The final hurdles for acute phase protein analysis in small animal practice


  • Peter D. Eckersall,

    1. Institute of Biodiversity, Animal Health & Comparative Medicine, University of Glasgow, Glasgow, UK
    Search for more papers by this author
  • Elizabeth Moreira dos Santos Schmidt

    1. Institute of Biodiversity, Animal Health & Comparative Medicine, University of Glasgow, Glasgow, UK
    2. Departamento de Clínica Veterinária, FMVZ – Unesp, Botucatu, Botucatu, SP, Brazil
    Search for more papers by this author

Studies on the acute phase proteins (APP) in small animal medicine have proliferated recently with particular emphasis on their value as biomarkers of infectious, inflammatory and neoplastic diseases. There have been several reviews on the extensive knowledge that has thus accumulated (Eckersall & Conner 1988, Ceron et al. 2005, Eckersall & Bell 2010, Cray et al. 2009). However, in many parts of the world the majority of veterinary practitioners do not have tests for APPs available on a regular basis. A number of pioneering laboratories have established methods that can be used for routine analysis for small animal practice (Jensen & Kjelgaard-Hansen 2006, Caldin et al. 2009, Martinez-Subiela et al. 2003, Shida et al. 2011). For less specialised laboratories there are significant problems to be overcome prior to wide spread availability of APP assays. Notable are the lack of a universally accepted calibration standard and the availability of a quality assurance (QA) scheme to ensure commonality between reported results. Here we review selected recent publications on APP in companion animals and thereafter discuss how to overcome such hurdles to the full use of APP assays in these species.

Only a few studies have evaluated temporal changes of APP in companion animals. In one such study, dogs with severe clinical signs of parvovirus enteritis that did not survive had higher C-reactive protein (CRP) concentrations at admission, and at 12 and 24 hours after admission when compared with the dogs that survived. Higher CRP was also associated with longer hospitalisation time and the sensitivity and specificity of CRP to differentiate between survivors and non-survivors at 24 hours after admission was 86·7 and 78·7%, respectively (McClure et al. 2013). In a complementary investigation, dogs that did not survive the infection and had leukopenia and also severe clinical signs of parvovirus enteritis had a 72-fold increase in CRP and a 3·1-fold increase in haptoglobin concentrations. In this study, CRP had 91% sensitivity and 61% specificity to predict mortality while haptoglobin had 52% sensitivity and 63% specificity (Kocaturk et al. 2010).

Dogs naturally infected with Spirocerca lupi showed a sustained increase in CRP associated with neoplastic transformation of the oesophageal nodules which means a poor prognosis for outcome (Mukorera et al. 2011). Haptoglobin and CRP concentrations were 95 and 68% increased, respectively, in dogs with clinical signs and S. lupi ova on faecal specimens (Mylonakis et al. 2012). CRP showed 93% sensitivity for detecting Leishmania infantum symptomatic dogs and 82% sensitivity for detecting asymptomatic dogs (Martinez-Subiela et al. 2002). In experimental Leishmania infantum infection in dogs, CRP, serum amyloid A (SAA) and haptoglobin concentrations were higher before immunoglobulin G and immunoglobulin M increases and the appearance of clinical signs. During the entire infection period, haptoglobin remained increased and the treatment induced a significant decrease in all APPs (Martinez-Subiela et al. 2011). Dogs naturally infected with Babesia rossi and with clinical signs of acute babesiosis had higher CRP concentrations that decreased following treatment with imidocarb dipropionate (Rafaj et al. 2013). CRP was not associated with outcome in Babesia rossi infection in dogs but none of the survivors had on the admission day CRP concentrations less than 63·2 mg/L (Koster et al. 2009).

In pyometra, SAA, CRP and haptoglobin were valid indicators of the inflammatory state of the uterus, making it possible to differentiate between open and closed cervix infections (Dabrowski et al. 2013) and to evaluate the severity of the inflammatory process especially with post-surgery complications (Dabrowski et al. 2009). In another study, 12 hours after standard soft-tissue surgeries (vasectomy, laparoscopic-assisted ovariohysterectomy and open approach ovariohysterectomy), CRP concentrations were significantly different and higher than prior to surgery (Kjelgaard-Hansen et al. 2013).

Investigation of APP in feline medicine has been relatively neglected and deserves to be further explored. The majority of investigations on feline APP have focused on pathophysiological conditions and have not generally examined time-series samples from cases. α1 Acid glycoprotein (AGP), SAA and haptoglobin concentrations in groups of hospitalised cats, submitted to surgery and with induced inflammation were 7 to 11 times higher than in healthy cats but there was no significant change of CRP concentrations between the evaluated groups (Kajikawa et al. 1999). AGP and haptoglobin and also SAA and AGP showed strong correlations between sick cats (Kann et al. 2012), demonstrating that APPs should be evaluated as an auxiliary diagnostic tool in this species.

AGP concentrations were significantly higher in tumour-bearing cats when compared to healthy cats, although no significant differences were found among the different groups of tumours (carcinomas, sarcomas and discrete round cell tumour) (Selting et al. 2000). In cats with lymphoma serum AGP concentrations did not provide useful information on response or survival, but the pretreatment AGP concentrations were significantly higher for cats with lymphoma than for healthy cats (Correa et al. 2001).

A recent study with Mycoplasma haemofelis and “Candidatus Mycoplasma haemominutum” demonstrated that experimental infections associated or not with FIV infection in cats induced significant increases in APP concentrations and showed variable concentrations of haptoglobin, SAA and AGP according to the stage of infection (Korman et al. 2012). In cats with high antibody titers for chlamydiae, the SAA concentration was significantly higher when the bacteria were detected in conjunctival swab samples than when no chlamydiae were detected (Holst et al. 2011).

For some time, AGP has been recognised as a biomarker of feline infectious peritonitis (FIP) (Duthie et al. 1997, Giordano et al. 2004, Paltrinieri et al. 2007) and has been reported to have 100% specificity and sensitivity in the diagnosis concordance for FIP (Giori et al. 2011). One of the few investigations of time-series monitoring of APP in cats has been carried out by following long-term pancreatitis in a cat in which the SAA concentration was useful as a marker of evaluating response to treatment (Tamamoto et al. 2009). The most recent time-series study, described in this volume (Gil et al. 2014), demonstrated that AGP and SAA are good predictors of immunomodulation in FIV and FeLV positive cats undergoing recombinant interferon omega (rFeIFN-ω) therapy and were influenced by the therapy protocol.

Although many studies have determined the APP concentrations in different diseases, the more clinically relevant have compared temporal changes in concentrations following the course of disease or treatment and show important information on sensitivity and specificity about the clinical outcome. Thus, future investigations of the clinical uses of APPs in dogs and cats should focus on their use as biomarkers during the course of inflammation. The dynamics of these proteins during the acute phase response reflects the underlying disease activity and can contribute to prediction of the outcome in diseased animals.

However, the potential of APP measurement is not universally available. Where APP assays have been established, it has been by the endeavour of specialist laboratories, but this is not possible in most diagnostic laboratories. The APP will only attain their due status once methodology is available for use by non-specialist veterinary diagnostic laboratories. There are a number of significant barriers to this process which can be exemplified by the status of assays for CRP in canine serum. First, rapid and reliable diagnostic assay kits need to be available at an economic price for use by diagnostic laboratories. This may be achievable in the near future as suitable immunoturbidimetric assays (Eckersall et al., 1991) become available from manufacturers. Second, while different research laboratories and diagnostic companies have been developing methods for CRP, there has been no common standard material available to calibrate the assays and without such a reference standard there is the possibility that assay results will vary. This has been undertaken for bovine and porcine APP (Skinner 2001). Third, there is no external QA scheme that includes CRP in circulation and without such a scheme the assay of CRP cannot be performed at the highest level of quality control to which laboratories aspire.

In human laboratory medicine, the International Federation of Clinical Chemistry has supported collaboration between researchers, industry and end-users to develop an international reference preparation for human serum proteins (Zegers et al. 2010). In veterinary clinical pathology this may be a role that could be developed by the International Society of Animal Clinical Pathology and the forthcoming XVIth Congress in Copenhagen would be an opportune time to instigate this process with a combination of academia, manufacturers and clinical pathology laboratories. A continuing collaboration across the stakeholders in veterinary diagnostics will be needed to ensure that the relevant APP are subsequently included in QA schemes.

The developments that have taken place in diagnostic measurement of APP are likely to be replicated in the future as biomarkers of diseases such as cardiac, neurological and neoplastic disease are developed and validated. The diagnostic laboratory community will have to face similar hurdles as further novel biomarkers, or combinations of biomarkers become available. Establishing the protocol for harmonised calibration and QA of the APP will create a precedent that can be extended to future assays as they convert from research laboratory into valuable diagnostic procedures for small animal practices.

Professor David Eckersall graduated with BSc in Biochemistry from the University of Liverpool in 1973 and from the University of Edinburgh with a PhD in 1977 having undertaken his studies in the Animal Breeding Research Organisation, the forerunner of the Roslin Institute. He was appointed to a lectureship in Veterinary Clinical Biochemistry at the Faculty of Veterinary Medicine, University of Glasgow in 1980 and is currently the Professor of Veterinary Biochemistry in the Institute of Biodiversity, Animal Health and Comparative Medicine at the same University. He was the recipient of the Heiner Sommer Prize of the International Society for Animal Clinical Pathology in 2008 for lifetime contributions to animal clinical biochemistry and the Siemens Prize of the American Association of Clinical Chemistry in 2010 for outstanding contributions to animal clinical chemistry. He is currently the Chair of the COST Action on Farm Animal Proteomics.

Professor Elizabeth Schmidt graduated with BSc in Veterinary Medicine from the University of Parana State (UFPR), Brazil in 1997; with a Master Degree in Veterinary Sciences in 2000 and from the Sao Paulo State University, Brazil (UNESP) in Veterinary Clinical Pathology with a PhD in 2008. She undertook a post-doctoral research study for two years (2008–2009) at Sao Paulo State University, Brazil in Veterinary Pathology. Since 2009 she is an Assistant-Professor of Veterinary Clinical Pathology and Parasitic Diseases at Sao Paulo State University (FMVZ – UNESP), Brazil.