courses, have been suggested for cases of neurological or ocular FIP; however, more studies are warranted.

We are aware that, in the absence of commercially available licensed products, some UK cat owners have obtained black-market forms of both GS-441524 and GC376 via the internet for the treatment of FIP in their pet. By their nature, these black-market products are of unknown quality, efficacy, toxicity and longevity, and therefore cannot be prescribed by veterinary surgeons for their patients.


A ‘nutritional supplement’ (Mutian; Nantong Mutian Biotechnology) containing a novel adenosine nucleoside analogue (Mutian Xraphconn [Nantong Mutian Biotechnology]; reported to be different to GS-441524) has been marketed worldwide, primarily at cat owners, for the treatment of FIP (Addie and others 2020a). However, there is only limited published research describing its use to stop faecal shedding of virus (Addie and others 2020a). There is currently no peer-reviewed evidence base upon which to recommend its use in cats with FIP. Furthermore, according to both the Veterinary Medicines Directorate (UK) and the Food and Drug Administration (USA), nutritional supplements may not be presented with medicinal claims (eg, ‘the ability to cure cats of FIP’), otherwise they would be considered as a veterinary medication requiring authorisation.

Prevention and in-contact cats

One of the most frequent questions from owners following the diagnosis of FIP in one of their cats is what do with the other cats in the household. For the major considerations see Box 3. As spread of FCoV is most of a concern among large groups, reduction in environmental viral load through improved hygiene is key. Appropriate care in cleaning and use of disinfectants (see Addie and others 2015 for more information) to reduce environmental, including fomite, contamination is necessary. This includes maintenance of toileting facilities including:

Box 3:

When one cat gets Feline infectious peritonitis, what do you do with the other cat(s) in the household?

  • Where more than one cat is being considered (ie, the one with feline infectious peritonitis (FIP) and any in-contact cats), an accurate diagnosis becomes more important in order to guide advice.

  • In-contact cats will be at slightly increased risk compared to the general population, particularly if they are direct siblings (estimated 2x risk), due to shared viral, environmental and possibly genetic factors – but none of these can be changed at the time of diagnosis

  • Although cats can pass feline coronavirus (FCoV) that causes intestinal infection between each other, they are not thought to be able to horizontally pass the mutated FCoV that directly causes FIP between themselves (ie, the mutation is a spontaneous event that happens within individuals). Given that the cat with FIP would have had historical intestinal infection with FCoV, it is likely that the other cat(s) in the household would have been infected historically too.

  • Removal or isolation of a cat with (suspected/proven) FIP from a household is not indicated, and would likely negatively impact on the sick cat.

  • As the household environment (including fomites) is likely contaminated with FCoV (particularly if there are five or more cats in the household, to maintain continuous infection), improved hygiene (particularly regarding the litter tray) is strongly recommended.

  • As stress is associated with the development of FIP, reducing household stress (eg, due to conflict, overcrowding, or continued breeding) is recommended, as is deferral of non-essential, elective procedures (eg, microchipping, neutering).

  • As FCoV can survive under appropriate conditions for up to seven weeks in the environment and the loss of a house-mate will be stressful for the remaining cats (and therefore may temporarily induce FCoV shedding in carriers), immediate ‘replacement’ of the deceased cat is strongly discouraged (for at least three months). These replacements may be naive to the FCoV isolate circulating in the household, are typically young (ie, in the highest risk category for going on to develop FIP following exposure to the virus), may share some genetic risk factors (ie, if from the same source as the deceased cat), and may well cause stress and conflict within the household (ie, owners often have the misconception that the remaining cat(s) need the company of another cat).

  • Faecal shedding by remaining cats (eg, by weekly faecal PCRs on three to four occasions) could be considered before the introduction of a new cats, but this would not completely eliminate risk (as shedding is intermittent), and a significant number of cats are infected with FCoV from their original household such that the incoming cat may have already been exposed to FCoV).

  • Sufficient litter-trays per cat;

  • Siting litter-trays away from food and water; and

  • Use of cat litter that may have enhanced neutralisation of FCoV to limit fomite spread (eg, dust-free clumping Fuller’s earth litter) (Addie and others 2020b).

Although achievement of a FCoV-free household or kittens (eg, via early weaning) is technically possible, it is not without significant cost and potential welfare concerns – more information can be found in the latest European Advisory Board on Cat Diseases (ABCD) guidelines on FIP (

Minimising host risk factors, particularly relating to stress of conflict or overcrowding, is also recommended. Furthermore, in breeding situations where particular sire and queen combinations have resulted in cases of FIP across multiple litters, retiring one or both cats from the breeding programme should be considered in case they are conferring increased genetic risk of FIP.

The FIP vaccine (Felocell FIP; Zoetis), where available, is not recommended for use in cats under 16 weeks age, nor is it considered to be of benefit to cats that are already FCoV antibody-positive cats. Therefore, it is of limited usefulness as initial exposure to FCoV is considered to be much earlier than 16 weeks in most cases. Furthermore, as current serological tests are unable to differentiate vaccinated from naturally exposed cats, it is of no benefit if trying to maintain a FCoV-free household. The FIP vaccine is either ‘not recommended’ or considered ‘non-core’ under current American Association of Feline Practitioners, ABCD, and World Small Animal Veterinary Association vaccination guidelines (and not even discussed in the British Small Animal Veterinary Association’s vaccination guidelines).


FIP is a common differential for disease in cats, particularly younger animals. Obtaining a definitive diagnosis by minimally invasive means can be difficult, and a balance of probability might need to be used to guide further testing. Although currently treatment is limited, novel antiviral agents show real promise for the future.


Further reading and useful resources

View Abstract

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.