• Open Access

Nested Case-Control Study of Feline Calicivirus Viruria, Oral Carriage, and Serum Neutralizing Antibodies in Cats with Idiopathic Cystitis

Authors

  • J. Larson,

    1. Department of Small Animal Clinical SciencesCollege of Veterinary Medicine, Michigan State University Veterinary Teaching Hospital, East Lansing, MI
    2. Animal Emergency and Treatment Center, 3927 W Belmont, Chicago, IL 60618
    Search for more papers by this author
  • J.M. Kruger,

    1. Department of Small Animal Clinical SciencesCollege of Veterinary Medicine, Michigan State University Veterinary Teaching Hospital, East Lansing, MI
    Search for more papers by this author
  • A.G. Wise,

    1. Diagnostic Center for Population and Animal HealthCollege of Veterinary Medicine, Michigan State University Veterinary Teaching Hospital, East Lansing, MI
    Search for more papers by this author
  • J.B. Kaneene,

    1. Center for Comparative EpidemiologyCollege of Veterinary Medicine, Michigan State University Veterinary Teaching Hospital, East Lansing, MI
    Search for more papers by this author
  • R. Miller,

    1. Center for Comparative EpidemiologyCollege of Veterinary Medicine, Michigan State University Veterinary Teaching Hospital, East Lansing, MI
    Search for more papers by this author
  • S.D. Fitzgerald,

    1. Department of Pathobiology and Diagnostic InvestigationCollege of Veterinary Medicine, Michigan State University Veterinary Teaching Hospital, East Lansing, MI
    2. Diagnostic Center for Population and Animal HealthCollege of Veterinary Medicine, Michigan State University Veterinary Teaching Hospital, East Lansing, MI
    Search for more papers by this author
  • M. Kiupel,

    1. Department of Pathobiology and Diagnostic InvestigationCollege of Veterinary Medicine, Michigan State University Veterinary Teaching Hospital, East Lansing, MI
    2. Diagnostic Center for Population and Animal HealthCollege of Veterinary Medicine, Michigan State University Veterinary Teaching Hospital, East Lansing, MI
    Search for more papers by this author
  • R.K. Maes

    1. Department of Microbiology and Molecular GeneticsCollege of Veterinary Medicine, Michigan State University Veterinary Teaching Hospital, East Lansing, MI
    2. Diagnostic Center for Population and Animal HealthCollege of Veterinary Medicine, Michigan State University Veterinary Teaching Hospital, East Lansing, MI
    Search for more papers by this author

  • The work was performed at the Michigan State University College of Veterinary Medicine. An abstract of this study was presented at the 25th Annual ACVIM Forum in Seattle, WA, 2007.

Corresponding author: John M. Kruger, DVM, PhD, Department of Small Animal Clinical Sciences, Room D208, College of Veterinary Medicine, Michigan State University Veterinary Teaching Hospital, East Lansing, MI 48824; e-mail: kruger@cvm.msu.edu.

Abstract

Background: The epidemiology of feline calicivirus (FCV) infection in cats with idiopathic cystitis (FIC) has not been investigated by contemporary molecular biologic methods.

Objectives: To determine the prevalence of and evaluate risk factors for FCV viruria, oral carriage, and virus neutralizing (VN) antibodies in cats with and without FIC.

Animals: Cats with nonobstructive FIC (n = 47), obstructive FIC (n = 22), and FCV upper respiratory tract infection (URI; n = 25), and healthy client-owned (n = 18) and colony-housed (n = 24) cats.

Methods: Oropharyngeal secretions and urine were evaluated with a FCV p30 gene-based real-time reverse-transcriptase polymerase chain reaction (RT-PCR) assay. Serum VN antibody titers were determined by a modified microtiter assay. Associations of risk factors with log-transformed antibody titers were determined by multivariable generalized linear regression.

Results: FCV viruria was detected in 4 (6%) and 3 (12%) cats with FIC and URI, respectively. In 3 FIC cats, viruria was unassociated with detectable oral virus carriage. Oral FCV carriage was detected in 7 (10%) FIC cats. Median antibody titers were significantly higher in cats with obstructive FIC (1 : 256), nonobstructive FIC (1 : 128), and URI (1 : 512) compared with healthy client-owned (1 : 16) and colony-housed (1 : 4) cats (P < .001). Other than disease, multivariate analysis did not identify any other explanatory variables for increased titers in cats with FIC or URI.

Conclusions and Clinical Importance: FCV viruria was detected in cats with FIC and URI, however, its etiologic significance is uncertain. Serologic results suggest increased FCV exposure in FIC cats compared with controls. Further investigations are needed to clarify the potential role of FCV in FIC.

Abbreviations:
FCV

feline calicivirus

FIC

feline idiopathic cystitis

RT-PCR

reverse-transcriptase polymerase chain reaction

SPF

specific-pathogen-free

URI

upper respiratory tract infection

VN

virus neutralizing

Feline idiopathic cystitis (FIC) is a common disorder that accounts for nearly two thirds of all lower urinary tract diseases encountered in young to middle aged cats.1 The disorder is characterized by varying combinations of hematuria, dysuria, pollakiuria, periuria, and urethral obstruction.1 The cause(s) of FIC is uncertain and proposed etiologies have included dysfunctional urothelium, neurogenic inflammation, neuroimmune disorders, systemic psychoneuroendocrine dysfunction, and viral urinary tract infections.2 Feline calicivirus (FCV) was first implicated as a potential causative agent of FIC in 1969, yet its etiopathogenic role in the disorder remains uncertain.3 Isolation of an FCV from a Manx cat with urethral obstruction and experimental induction of urethral obstruction in 80% of conventionally reared cats after urinary bladder, aerosol, or contact exposure with this virus directly supported the concept of a viral etiology.3,4 In subsequent studies, virus-like particles that were similar in size and morphology to FCV were observed by electron microscopy in 38% of 92 urethral crystal-matrix plugs obtained from male cats with obstructive idiopathic cystitis.5 However, the identity of these virus-like particles was not confirmed by antigenic or molecular biologic analyses. Moreover, FCV has been isolated only sporadically from cats with FIC.4,6 In 1 study, FCV was detected by virus isolation in 8% of male cats with obstructive and 4% cats with nonobstructive FIC.6

Uncertainty regarding the causative role of FCV in FIC is due, at least in part, to lack of large-scale epidemiologic studies utilizing contemporary molecular biologic viral detection methods optimized for detection of FCV urinary tract infections. Based on the unified concept of causation proposed by Evans,7 criteria for establishing a causative relationship between FCV and FIC should include demonstrating that exposure to FCV is more common in cats with FIC than in cats without the disease when all risk factors are held constant.8 The objective of this study was to determine if FCV viruria, oral carriage, and magnitude of virus neutralizing (VN) antibody titers were higher in cats with FIC compared with cats without FIC. To achieve this objective, a nested case-control study was designed for cases of FIC (divided into nonobstructive and obstructive FIC groups) and controls without FIC (divided into groups with different potential exposures to FCV).

Materials and Methods

Study Populations

The prevalence of FCV viruria, oral carriage, and VN antibodies was determined in 3 groups of cats. Group 1 consisted of 69 client-owned cats with FIC divided into 2 subgroups: 47 cats with nonobstructive FIC and 22 cats with obstructive FIC.1 Cases were recruited from the Michigan State University Veterinary Teaching Hospital (MSUVTH) and local private feline practices. A diagnosis of nonobstructive FIC was based on the presence of 3 of 5 clinical signs (hematuria, pollakiuria, dysuria, strangury, periuria) within the past 24 hours and exclusion of other causes of these signs by evaluation of a CBC, serum biochemistry profile, complete urinalysis, quantitative culture of urine for aerobic bacteria, survey abdominal radiographs, and contrast urethrocystography (double-contrast cystogram and contrast urethrogram), abdominal ultrasonography, or urethrocystoscopy.9 A diagnosis of obstructive FIC was based on clinical signs and physical examination findings, including a large, overdistended urinary bladder that could not be expressed by manual compression and exclusion of urolithiasis and bacterial urinary tract infection by evaluation of a urinalysis, culture of urine for aerobic bacteria, and survey abdominal radiographs. In 8 cases, the lower urinary tract was further evaluated with ultrasonography or contrast radiography. This group consisted exclusively of male cats in which obstructive uropathy was caused by crystalline-matrix urethral plugs (n = 7) or idiopathic urethral obstructive disease (n = 15). Urethral crystalline-matrix plugs were identified by physical or radiographic procedures. All cats recovered after appropriate medical management for urethral plug-induced or idiopathic obstructive uropathy.

Group 2 consisted of 25 shelter-housed (n = 23) or client-owned (n = 2) cats diagnosed with FCV-upper respiratory tract infection (URI). A total of 56 cats with active signs of upper respiratory tract disease were recruited from the MSUVTH and 4 regional animal shelters. Twenty-five (45%) cats were diagnosed with FCV infection: 0 of 2 MSUVTH cats, 4 of 5 shelter A cats, 6 of 15 shelter B cats, 2 of 14 shelter C cats, and 13 of 20 shelter D cats. Shelter cats were observed for a minimum of 7 business days. Cats in this group were evaluated with a minimum of a physical examination, complete urinalysis, and quantitative culture of urine for aerobic bacteria. FCV infection was confirmed by detection of viral nucleic acids in oropharyngeal secretions by a FCV reverse-transcriptase polymerase chain reaction (RT-PCR) assay. Diagnostic imaging was not performed in this group of cats because of the infectious nature of feline upper respiratory tract disease. Cats were excluded if they had any historical or clinical findings indicative of urinary tract illness (eg, gross hematuria). However, cats with microscopic hematuria were not excluded because of the inability to reliably distinguish cystocentesis-induced traumatic hematuria from that associated with urinary tract disease. Cats in group 2 served as positive FCV-infection controls.

Group 3 initially consisted of 46 clinically healthy cats without a history of urinary or respiratory tract disease. Four of these asymptomatic cats were subsequently found to be RT-PCR positive for FCV in oropharyngeal secretions. Because asymptomatic carriers constitute a heterogeneous population of FCV-infected cats whose serologic responses have not been well characterized, these 4 cats were exclude from analyses. The remaining 42 healthy group 3 cats were divided into 2 subgroups: 18 client-owned cats and 24 colony-housed cats. Colony-housed cats were conventionally reared and were part of a research (n = 8) or commercial blood-bank (n = 16) cat colony. Both colonies had histories of endemic FCV and all resident cats received routine FCV vaccinations. Cats in both subgroups were evaluated with a minimum of a history, physical examination, complete urinalysis, and quantitative culture of urine for aerobic bacteria. All client-owned cats were evaluated with survey abdominal radiographs; radiographs were not performed in colony-housed cats. These cats served as clinically healthy control cats.

This study was reviewed and approved by the Michigan State University All University Committee on Animal Use and Care.

Sample Collection and Analyses

Six milliliters of urine were collected aseptically by cystocentesis or by catheterization from each cat. All FIC cats had clinical signs within the previous 24 hours. All FCV-URI cats had active clinical signs at the time sampling. Before cystocentesis, the skin was disinfected with a povidone-iodine solution (30 seconds) followed by 70% ethanol (30 seconds) to reduce risk of contamination of cystocentesis sites with virus originating from saliva.10 Oropharyngeal secretions were collected with a premoistened sterile Dacron swab.a The swab was immediately immersed in a sterile vial containing 2 mL of viral transport medium (Bovarnick's diluent). All swabs, syringes, tubes, and specimens were handled with gloves to prevent ribonuclease contamination. Urine and oropharyngeal swabs were immediately cooled to 4°C for storage until RNA isolation was performed. A venous blood sample (3 mL) was collected from all cats; serum was separated by centrifugation for 10 minutes at 1,250 ×g and stored at −70°C until serologic analysis.

Total RNA was isolated from 140 μL of urine with a commercial silica gel-based extraction column.b,11 Oropharyngeal secretions were eluted from the swab and total saliva RNA was isolated from 140 μL of the eluate with the silica-gel extraction column described above. All RNA isolations were completed within 24 hours of sample collection. Isolated RNA was stored at −70°C until RT-PCR amplification. Total RNA samples were assayed for FCV nucleic acids by a FCV p30 gene-based real-time RT-PCR assay.12 A sample of noninfected tissue culture medium was extracted in parallel with each set of specimen RNA extractions and served as a FCV-negative extraction control. In addition, FCV-positive and -negative (water) RT-PCR reaction controls were included in each set of RT-PCR reactions.

Serum specimens were analyzed for FCV neutralizing (VN) antibodies with a modified microtiter assay at the Michigan State University Diagnostic Center for Population and Animal Health.13 Serum VN titers were expressed as the reciprocal of the highest serum dilution resulting in complete inhibition of cytopathic effect. A positive result was defined as a titer ≥4.

Statistical Analyses

Univariate analyses of population characteristics and clinical findings were conducted to identify significant differences between study populations, using P < .05 as the measure of significant association. Categorical risk factors (gender, outdoor access, multiple cat environment, FCV vaccination status, hematuria, pyuria, and crystalluria) were tested with Mantel-Haenszel χ2 statistic and Fisher's exact 2-tailed test (SAS 9.1.3),c and continuous risk factors (age, time since last vaccination, urine specific gravity, and urine pH) were assessed with the nonparametric Kruskall-Wallis χ2 statistic.

The univariate association between natural log-transformed titers and selected population risk factors was assessed by the Mantel-Haenszel χ2 statistic (SAS 9.1.3),c and odds ratios with 95% confidence intervals were calculated. Selection of risk factors was based on previous experimental and epidemiologic studies,d,3–6,13–17 our assessment of the likelihood that a variable might influence the frequency, magnitude, or both, of FCV antibody titers, and adequacy of data for analysis. Selected variables included disease category, sex, age, vaccination status, time since last vaccination, outdoor access, multiple-cat environment, hematuria, and prior unspecified lower urinary tract disease. Spearman's rank-correlation coefficients (r) between all risk factors were calculated to identify any potential interactions that would need to be taken into consideration during multivariable analysis (r > 0.75).

A multivariable generalized linear regression model was used to identify relationships between log titer and risk factors. Specific risk factors evaluated included disease category, age, sex, vaccination status, time since last vaccination, outdoor access, multiple cat environment, and hematuria. A backward model building approach was used. First, a full-rank model, containing all possible risk factors with interaction terms, was executed. Next, risk factors or interaction terms were removed from the model one at a time, based on P-values and comparison of model coefficient of determination (R2) before and after removal of the factor. The final model was that which had the best combination of high model R2 and low P-values for risk factors. Associations between risk factors and log titer were reported as odds ratios with 95% confidence intervals.

Results

Cats affected with obstructive or nonobstructive FIC were typically young to middle aged (Table 1). Differences in median ages between disease and healthy control groups were not significant (P= .20); however, colony-housed healthy cats tended to be somewhat younger than cats in other groups (P= .086). The majority of FIC cats and healthy control cats were male, mixed breed, and lived exclusively indoors with at least 1 other cat (Table 2). There were significantly more males in the disease groups compared with healthy controls (P= .026; Table 2). Purebred cats accounted for 1 of 22 (5%) obstructive FIC cats, 4 of 47 (9%) nonobstructive FIC cats, and 1 of 18 (5%) healthy client owned cats; all FCV-URI and healthy colony-housed cats were of mixed breed. A small proportion of FIC cats and healthy client-owned cats had unsupervised outdoor access, whereas all healthy colony-housed cats resided exclusively indoors (Table 2). Access to outdoors was largely unknown in FCV-URI cats. The majority of FIC and healthy control cats resided in multicat environments, whereas the housing status of FCV-URI cats was largely unknown (Table 2). Forty-five percent of obstructed and nonobstructed FIC cats had a history of prior unspecified lower urinary tract disease. Only 2 (9%) obstructed and 3 (6%) nonobstructed FIC cats had histories of prior unspecified upper respiratory tract disease. With the exception of 1 healthy client-owned cat whose vaccination status was unknown, all FIC and healthy client-owned and colony-housed cats had been vaccinated previously for FCV. The vaccination status of cats with FCV-URI was largely unknown. Median time since last vaccination was significantly shorter in colony-housed cats compared with other groups (P < .001; Table 1).

Table 1.   Univariate comparisons of selected continuous population characteristics of cats with obstructive and nonobstructive idiopathic cystitis, feline calicivirus upper respiratory tract infection, and clinically healthy cats (client owned and colony housed).
GroupAge (Months)Months from Vaccination
nMinimum25%Median75%MaximumnMinimum25%Median75%Maximum
  1. FCV, feline calicivirus; FIC, feline idiopathic cystitis; URI, upper respiratory tract infection.

FIC nonobstructed476.126.751.280.3144.946.34.312.524.854.7
FIC obstructed2211.829.345.986.6119.5222.27.518.627.983.4
FCV-URI122.05.99.315.636.712UnknownUnknownUnknownUnknownUnknown
Client owned180.85.111.418.439.9186.710.130.549.285.5
Colony housed241.43.45.28.311.8241.43.45.28.311.8
Kruskall-Wallis χ2 (P-value)25.9 (<.001)16.9 (<.001)
All healthy controls423.011.042.561.5177.1420.83.46.111.139.9
All others812.018.444.079.0144.9680.35.712.826.483.4
Kruskall-Wallis χ2 (P-value)1.6 (.20)11.9 (<.001)
Healthy client owned180.85.111.418.439.9186.710.130.549.285.5
All others1052.017.343.266.9144.9920.34.510.824.083.4
Kruskall-Wallis χ2 (P-value)0.05 (.82)0.1 (.72)
Healthy colony housed241.43.45.28.311.8241.43.45.28.311.8
All others992.017.145.880.3177.1860.35.612.324.083.4
Kruskall-Wallis χ2 (P-value)2.9 (.086)14.1 (<.001)
Table 2.   Univariate comparisons of the proportions of cats in each study group with selected categorical population characteristics and laboratory findings.
GroupN% Male% Outdoor
Access
% Multicat
Home
% Microscopic
Hematuria
% Crystalluria
  1. FCV, feline calicivirus; FIC, feline idiopathic cystitis; URI, upper respiratory tract infection.

FIC nonobstructed4772.319.178.753.225.5
FIC obstructed22100.027.359.1100.072.7
FCV-URI2564.0UnknownUnknown24.08.0
Healthy client owned1850.027.883.333.35.6
Healthy colony housed2462.50100.020.845.8
Mantel-Haenszel χ2 (P-value)3.9 (.049)3.5 (.060)1.1 (.29)13.7 (<.001)0.03 (.84)
Fisher's exact 2-tailed P-value.0015.0013<.001<.001<.001
All healthy controls4257.111.992.926.228.6
All others9476.616.053.256.431.9
Mantel-Haenszel χ2 (P-value)5.3 (.021)0.4 (.53)20.0 (<.0001)10.5 (.0012)0.2 (.69)
Fisher's exact 2-tailed P-value.0261.6104<.0001.0015.8411
Healthy client owned1850.027.883.333.35.6
All others11873.712.762.749.134.7
Mantel-Haenszel χ2 (P-value)4.2 (.040)2.8 (.093)2.9 (.087)1.6 (.21)6.2 (.012)
Fisher's exact 2-tailed P-value.0524.1438.1124.3108.012
Healthy colony housed2462.517.86100.020.845.8
All others11272.3058.052.727.7
Mantel-Haenszel χ2 (P-value)0.9 (.339)5.0 (.025)15.3 (<.001)8.0 (.0047)3.0 (.081)
Fisher's exact 2-tailed P-value.3355.0239<.001.0062.092

Microscopic hematuria was detected significantly more frequently in urine specimens obtained from FIC cats compared with FCV-URI cats and healthy control cats (P < .001; Table 2). Pyuria was an uncommon finding and the frequency of pyuria was not significantly different between groups (P= .20). The frequency of crystalluria was significantly different between groups and was observed more commonly in cats with obstructive FIC and healthy colony-housed cats (P < .001; Table 2). Struvite crystals were identified in 93% of specimens with crystalluria.

FCV nucleic acids were detected in oral swab samples by RT-PCR in all (100%) URI cats, in 2 (9%) cats with obstructive FIC, and in 5 (11%) cats with nonobstructive FIC (Table 3). FCV viruria was detected in 3 (6%) nonobstructive FIC cats, 1 (5%) obstructive FIC cat, and 3 (12%) URI cats; viruria was not detected in healthy control cats (Table 3). However, limited numbers of viruric cats precluded statistical evaluation. Three of the 4 FIC cats with viruria did not have detectable concurrent oral carriage of FCV. All viruric FIC cats were males and ranged in age from 23.5 to 149.9 months (mean of 59.7 ± 44 months). However, 2 of 3 viruric URI cats were female. Time since last vaccination in viruric FIC cats ranged from 3.6 to 39.9 months (mean 20.7 ± 10.7 months).

Table 3.   Results of FCV RT-PCR assay of urine and oropharyngeal secretions and FCV virus neutralizing antibody tests from cats with idiopathic cystitis, FCV-upper respiratory infection (URI), and clinically healthy client-owned and colony-housed cats.
FactorGroup
FIC ObstructedFIC NonobstructedFCV-URIClient OwnedColony Housed
  • NA, not applicable; VN, virus neutralizing; FCV, feline calicivirus; FIC, feline idiopathic cystitis; RT-PCR, reverse-transcriptase polymerase chain reaction.

  • a

    Selection criteria excluded cats that were oral swab positive.

  • b

    b Selection criteria excluded cats that were oral swab negative.

  • c

    Significantly different from other groups (P= .011).

  • d

    Significantly different from healthy controls (P<.001).

Number of cats2247251824
FCV RT-PCR positive
 Oral swab only2422NAaNAa
 Urine only12NAb00
 Oral swab and urine013NAaNAa
FCV VN antibody titer
 Number positive (≥4)21 (96%)43 (92%)24 (96%)14 (78%)17 (70%)c
 Minimum<4<4<4<4<4
 25%16161284<4
 Median256d128d512d164
 75%102451220486416
 Maximum≥40964096≥40962048≥4096

Overall, the seroprevalence of FCV was significantly higher in diseased cats compared with healthy control cats (P= .0034; Table 3). Likewise, median VN titers for FCV-URI or FIC cats were significantly higher than healthy control cats (P < .001; Table 3). Univariate analysis of relationships of antibody titers to explanatory variables revealed significant associations between VN titers and disease category (FIC, P= .025; FCV-URI, P < .001; healthy, P < .001), vaccination status (P= .0014), outdoor access (P= .001), and multiple cat environments (P= .0021). Other variables evaluated, including age, sex, hematuria, time since last vaccination, and prior lower urinary tract disease, were not significantly associated with VN titers. When multiple explanatory variables, including disease, age, sex, vaccination status, time since last vaccination, multiple cat environment, outdoor access, and hematuria, and their relationship to FCV antibody titers were simultaneously evaluated by multivariable regression, the only significant association with an increased antibody titer identified was disease category (P < .001; Table 4). No significant interactions were identified between risk factors considered for inclusion in the multivariable model. In terms of relative risk, FCV-URI cats were approximately 38–57 times more likely obstructive FIC cats 11–17 times more likely, and nonobstructive FIC cats 5–8 times more likely to have an increased titer compared with healthy client-owned or colony-housed healthy control cats, respectively.

Table 4.   Multivariable analysis of the association of risk factors with FCV virus neutralizing antibody titer results using a generalized linear regression model (GLM) for log-transformed antibody titer values.
Risk FactorModel F (P-value)Level/UnitsT (P-value)Odds Ratio95% Confidence Interval
  1. FCV, feline calicivirus; FIC, feline idiopathic cystitis; URI, upper respiratory tract infection.

  2. Model F = 8.35, P < .0001, R2= 26.3%.

Age3.5 (.0647)Months1.9 (.064)1.00.99–1.02
Group9.8 (<.0001)FIC nonobstructive3.5 (<.001)7.72.5–24.0
 FIC obstructive4.2 (<.001)16.84.5–62.7
 FCV-URI5.0 (<.001)56.511.7–273.2
 Healthy client owned0.6 (.55)1.50.4–6.1
 Healthy colony housedBaseline

Discussion

In the present study, an FCV RT-PCR assay optimized for urine specimens11 in conjunction with conventional serologic methods was used to investigate the prevalence of FCV shedding and exposure in cats with obstructive and nonobstructive FIC, cats with FCV-URI, and clinically healthy cats. Viruria was detected in low numbers of cats with FIC or FCV-URI. Median antibody titers for cats with obstructive FIC, nonobstructive FIC, and FCV-URI were significantly higher than those of healthy control cats. Other than disease, multivariate analysis did not identify any other explanatory variables for increased titers in cats with FIC or FCV-URI.

FCV viruria was detected in a low proportion of FIC cats (6%) and a somewhat higher proportion of positive control cats with FCV-URI (12%). The relatively low frequency of viruria in FIC cats in the present study was similar to a previous study that used a modified virus isolation technique for detection of FCV in urine from cats with FIC.6 In that study, FCV viruria was detected in 1 of 28 (4%) cats with nonobstructive FIC and in 1 of 12 (8%) male cats with obstructive FIC. The frequency of viruria in the present study, however, was substantially lower than that predicted by earlier electron microscopic studies where virus-like particles similar in size and morphology to FCV were observed in the matrix of 38% of 92 urethral plugs obtained from male cats with obstructive FIC.5 Reasons for this disparity are unknown, but may be related to sample type, viral urine shedding patterns, virus strain variation, infectious dose, or sensitivities of detection methods. In previous experimental studies of conventionally reared cats, FCV viruria was detected by virus isolation for only 1–4 days after urinary bladder FCV exposure4 and for 1 day after intranasal exposure.18 Similarly, viruria was detected sporadically by RT-PCR for 1–18 days after oropharyngeal exposure of specific-pathogen-free (SPF) cats with a urinary or respiratory FCV isolate.d It is plausible that urethral plugs form at a time that coincides with urine virus shedding, thus increasing the likelihood of detection by electron microscopy.

The disparity between RT-PCR and electron microscopy-based studies may also reflect differences in sensitivities and specificities of the 2 viral diagnostic methods. Although virus-like particles observed by electron microscopy in urethral plug matrix5 were of similar size and morphology to FCV,19 it is possible that these particles represent a different viral agent or noninfectious cellular elements. Molecular diagnostic assays for FCV have been shown to have a sensitivity, specificity, and diagnostic range comparable to that of conventional virus isolation and have largely replaced virus isolation as the diagnostic method of choice for detection of FCV infection.12,20–22 The RT-PCR FCV assay used in the present study12 had a sensitivity, specificity, and diagnostic range similar to that reported for other FCV RT-PCR assays.20,22 However, urine has also been recognized as a particularly difficult substrate for recovery and amplification of nucleic acids.11 False-negative RT-PCR assay results may be because of low virus concentrations, loss of viral target RNA integrity in feline urine, or the presence of substances in feline urine that are inhibitory to reverse transcription and PCR amplification. Competitor RNA (mimic RNA) has been used as an exogenous internal control to identify RT-PCR assay inhibition in detection of RNA viruses in complex samples.23 Further studies confirming the identity and prevalence of caliciviruses in feline urethral plugs are essential and should utilize RT-PCR assays that incorporate exogenous competitor RNA controls to identify potential assay inhibition.

Unfortunately, the low prevalence of FCV viruria precluded statistical evaluations and limited our ability to assess its etiologic significance in FIC cats. It is clear from our observations and those of others that ororespiratory FCV strains may be associated with infection of kidney and urinary bladder tissues and concurrent viruria.d,18,24–26 Consequently, we cannot determine whether viruria detected in FIC cats represents urinary tract infection with FCV strains that directly or indirectly induce urothelial injury and subsequent signs of lower urinary tract disease, or whether it represents innocent coincidental urine shedding associated with a concurrent clinically inapparent and unrelated FCV infection. Persistent oral shedding of FCV in healthy asymptomatic cats is a well-known phenomenon.18,21,25–30 Similarly, asymptomatic oral shedding may occur after reexposure to virulent FCV strains in cats previously infected or vaccinated.13,21,27–29,31 It is possible that a similar phenomenon of asymptomatic shedding may occur in the urinary system of infected cats. However, in the small number of asymptomatic cats with persistent oral shedding that have been evaluated, urine was consistently negative for FCV viruria.25,30 In studies of SPF cats exposed to a urinary strain or virulent respiratory FCV strain, urinary bladder lesions tended to be more severe and were seen more frequently in urinary strain infected cats, whereas oral and respiratory tract lesions predominated in the respiratory strain infected cats.d These observations raise the possibility that in some FIC cases, FCV urinary tract infections may be associated with pathologic changes in the lower urinary tract. A limitation of the present study is that detection of viruria does not distinguish between coincidental or pathologic virus shedding. Therefore, future studies should utilize methods that allow localization of viral antigens or nucleic acids in urinary bladder lesions.

Because of natural infection and widespread vaccination, the seroprevalence of FCV is generally high, ranging from 57 to 99% in client-owned healthy cats, and nearly 100% in cats with upper respiratory tract disease.14,15,32 Our results were similar, with FCV VN antibodies detected in 96% of URI cats, 92% of FIC cats, and 74% of healthy control cats. However, median antibody titers for cats with obstructive and nonobstructive FIC were significantly higher than those of vaccinated healthy control cats and were not significantly different from the median for FCV-URI cats. Cats exposed to virulent FCV rapidly develop homologous VN antibody titers that peak at approximately 4 weeks postinfection and gradually decrease over months to years.13,16,17,27 Reexposure to field or vaccine virus results in an anamnestic response and substantial increases in VN titers.13,15,17,27,33 Higher titers observed in FCV-URI cats are likely consistent with persistent FCV infection or reexposure. Increased FCV titers in FIC cats may likewise suggest persistent infection or reexposure. Multivariate analysis did not identify any other explanatory variables for increased titers in cats with FIC. Similarly, previous surveys of feral cats in the United Kingdom and healthy client-owned cats in North America did not identify associations between FCV VN titers and age, sex, social status, health status, feeding group, or lifestyle.15,34 Although our serologic results indirectly support increased exposure of FIC cats to FCV, we caution that interpretation of our results is limited by the cross-sectional nature of the study and lack of sequential titer determinations. Diagnosis of recent FCV exposure in FIC cats may be facilitated by analysis of paired serum specimens for increasing titers of VN antibodies. Identification of other potential explanatory variables for increased FCV titers in cats with FIC requires further investigations.

A central question arising from results of this and previous studies is whether FCV is causatively related to FIC or represents a concurrent, but unrelated infection. We emphasize that our observations only identified an association between FIC and FCV exposure; they do not define the etiologic relationship between FIC and FCV. Nevertheless, detection of high levels of VN antibodies supports that the notion that FCV may, in some way, be directly or indirectly related to the pathogenesis of FIC. FCV could cause lower urinary tract disease directly by inducing uroepithelial cell death, vascular endothelial injury, or uroepithelial barrier dysfunction,19,35,36 or indirectly by triggering immunopathologic injury or epigenetic reprogramming of uroepithelial stem cells.37–41 Alternatively, FCV infection could be a trigger of systemic psychoneuroendocrine dysfunction. In this scenario, an unrelated FCV infection may serve as a physiologic stressor of sufficient magnitude to induce central dysregulation of autonomic neurons regulating bladder contraction, directly induce urothelial injury, or both and subsequent signs and sequelae of idiopathic cystitis.42 Finally, we cannot exclude the possibility that FIC may in some way predispose to FCV infection or that FCV infection may nonspecifically predispose to disease in general. Regardless, our observations emphasize the need for further studies to establish the etiologic significance of FCV exposure in FIC and to characterize the potential cause-and-effect relationship between FCV and FIC.

Footnotes

a Fisherbrand, Thermo Fischer Scientific, Waltham, MA

b QIAamp Viral RNA Mini Kit, QIAGEN Inc, Valencia, CA

c SAS 9.1.3, SAS Institute Inc, Cary, NC

d Kruger JM, Pfent CP, Clark AK, et al. Feline calicivirus-induced urinary tract disease in specific-pathogen-free cats. J Vet Intern Med 2007;21:684 (abstract)

Acknowledgments

The authors thank Marlee Richter, Michelle Granzow, Michelle Fritz, and Drs Catherine Pfent, April Clark, Phillip Strom, Julie Thompson, and Naomi Balduff for their technical assistance, and Drs Ann Hale and Mari Nichol for their referral of cases. This work was supported by the American College of Veterinary Internal Medicine Foundation, the Morris Animal Foundation, and the Michigan State University Center for Feline Health and Well-Being.

Ancillary