Usefulness of acute phase proteins in differentiating between feline infectious peritonitis and other diseases in cats with body cavity effusions

Abstract

Objectives

The aim of this study was to evaluate the measurement of acute phase proteins (APPs) as a diagnostic tool to differentiate between feline infectious peritonitis (FIP) and other diseases in cats with body cavity effusions.

Methods

Cats with pleural, abdominal or pericardial effusion were prospectively enrolled. Cats were classified as having or not having FIP based on immunohistochemistry (if available) or a sophisticated statistical method using machine learning methodology with concepts from game theory. Cats without FIP were further subdivided into three subgroups: cardiac disease, neoplasia and other diseases. Serum amyloid A (SAA), haptoglobin (Hp) and α₁-acid glycoprotein (AGP) were measured in serum and effusion, using assays previously validated in cats.

Results

Serum and effusion samples were available for the measurement of APPs from 88 and 67 cats, respectively. Concentrations of the APPs in serum and effusion were significantly different in cats with and without FIP (P <0.001 for all three APPs). The best APP to distinguish between cats with and without FIP was AGP in the effusion; a cut-off value of 1550 µg/ml had a sensitivity and specificity of 93% each for diagnosing FIP.

Conclusions and relevance

AGP, particularly if measured in effusion, was found to be useful in differentiating between FIP and other diseases, while SAA and Hp were not. The concentration of all three APPs in some diseases (eg, septic processes, disseminated neoplasia) was as high as in cats with FIP; therefore, none of these can be recommended as a single diagnostic test for FIP.

Introduction

Feline infectious peritonitis (FIP) is a lethal infectious disease which can occur in two clinically distinct forms, the more common effusive (wet) form and the granulomatous (dry) form.^1–3 Abdominal distension and dyspnoea are common physical examination findings in cats with effusive FIP. Ascites or pleural effusion due to FIP have to be differentiated from other potential causes such as cardiac disease, neoplasia or septic effusion.^4,5 Although several diagnostic tests have been developed to diagnose FIP, differentiation between FIP and diseases with similar clinical presentation remains challenging in clinical settings. Although recent advances in the development of antiviral drugs might change the outcome of the disease in the future, ⁶ currently there is no widely available effective therapy for FIP. It is crucial to obtain a correct diagnosis because the ensuing consequences are currently fatal. In the majority of cases a combination of several diagnostic tests is necessary. ³

As FIP is an inflammatory condition, the concentrations of acute phase proteins (APPs) are expected to be increased. APPs are produced by hepatocytes as a part of the acute phase reaction that is an early and unspecific but highly complex reaction of the organism to a variety of injuries (infection, trauma, necrosis, malignant growth, etc).^7–9 Depending on the magnitude of their response to the triggers, APPs can be classified as major (10- to 100-fold increase), moderate (two- to 10-fold increase) and minor (<two-fold increase). In the cat the major APPs are serum amyloid A (SAA) and α1-acid glycoprotein (AGP); haptoglobin (Hp) is a moderate APP. ¹⁰

Several studies have investigated the diagnostic potential of APPs in diagnosing FIP, as well as the possible role of APPs in the pathogenesis of FIP.^11–17 These investigations had some drawbacks, limiting the implementation of the results into practice. First, APPs in cats with FIP were compared with APPs in healthy cats or cats exposed to feline coronavirus (FCoV) (based on a positive FCoV titre) instead of making comparison to cats potentially suffering of FIP based on similar clinical presentation.^11,13 Second, it was unclear if effusion or serum was used to measure APPs. ¹² Finally, the sample size was much too small to draw meaningful conclusions, with only four cats without FIP and eight cats with FIP included in one study. ¹⁴

The aim of the present study was to evaluate the ability of the APPs (measured in serum and effusion) to distinguish FIP from other diseases.

Materials and methods

Cats with pleural, abdominal or pericardial effusion or a combination thereof presented to the Small Animal Clinic of the Justus-Liebig-University Giessen, Germany, and Tierklinik Hofheim, Germany, over a 2 year period were prospectively included into the study. Abdomino-, thoraco- or pericardiocentesis were performed as routine diagnostic procedures to investigate the nature of the disease process in each cat. The cats from which <5 ml effusion could be obtained were excluded from this study.

Routine haematology and plasma biochemistry, tests for the detection of feline immunodeficiency virus (FIV) antibodies and feline leukaemia virus (FeLV) antigen (SNAP FIV/FeLV Combo Test; IDEXX Laboratories), as well as standard laboratory analysis (total protein, albumin and total nucleated cell count) and cytological examination of the effusion, were performed in all cats. Residual serum and effusion samples were stored for up to 24 months at −80°C until APPs were analysed.

Several diagnostic tests were undertaken with reference to FIP as a part of another investigation, namely Rivalta test in effusion, anti-FCoV antibody test in serum and effusion, immunofluorescent staining of FCoV antigen in macrophages in effusion, PCR in EDTA-blood and effusion. ¹⁸ The Rivalta test was performed in the Central Laboratory at the Small Animal Clinic, as described previously. ¹⁹ The direct and indirect detection of FCoV was carried out in the Institute for Veterinary Virology of the Justus-Liebig-University Giessen with the exception of the immunofluorescenting staining of FCoV antigen in macrophages in effusion, which was performed in an external laboratory (Landesbetrieb Hessisches Landeslabor, Giessen, Germany). Anti-FCoV antibody titres were determined by indirect immunofluorescence assay using methodology similar to a previous study. ²⁰ Nested reverse transcriptase PCR in EDTA blood and effusion were performed by the method of Herrewegh et al. ²¹ Immunofluorescent staining of FCoV antigen in macrophages in effusion was performed by a method identical to the one used by Parodi et al. ²² Additional diagnostic procedures (thoracic and/or abdominal radiography, abdominal ultrasonography, CT, echocardiography, cytology, histopathology, surgical exploration of the thorax or abdomen) were performed depending on the medical condition. The cats that died or were euthanased during hospitalisation underwent post-mortem examination at the Institute for Veterinary Pathology of the Justus-Liebig-University Giessen if owner consent was available. The pathohistological examination included immunohistochemistry as the gold standard to diagnose FIP, according to the method described by Kipar et al. ²³

Final diagnosis (FIP vs non-FIP) in the previous study was made based on immunohistochemistry results, if possible. ¹⁸ The remaining cats, for which no immunohistochemistry was available, either because cats were discharged from the clinic or owners declined post-mortem examination, were classified as either FIP-positive or FIP-negative by means of a sophisticated statistical method. ²⁴ This method combines semi-supervised machine-learning methodology with concepts from cooperative game theory using the results of several diagnostic tests in individuals where the true health condition is unknown. ²⁴ The method takes into account diagnostic accuracy and clinical relevance of individual tests and their combinations to establish the final diagnosis. Briefly, in 29/100 cats included in the previous study, ¹⁸ post-mortem examination (including immunohistochemistry) was performed and 11/29 cats were diagnosed with FIP. The results of several diagnostic tests (Rivalta test in effusion, anti-FCoV antibody test in serum and effusion, immunofluorescent staining of FCoV antigen in macrophages in effusion, PCR in EDTA blood and effusion) from 16/29 cats for which the FIP/non-FIP status was known based on immunohistochemistry were used to train the machine. The remaining 13/29 cats were used as test samples. The FIP/non-FIP status of the remaining 71 cats in the previous study was then evaluated using this statistical method.

Cats without FIP were further subdivided according to results of all diagnostic tests into three subgroups: (1) cardiac disease – cats in which echocardiography diagnosed a cardiomyopathy causing the effusion; (2) neoplasia – cats in which a tumour causing the effusion could be diagnosed based on cytological and/or histological examination; and (3) others – cats with effusion caused by other disease than FIP, cardiac disease or tumour.

The cats included in this study represent a subset of the population of the previous investigation. ¹⁸ The classification of cats into the FIP/non-FIP category, as well as further classification of cats without FIP into the three subgroups (cardiac disease, neoplasia, others), was performed in the same manner as in that study.

Measurement of acute phase proteins

Before the measurements of APPs, the serum and effusion samples were thawed at room temperature. The presence of haemolysis was evaluated visually.

Haptoglobin and SAA were measured with an automated analyser ABX Pentra 400 (Axon Lab) using the reagents from Phase Range Haptoglobin kit (Second Generation) (Tridelta Development) and LZ Test ‘Eiken’ SAA (Eiken Chemical), respectively. Both assays have been used for the measurement of APPs in cats.^11,25,26 AGP was measured with a manual method using the reagents from Phase Feline α 1 Acid Glycoprotein SRID Assay Kit (Tridelta Development). This test has also been validated and used in cats.^13,14

Statistical analysis

Data were tested for normality using the D’Agostino and Pearson omnibus normality test and were found not to be normally distributed. The numerical values are therefore presented as median and range. APPs in cats with and without FIP were compared using the Mann–Whitney U-test, while the four groups (depending on the final diagnosis) were compared using the Kruskal–Wallis test followed by Dunn’s comparison. For each parameter tested, a receiver operating characteristic (ROC) curve and the area under the curve (AUC) were calculated. Statistical analysis was performed using commercial software (GraphPad Prism version 6). Statistical significance was defined as P ⩽0.05.

Results

Serum samples were available for the measurement of APPs from 88 cats. In 67/88 cats, APPs were also measured in the effusion. The median age of the 88 cats was 8.3 years (range 0.3–17.9 years); in four cats the age was unknown. Thirty (34%) cats were female (four intact, 26 spayed) and 57 (65%) were male (seven intact, 50 castrated); in one (1%) cat, sex was not recorded. The most common breed was domestic shorthair (63 cats [72%]); 11 other breeds were recorded. Thirty-seven (42%) cats had ascites, 44 (50%) cats had pleural effusion, five (6%) cats had both ascites and pleural effusion, and two (2%) cats had pericardial effusion. Of the 88 cats, 20 (23%) cats had FIP (in nine cats this was confirmed by immunohistochemistry; the remaining 11 cats were determined statistically) and 68 (77%) cats were suffering from other diseases (in 16 cats the diagnosis was confirmed by histopathology). Among those, 22 (25%) cats had a cardiac disease, 24 (27%) had a tumour and 22 (25%) cats were suffering from other diseases. In the latter group, diagnosis was made in all but two cats. Cats were diagnosed with one of the following diseases: septic peritonitis or pythorax (n = 10), sepsis (with thoracic effusion classified as transudate based on cell count and total protein; n = 1), idiopathic chylothorax (n = 3), renal disease (n = 2), hepatic amyloidosis (n = 1), cholangiohepatitis (n = 1), permethrin intoxication (n = 1) and trauma (n = 1).

Two cats were positive for FIV, two for FeLV and one was positive for both FIV and FeLV. The underlying diseases responsible for the formation of effusion in these cats were FIP, septic peritonitis, carcinoma, lymphoma and trauma.

Three serum and 12 effusion samples were grossly haemolytic.

Concentrations of the three APPs in serum and effusion were significantly different in cats with and without FIP (P <0.001 for all three APPs) (Table 1).

Table 1.

Haptoglobin (Hp), α₁-acid glycoprotein (AGP) and serum amyloid A (SAA) concentration in serum (20 cats with FIP and 68 non-FIP cats) and effusion (14 cats with FIP and 53 non-FIP cats)

		FIP	Non-FIP	P value
Serum	Hp (mg/ml)	2.0 (2.0–9.0)	1.8 (0–2.0)	<0.001
	AGP (µg/ml)	2900 (960–5040)	690 (120–4500)	<0.001
	SAA (µg/ml)	98.5 (1.3–163.4)	7.6 (0.1–163.8)	<0.001
Effusion	Hp (mg/ml)	2.2 (0.1–9.3)	0.8 (0.1–2.5)	<0.001
	AGP (µg/ml)	2570 (1300–5760)	480 (190–3800)	<0.001
	SAA (µg/ml)	80.4 (0.1–207.4)	0.1 (0.1–182.7)	<0.001

Data are median (range)

There was a significant difference in concentrations of all three APPs in both serum and effusion between cats with FIP, cardiac disease, neoplasia and others (Figures 1a–c and 2a,b) with the sole exception of SAA concentration in serum which was not different between cats with FIP and other diseases (Figure 2c).

Figure 1.

(a) Haptoglobin, (b) α1-acid glycoprotein (AGP) and (c) serum amyloid A (SAA) concentration in effusion of cats with feline infectious peritonitis (FIP; n = 14), cardiac disease (n = 17), neoplasia (n = 21) and other diseases (n = 15). The boxes represent the 25th and 75th quartiles, with a horizontal line at the median. The whiskers represent the range of the data. Stars represent the significance levels (***P <0.001, **P <0.01, *P <0.05) when comparing the group with cardiac disease, neoplasia and other diseases with the FIP group

Figure 2.

(a) Haptoglobin, (b) α1-acid glycoprotein (AGP) and (c) serum amyloid A (SAA) concentration in serum of cats with feline infectious peritonitis (FIP; n = 20), cardiac disease (n = 22), neoplasia (n = 24) and other diseases (n = 22). The boxes represent the 25th and 75th quartiles, with a horizontal line at the median. The whiskers represent the range of the data. Stars represent the significance levels (***P < 0.001, **P <0.01, *P <0.05) when comparing the group with cardiac disease, neoplasia and other diseases with the FIP group

ROC curves for the three APPs to diagnose FIP are depicted for serum in Figure 3 and for effusion in Figure 4.

Figure 3.

Receiver operating characteristic curves of the three acute phase proteins, haptoglobin (Hp), α₁-acid glycoprotein (AGP) and serum amyloid A (SAA) in the serum of cats with feline infectious peritonitis (FIP; n = 20) and cats without FIP (n = 68)

Figure 4.

Receiver operating characteristic curves of the three acute phase proteins haptoglobin (Hp), α₁-acid glycoprotein (AGP) and serum amyloid A (SAA) in the effusion of cats with feline infectious peritonitis (FIP; n = 14) and cats without FIP (n = 53)

Table 2 shows the AUC for each APP including the best cut-off values based on the best likelihood ratio with its sensitivity and specificity.

Table 2.

Area under the curve (AUC), optimal cut–off values and sensitivities and specificities of the three acute phase proteins in serum and effusion

		AUC	Cut off	Sensitivity (%)	Specificity (%)
Serum	Hp	0.777	2.0 mg/ml	55	82
	AGP	0.899	2260 µg/ml	85	90
	SAA	0.800	97.3 µg/ml	55	87
Effusion	Hp	0.870	2.1 mg/ml	79	87
	AGP	0.950	1550 µg/ml	93	93
	SAA	0.885	43.6 µg/ml	71	91

Hp = haptoglobin; AGP = α1-acid glycoprotein; SAA = serum amyloid A

AGP in the effusion was shown to be the best marker to distinguish between cats with and without FIP; a cut-off value of 1550 µg/ml had a sensitivity and specificity of 93% each for diagnosing FIP.

Discussion

This is the first study to evaluate all three important APPs in cats, measured in both serum and effusion, as a diagnostic tool to differentiate FIP from other diseases. The best APP with which to distinguish between cats with and without FIP was AGP in effusion. The AUC of the ROC curve for this analyte was 0.95; a cut-off value of 1550 µg/ml had a sensitivity and specificity of 93% each for diagnosing FIP. The cut-off values for the tested parameters were chosen preferably to obtain a high specificity because a false-positive result could be potentially fatal for the cat. In the present study, only four cats with diseases other than FIP had AGP concentrations in the effusion higher than 1550 µg/ml and would therefore be falsely diagnosed as having FIP. Among these cats, three had a septic effusion and one had a metastatic pancreatic carcinoma, endocarditis and purulent bronchopneumonia. This cat had ascites, and although no tumour cells were found on cytological examination, this was most likely caused by the metastatic pancreatic carcinoma and the cat was included in the ‘tumour group’. However, this cat also tested positive for FIV and FeLV, which might have had an effect on APP levels. To what extent FIV or FeLV contributed to the concentrations of APPs in the remaining four cats with positive FIV and/or FeLV results cannot be commented on using current methods.

For all APPs tested in serum and effusion, some cats with septic processes and a few cats with disseminated neoplasias had APPs levels as high as cats with FIP. However, cats with cardiac disease had low APPs, with the sole exception of one cat with endocarditis with an AGP concentration in the effusion of 1500 µg/ml. This cat also had evidence of bronchopneumonia on postmortem examination, likely contributing to the high AGP concentration. However, this cat was included in the category ‘cardiac disease’, because AGP was measured in ascites, which was considered to have occurred secondary to heart failure based on the results of postmortem and cytological examination (modified transudate). Interestingly, for all APPs, there was less overlap between cats with FIP and cats with a septic process or neoplasia when APPs were measured in the effusion. The reason for this finding is currently unknown.

Based on these findings, a diagnostic algorithm useful in clinical practice could be proposed. In cats with body cavity effusions AGP in the effusion should be determined. If AGP levels are high, FIP, or septic or disseminated neoplastic disease, should be considered. Septic effusion can usually be recognised easily based on the results of haematology and cytological examination of the effusion. If there is no evidence of a septic process, further diagnostic tests for FIP should be performed. The major drawback to this diagnostic algorithm is the poor availability of AGP testing in comparison with other methods, including PCR. However, this might improve in the future with the introduction of assays that can be used with automated analysers. ²⁷

Several studies found increased levels of APPs in cats with FIP;^11–14 however, high levels of APPs were also reported in a variety of other inflammatory, as well as non-inflammatory, conditions such as neoplasia.^{25,26,28–30} Two investigations in cats with FIP evaluated AGP only,^13,14 and one study evaluated both AGP and Hp. ¹² All three APPs were measured in another report; however, the results were only compared with those in healthy cats. ¹¹ In one study, the ROC curve for serum AGP showed an AUC of 0.850, ¹³ which is similar to the value found in the present investigation (0.899). However, clinically healthy cats were used to calculate the ROC in that study, hampering direct comparison of the results. ¹³ An excellent sensitivity and specificity of 100% each was reported for serum AGP to diagnose FIP, using a cut-off value of 1.5 mg/ml (this equates to 1500 µg/ml). ¹⁴ However, only eight cats with FIP and four cats with other diseases were included and these results have to be interpreted with caution. Using the same cut-off value, AGP had a sensitivity of 85% and a specificity of 100% in diagnosing FIP in another report. ¹² This excellent specificity might be attributable to the fact that no cats with septic effusion were included. ¹² Septic cats were shown to have AGP values as high as cats with FIP in the present study, reducing the specificity of this parameter.

There are several limitations in the present study. First of all, immunohistochemistry, the gold standard method to diagnose FIP, could only be performed in 11 cats to confirm and in 18 cats to exclude FIP (cats suffering from diseases other than FIP) because post-mortem examination was only performed in this subset of cats. ¹⁸ However, in the latest comprehensive review on FIP diagnosis, histological examination and immunohistochemistry were not considered absolute requirements for diagnosis of FIP. ³¹ Instead, the importance of typical signalment, history and clinical signs of FIP were emphasised as important tools in achieving diagnosis, and the limitations of any laboratory method, including immunohistochemistry, were highlighted. ³¹ Other authors based the diagnosis of FIP on the findings of cytological examination of the effusion in combination with other laboratory parameters and typical history for cases where histology was not available.^32,33 To overcome the problem of not having post-mortem examination performed for all cats in this study, a sophisticated statistical method combining the results of multiple tests was used to classify cats as having or not having FIP with high confidence. ²⁴

The assays used for the measurement of APPs were not specifically evaluated for the measurement of these analytes in the effusion. However, according to the manufacturer’s instructions, the AGP assay used in this study can be used to determine AGP concentration in feline serum or other specimens. A similar assay has previously been used for the measurement of AGP in effusion. ¹² Prolonged storage might have influenced the concentrations of APPs in our samples as some had been stored for up to 2 years at −80°C prior to analysis. No information regarding stability of APPs in cats during prolonged storage has been published and the impact of the storage conditions and time is unknown.

Some laboratory methods are known to be influenced by the presence of haemolysis, lipaemia and bilirubinaemia in the analysed samples. To date, little is known about the influence of these substances on the measurement of APPs in dogs and cats. ¹⁰ To our knowledge, no specific precautions have been published for the immunoturbidimetric assay used for the measurement of the SAA concentration or the single radial immunodiffusion method used for the measurement of the AGP concentration in cats in this study. However, when the Hp concentration is measured, haemolysis could be a concern as free haemoglobin could bind haptoglobin in the sample and as a result of this falsely decreased Hp concentrations could be found. ³⁴ As haemolytic samples were not excluded from the analysis, their use might have influenced the results and might be responsible for the limited usefulness of Hp in diagnosing FIP.

Finally, it is important to note that the findings of the present study are applicable only in cats with body cavity effusions that might be suffering from the effusive (wet) form of FIP. Further studies are necessary to evaluate APPs in cats with granulomatous (dry) FIP.

Conclusions

While AGP was found to be a useful diagnostic tool in differentiating between FIP and other diseases causing effusions, the performance of SAA and Hp was not sufficient in this regard. Measurement of AGP in effusion provided the highest diagnostic yield among the APPs tested in both serum and effusion. However, because some overlap was seen for AGP between cats with FIP and a septic disease or disseminated neoplasia, AGP cannot be used as a single test for FIP.