Gems from ACVIM 2005

Alice M. Wolf, DVM, DACVIM, DABVP

Adjunct Professor – TAMU

Chief Consultant: Veterinary Information Network

A New Blood Group Antigen in Domestic Shorthair Cats: the Feline Mik Red Cell Antigen

ACVIM 2005

N.M. Weinstein; M.C. Blais; K. Greiner; D.A. Oakley; A. Hyson; U. Giger
Section of Medical Genetics, University of Pennsylvania, Philadelphia, PA

The feline blood group system is currently defined by blood types A, B, and AB, and, thus far is the only system recognized in cats. Naturally-occurring alloantibodies have been well documented in type A and type B cats and require that blood typing be performed prior to both blood transfusion and breeding to assure appropriate blood compatibility. Blood incompatibilities, unrelated to the AB blood group system, have also been recognized following blood transfusion through crossmatching cats or as a result of acute hemolytic transfusion reactions. Utilizing standard tube and novel gel column crossmatching techniques, the presence of a clinically relevant alloantibody, formed against a newly discovered feline red blood cell antigen, has been identified.

Evidence of this alloantibody was demonstrated through routine crossmatch testing in a healthy, previously never transfused, type A, DSH, blood donor cat, named Mike. Mike's plasma was incompatible with more than 50 type A, B, and AB feline patients and donors, with the exception of two, likely related, type A, DSH, blood donor cats. It was, therefore, concluded that Mike lacks the red cell antigen, referred to as Mik, as do these two additional blood donor cats. The anti-Mik alloantibody formed by Mike and one of the other Mik red cell-negative cats resulted in equivalent incompatible crossmatch results with over 30 additional type A, B, and AB cats. Furthermore, the anti-Mik plasma titers were 1:64 and 1:32, respectively, for Mike and for the second Mik red cell-negative blood donor cat. Plasma from the remaining Mik- negative blood donor cat did not react or only reacted weakly with red blood cells from over 30 cats.

Further evidence documenting the clinical relevance of the anti-Mik alloantibody was seen in a DSH, feline renal transplant candidate with blood type A and a prior transfusion history. Following an AB-matched blood transfusion, this cat experienced an acute hemolytic transfusion reaction. Subsequent crossmatching results revealed that patient plasma, obtained both pre- and post transfusion, was incompatible with the blood donor used, as well as with numerous other cats, but not with any of the three Mik red cell antigen-negative blood donors. Plasma from both blood donors with a documented anti-Mik alloantibody was crossmatch-compatible with this renal transplant patient, suggesting this patient's red blood cells also lack the Mik red cell antigen.

The absence of a Mik red cell antigen in some type A, DSH cats may result in formation of anti-Mik alloantibodies which can occur naturally. The presence of these alloantibodies may elicit an acute hemolytic transfusion reaction following an AB-matched blood transfusion. Additional studies to determine both the frequency of Mik red cell antigen-negative cats and the presence of anti-Mik alloantibodies in the general feline population are needed as is molecular characterization of the Mik-red cell antigen. Screening feline blood donors and patients for the presence of this apparently common red cell antigen and corresponding alloantibody may prove necessary in clinical practice.

Comparison of Glargine and Lente Insulins in Cats with Diabetes Mellitus

ACVIM 2005

K.E. Weaver; E.A. Rozanski; O. Mahony; D.L. Chan; L.M. Freeman
Tufts University School of Veterinary Medicine, North Grafton, MA

Most cats with diabetes mellitus require insulin therapy to prevent symptomatic hyperglycemia and the development of ketosis. Recently, the use of insulin Glargine (Lantus®) has been described in cats. Glargine is a genetically modified human recombinant insulin designed to function as a basal insulin with relatively peak-less activity and sustained duration of action. Glargine has been proposed as a once-daily insulin for cats, which would simplify treatment for clients. The two goals of this study were to 1) compare the efficacy of once-daily administered Glargine insulin to twice-daily administered Lente insulin in cats with diabetes mellitus and 2) to describe the effects of a commercial high protein, low carbohydrate diet designed for the management of feline diabetes mellitus (Purina DM®).

The study was a prospective randomized trial. Cats with naturally-occurring diabetes mellitus which was either newly diagnosed or poorly responsive to current therapy were eligible for study inclusion. Baseline testing included a physical examination, biochemistry profile, urinalysis and urine culture, T4 and fructosamine concentrations. Body weight and body condition scores (BCS, 1-9 scale) were recorded and a thorough dietary history was obtained. All owners were instructed to feed the cats the high protein, low carbohydrate diet exclusively once the study began. Cats were randomized to receive either Lente or Glargine insulin. The initial starting doses were 0.5 U/kg SQ Lente q 12 hours or 0.5 U/kg Glargine q 24 hours. Re-evaluations were performed on all cats at weeks 1, 2, 4, 8, and 12 and included an assessment of clinical signs (e.g., polyuria, polydipsia and appetite), physical examination including BCS and weight, a 16-hour blood glucose curve and fructosamine concentrations. Insulin dosage was increased, left the same, or decreased based upon the findings at each re-examination. Differences in the number of cats with effective control between the two insulin groups were compared using Chi Square analysis and glucose and fructosamine concentrations were compared using analysis of variance with repeated measures.

Seventeen cats were enrolled in the study. One cat was excluded (aggression, n=1), and three died during the study (diabetic ketoacidosis n=1; anesthetic accident n=1; cancer n=1). Therefore, 13 cats completed the 12-week study (Lente n=7; Lantus n=6). There was a significant improvement in the fructosamine and glucose concentrations in all cats but there was no significant difference between the two insulin groups. Four of the thirteen cats achieved complete remission, and two cats had a marked decrease in their insulin needs. There was no difference between insulin types in regard to treatment response or remission rates. The results of the study suggest that Glargine in combination with a high protein, low carbohydrate diet represents a viable option for treatment of feline diabetes mellitus.

Treatment with Glargine Results in Higher Remission Rates than Lente or Protamine Zinc Insulins in Newly Diagnosed Diabetic Cats

ACVIM 2005

R.D. Marshall^1,2; J.S. Rand^1
1Centre for Companion Animal Health, University of Queensland; ²Creek Rd Cat Clinic, Brisbane, Australia

Insulin glargine is a synthetic insulin analogue that is very long-acting in human patients. The aim of this study was to compare the effectiveness of glargine, PZI and Lente insulins in newly diagnosed diabetic cats.

Twenty-four newly-diagnosed diabetic cats (17m,7f) were treated either with glargine, PZI or Lente (n=8 for each group) and fed a very low carbohydrate-high protein diet (Purina DM canned). Insulin was initially given at 0.5U/kg BID S/C if blood glucose was >360mg/dl, and 0.25U/kg BID S/C if blood glucose was <360mg/dl. Insulin dose was then adjusted based on serial blood glucose curves and water intake. Cats were defined as achieving diabetic remission if normoglycemia was maintained without insulin therapy for more than two weeks.

At diagnosis, there was no statistical difference between treatment groups for age, body weight, body condition score, or concentrations of fructosamine, blood glucose, B-hydroxybutyrate or bicarbonate. Four of eight cats in each group were Burmese.

There was a non-significant trend for glargine treated cats to have lower 12-hour glucose concentrations after 10 and 17 days, than those treated with PZI or Lente. Mean 12-hour blood glucose at four weeks was significantly lower for glargine (239 ±61mg/dl) than PZI (389 ±20mg/dl) and Lente (410 ±38mg/dl) treated cats. Fructosamine concentration after four weeks of treatment was significantly lower than at diagnosis for glargine treated cats (343 ±38 and 553 ±21umol/l respectively) but not for PZI (469 ±42 and 570 ±22umol/l) or Lente (465 ±49 and 574 ±34umol/l).

All eight cats treated with glargine went into diabetic remission within four months of beginning treatment (mean=5.2 ±1.6 weeks, median=3 weeks), while three cats treated with PZI (mean=3.3 ±0.7 weeks) and two cats treated with Lente (mean=2 ±0 weeks) achieved diabetic remission. Of the seven glargine treated cats alive, six cats remain in remission at the time of publication (mean remission time=13 ±3.5 months, range=3-27 months). One of the remaining two cats alive treated with PZI (mean remission time=8.3+3.3 months, range=6-10months) and both cats treated with Lente (mean remission time=8 ±2 months, range=6-10 months) remain in remission.

Only one cat treated with glargine required an increase in insulin dose above 0.5U/kg BID, and seven of eight cats had their insulin dose reduced in the first three days of treatment. The mean dose of glargine at day 3 was 0.3 ±0.04U/kg BID. Clinical hypoglycemia occurred in two cats treated with Lente and one cat treated with PZI, but in none of the cats treated with glargine.

In conclusion, glargine was safe to use and resulted in higher remission rates than Lente or PZI insulins in newly diagnosed diabetic cats. It is postulated that improved glycemic control with glargine resulted in better reversal of B-cell glucose toxicity and higher diabetic remission rates.

Prevalence of FeLV and FIV in North America

ACVIM 2005

Julie K. Levy¹; P. Cynda Crawford¹; Jessica L. Brien^2
1College of Veterinary Medicine, University of Florida, Gainesville, FL; ²IDEXX Laboratories, Westbrook, ME

FeLV and FIV are among the most common infectious diseases of cats. Over the past 20 years, prevalence of FeLV has decreased, presumably as a result of widespread test and removal programs and immunization against FeLV. Testing for FIV is less common than for FeLV, and the recently introduced FIV vaccine is not widely used. Whether prevalence of FIV is changing is unknown. Because testing is voluntary and results are not collected into a central database, determining the true prevalence of FeLV and FIV is difficult. A large national study published more than a decade ago reported a prevalence of 13% for FeLV and 7% for FIV in 27,976 diseased and "high-risk" pet cats. In contrast, the prevalence of infection in 1,876 unowned feral cats was reported to be only 4% for each virus. Prevalence was lowest in healthy pet cats in which 1.3% of 1,763 cats recently studied were positive for FeLV and 0.9% of 1,757 cats were positive for FIV. The purpose of this study was to update prevalence data for FeLV and FIV infection in pet cats in North America, including cats of all ages and risk factors.

Veterinary clinics and animal shelters in the United States and Canada were recruited to test kittens and cats for FeLV and FIV using a point-of-care ELISA test (IDEXX SNAP Combo FeLV antigen/FIV antibody) during August to November 2004. Confirmatory tests were not performed as part of the study. Prevalence was calculated as the percent of positive tests in the study population for each virus. The Chi Square test was used to compare prevalence rates between regions. P < 0.05 was considered to be statistically significant.

A total of 18,038 cats were tested, of which 446 (2.5%) were positive for FIV and 409 (2.3%) were positive for FeLV. Of these, 58 (0.3%) were coinfected with both viruses. Prevalence was significantly higher in cats tested at veterinary clinics (3.1% FIV, 2.9% FeLV) than at shelters (1.7% FIV, 1.5% FeLV), in mature cats ( 4.1% FIV, 4.3% FeLV) than in juveniles (1.0% FIV, 1.4% FeLV), in males (3.6% FIV, 2.7% FeLV) than in females (1.4% FIV, 1.9% FeLV), and in diseased cats (6.1% FIV, 6.3% FeLV) than in healthy cats (1.8% FIV, 1.6% FeLV). Owned cats with access to outdoors had higher infection rates (4.3% FIV, 3.6% FeLV) than cats kept exclusively indoors (0.9% FIV, 1.5% FeLV). For shelter cats, the source (stray, relinquished pet, or feral) had no effect on FeLV infection rate, but feral cats had a higher rate of FIV infection (3.9%) than cats found as strays or relinquished by their owners.

Although the prevalence reported here is lower than in previous reports, it is not possible to compare them to assess changes in infection rates over time because of differences in the study populations. The previous studies each selected a single population for testing: high-risk cats, feral cats, or healthy pets, whereas the present study included cats of all ages, lifestyles, and health conditions, tested contemporaneously under similar conditions and season. Extrapolation of prevalence rates beyond the study population should be made with caution, since cats were not selected randomly from the overall population. Despite increased surveillance for FeLV and FIV and the availability of antiretroviral vaccines, infections with these viruses are common in North America.

Effect of Chronic FIV Infection, and Efficacy of Marbofloxacin Treatment, on Mycoplasma haemofelis Infection

ACVIM 2005

S. Tasker; S.M.A. Caney; M.J. Day; R.S. Dean; C.R. Helps; T.G. Knowles; P.J.P. Lait; M.D.G. Pinches; T.J. Gruffydd-Jones
School of Clinical Veterinary Science, University of Bristol, Bristol, N. Somerset, UK

The purpose of this study was to investigate the effect of chronic FIV infection, and efficacy of marbofloxacin treatment, on Mycoplasma haemofelis infection.

Twelve adult cats were used. Six were chronically infected with FIV-Glasgow 8 (Group C) and the other six cats were FIV-free (Group D). Groups C and D were housed separately for the duration of the study. All cats were infected with M. haemofelis on Day 0 of the study by intravenous inoculation of blood collected from a carrier cat. Over the course of the study from Day 0 until Day 86 post-infection (pi), blood samples were collected three times weekly for PCV and M. haemofelis quantitative real-time polymerase chain reaction (PCR), and once weekly for full hematological examination. FIV provirus quantitative real-time PCR was performed once weekly for Group C cats, and on Days-7 and 77 pi for Group D cats. FIV antibody ELISA testing was performed on all cats on Days-7 and 77 pi. On Day 16 pi three of the six cats in each of Groups C and D were randomly selected to receive marbofloxacin treatment (2 mg/kg PO SID) until Day 43 pi, with the remaining six cats acting as controls with no antibiotic treatment.

The M. haemofelis copy numbers and hematological data were compared between Groups C and D, and between marbofloxacin-treated and control cats, using a Mann-Whitney U Test. Significance was taken as a P value of < 0.05.

M. haemofelis infection was associated with the development of macrocytic hypochromic anaemia. Marked variation in M. haemofelis copy number was seen over time (> 5 log fold difference within 48 hours in some cats). Cycling of M. haemofelis copy number was also evident in some cats. All Group C cats were FIV antibody positive whilst those in Group D were negative. No obvious correlation was found between FIV provirus copy number and M. haemofelis copy number or hematological variables. No significant effect of chronic FIV infection on M. haemofelis copy number kinetics or hematological changes due to M. haemofelis infection, other than mean cell hemoglobin concentration (MCHC) (P=0.03), was found. Marbofloxacin treatment was associated with a significant decrease in M. haemofelis copy number (P=0.002), and negative PCR results were obtained at various time points in treated cats, although clearance of infection was not thought likely. Marbofloxacin treatment was also associated with a significant effect on MCHC (P=0.04), platelet count (P=0.03) and punctate reticulocyte count (P=0.03).

Chronic FIV infection had no significant effect on M. haemofelis infection kinetics or pathogenicity. Further studies are required to elucidate the reasons for the interesting variation in M. haemofelis copy number kinetics demonstrated in this study. Although marbofloxacin was associated with a significant reduction in M. haemofelis copy number, further studies are needed to determine an antibiotic treatment regime appropriate for clearance of M. haemofelis infection.

Survival of Mycoplasma haemofelis and 'Candidatus Mycoplasma haemominutum' in Blood of Cats Used For Transfusions

ACVIM 2005

A.T. Gary; H.L. Richmond; T.B. Hackett; M.R. Lappin
Colorado State University, Ft. Collins, CO

Blood transfusions are commonly administered to cats; associated risks include the transmission of various infectious diseases including M. haemofelis (Mhf) and 'Candidatus M. haemominutum' (Mhm). In previous experimental studies, both Mhf and Mhm were reliably transmitted by IV inoculation of as little as 1ml of heparinized blood. In a recent study, DNA of Mhf or Mhm were amplified from blood of 14 of 146 (9.6%) active feline blood donors. Blood transfusions in citrate-phosphate-dextrose-adenine (CPDA-1) solution are commonly administered immediately or stored for up to one month at 4°C prior to administration. It is unknown whether Mhf or Mhm survive in this solution or temperature. The purpose of this study is to determine if Mycoplasma spp. remain viable after storage in CPDA-1 for varying periods of time.

Because both organisms are directly associated with feline red blood cells and cannot be cultured, cats must be inoculated to document organism viability. A chronic carrier of Mhf, a chronic carrier of Mhm, and six SPF cats were used in this study. A CBC and a PCR assay capable of amplifying DNA of both organisms were performed twice in all eight cats prior to initiating the study. Blood (60ml) was then collected from each of the carrier cats, placed into a CPDA-1 solution containing bag (Teruflex®, Terumo Co., Tokyo, Japan), and the bags were stored at 4°C. At one hour, one week, and one month of storage, 2.2ml of blood was aseptically collected from each of the two bags. At each time, the Mycoplasma spp. PCR assay was performed on 200 µl of blood and the remaining 2ml were inoculated IV into a Mycoplasma-negative cat. After inoculation, 2ml of blood were collected from each cat for CBC and PCR assay weekly for four weeks.

Both chronic carrier cats were positive for Mhf or Mhm DNA throughout the study. All six SPF cats were negative for DNA of Mhf and Mhm before inoculation. DNA of Mhf or Mhm, respectively, was amplified from the CPDA-1 bags after one hour, one week, and one month of storage. The SPF cat administered Mhf containing blood after one hour of storage was PCR positive weeks 1-4, the SPF cat administered Mhm containing blood after one hour of storage was PCR positive weeks 1-3, and the SPF cat administered Mhm containing blood after one week of storage was PCR positive on week 1 after inoculation. Mycoplasma spp. DNA was never amplified from the SPF cat administered Mhf containing blood after one week of storage or the 2 SPF cats inoculated with blood stored for one month.

The results provide evidence that Mhf and Mhm can be transmitted to Mycoplasma spp. negative cats by administration of infected feline blood that has been stored in CPDA-1 solution for variable time periods. These findings support the recommendation that cats used as blood donors be screened for Mhf and Mhm infections by PCR assay prior to use.

Efficacy of Ronidazole in vitro and in vivo for Treatment of Feline Tritrichomonas foetus Infection

ACVIM 2005

Jody Gookin; Christina Copple; Mark Papich; Matthew Poore; Michael Levy
North Carolina State University, College of Veterinary Medicine, Raleigh, NC

The protozoan Tritrichomonas foetus (TF) is a prevalent cause of chronic large bowel diarrhea in cats, for which no effective treatment has been reported. Two nitroimidazole antimicrobials, tinidazole (TDZ) and ronidazole (RDZ) were tested for activity against feline TF in vitro at concentrations ranging from 0.01 to 10 µg/mL. TDZ is registered for use in people and RDZ is available outside the U.S. Both TDZ and RDZ killed TF in vitro. Ronidazole was selected for further in vitro study. Ten 10-week-old neutered female, TF-negative cats were individually housed, acclimated for three weeks and orogastrically infected with 3 x 10⁶ TF organisms. Physical exam findings, fecal consistency (formed, semi-formed, cow-pie, or liquid) and fecal examination results (direct smear, culture, & single-tube nested PCR) were recorded weekly by a blinded observer for 35 weeks.

All cats became TF positive by culture and developed loose feces by two weeks post-infection. Cats were randomized to two groups (n=5 each) and treated with placebo (dextrose) or RDZ (10 mg/kg orally, twice daily for two weeks). If cats in the treatment group had a relapsing infection or if they received a placebo, they were re-treated with RDZ at a higher dose of 30 mg/kg (n=3) or 50 mg/kg (n=7) orally twice daily for two weeks. Treatment with RDZ at 10mg/kg caused initial improvement, but in 5/5 cats there was a relapse infection at 2, 3, 3, 17 & 20-wks after treatment was completed. The TF isolated from these cats remained susceptible to RDZ. At 30mg/kg or 50mg/kg twice daily 9/10 cats have been cured after monitoring for 13 weeks and 4-6 weeks post-treatment, respectively. Spontaneous remission, adverse drug events or clinicopathological abnormalities were not noted. We concluded that oral administration of RDZ at 30-50mg/kg twice daily for two weeks is capable of resolving diarrhea and eradicating infection (on the basis of PCR) in cats infected with TF.

The Association of Bartonella Spp. Infection with Chronic Stomatitis in Cats

ACVIM 2005

K.L. Dowers; M.R. Lappin
Colorado State University, Fort Collins, CO

Stomatitis is a debilitating disease in cats that leads to oral pain, anorexia, weight loss, and occasionally euthanasia in intractable cases. The syndrome is thought to have multiple causes; recent work suggests that Bartonella spp. may play a role in some cases. Establishing a causative link between bartonellosis and stomatitis would justify routine testing of stomatitis cases and might suggest use of alternative antibiotic therapies, such as azithromycin. The objective of this clinical study was to determine the prevalence of Bartonella spp. DNA in blood and B. henselae serum antibodies in client-owned cats with histopathologically documented stomatitis as well as age- and geographically-matched healthy control cats.

Blood and serum samples from 34 affected cats and 34 age-matched healthy control cats were submitted by veterinarians from around the United States. DNA of Bartonella spp. was amplified from blood via a previously validated polymerase chain reaction (PCR) assay and serum antibody titers against B. henselae were determined by ELISA. All cats were tested for FeLV antigen and FIV antibodies. For cases where oral biopsy samples were obtained at the time of blood sampling, the PCR assay was also performed on tissue samples. Survey information regarding housing status (multiple or single cat households), previous FeLV and FIV testing, flea exposure, vaccination history and history of upper respiratory infections (URI) were collected for both affected and control cats, and lesion details and treatment trials were collected for affected cats.

No significant differences in the prevalence rates for PCR-positive cats between affected (8.89%) and control cats (8.89%) or for antibody-positive cats between the affected group (67.6%) and the control group (58.8%) were found. The only survey factor with significant correlation with stomatitis was history of URI [p<0.05]. Of the 18 oral tissue samples submitted, only 1 was PCR-positive.

This study underscores the difficulty of correlating Bartonella spp. test results with clinical disease in individual cats because of the high prevalence rates of antibody-positive animals within the healthy population, as reported in this and other studies. Treatment with anti-Bartonella spp. antibiotics may still be appropriate in refractory stomatitis cases, but a larger scale epidemiologic study should be conducted to further assess the usefulness of Bartonella spp. antibody and PCR testing of cats with chronic stomatitis.

The Association of Bartonella henselae Antibodies and Uveitis in Cats

ACVIM 2005

J.P. Fontenelle; A.E. Hill; C.C. Powell; M.R. Lappin
Department of Clinical Sciences, Colorado State University, Ft. Collins, CO

We first reported B. henselae associated chronic uveitis in a cat with B. henselae antibodies in serum, local ocular production of B. henselae antibodies, and response to administration of doxycycline. In a follow-up study, we amplified B. henselae DNA from aqueous humor and demonstrated ocular production of B. henselae IgM and IgG antibodies in some cats with uveitis, but not healthy cats.In one other clinical study, several additional cats with uveitis and B. henselae serum antibodies were reported. However, numbers of cats with proven B. henselae associated uveitis are small to date because of difficulties associated with making a definitive diagnosis. The objective of this study was to determine the prevalence of B. henselae antibodies in cats with and without uveitis.

In a separate study performed between January 1, 2003 and January 1, 2004, veterinary ophthalmologists were asked to submit samples from cats with endogenous uveitis for infectious disease testing. Cases were classified as idiopathic uveitis (n = 75) if serum was negative for T. gondii antibodies, FIV antibodies, and FeLV antigen and aqueous humor was negative for Toxoplasma gondii DNA and FHV-1 DNA. Cats that were positive in one or more of these tests were classified as non-idiopathic (n = 34). Two groups of control cats were used in the analysis. Control group 1 consisted of serum samples from 109 clinically ill cats that were sent for infectious disease testing during the same time period; samples were excluded if ocular abnormalities or abnormalities consistent with feline bartonellosis (fever, lymphadenopathy, stomatitis, or seizures) were mentioned. Control group 2 consisted of serum from 64 healthy cats. Age was recorded for each cat and the cat categorized as high or low risk for flea exposure by state based on previously published work. IgG antibodies against B. henselae were measured in all sera by a previously validated ELISA. Samples were categorized as being positive for B. henselae using cutoff points of >1:64 and >1:128. The association between Bartonella status and uveitis was analyzed using logistic regression in two separate analyses accounting for age and risk of flea exposure in the model. While serological evidence of exposure to B. henselae was common in cats with uveitis (53.2% at the 1:64 cutoff), results were not significantly different between any of the groups regardless of which serological cutoff (1:64 or 1:128) was utilized.

These results indicate that as with other causative agents of uveitis (e.g., T. gondii), the presence of serum antibodies to B. henselae may not correlate with the precise cause of the intraocular inflammation in individual cats. Further work is needed to determine optimal diagnostic tests for documentation of bartonellosis in cats.

Prevalence of Bartonella henselae Specific Antibodies in Serum of Cats with and without Central Nervous System Disease

ACVIM 2005

L.K. Pearce¹; S.V. Radecki²; M. Brewer¹; M.R. Lappin^1
1Colorado State University, Ft Collins, CO; ²Statistical consultant, Ft. Collins, CO

Bartonella henselae is the most common cause of cat scratch disease in children and young adults. While neurological complications are thought to be rare, a California study found Bartonella spp. to be the most common bacterial agents associated with encephalitis. Additional signs of neurological dysfunction include seizures, aggression, excitability, myelopathy, neuropathy and meningitis. In some cats experimentally infected with B. henselae, neurological signs including seizures, nystagmus, tremors, aggressive behavior and abnormal response to stimuli have been observed. The purpose of this study was to determine if differences exist in the seroprevalence of B. henselae antibodies in client-owned cats with and without neurological signs.

The Infectious Diseases Laboratory database was searched between January 2002 and May 2004 for feline serum sample submissions that listed neurological disease as a presenting complaint (Group 1). The neurological complaint was then classified into seizures or other neurological signs. Sera were sequentially selected from cats with a non-neurological presenting complaint (Group 2) and healthy cats (Group 3). Age, breed, sex, and state were recorded for each case. The state was used to classify each cat as high or low risk for Ctenocephalides felis exposure. Samples were stored at -20°C or -80°C until thawed and assayed in an ELISA for the detection of antibodies against B. henselae. Samples with a titer of > 1:64 were considered positive. Relationships between B. henselae serological status (positive or negative) and clinical presentations were assessed by Fisher's exact test; logistic regression was used to assess the influence of age and risk of flea exposure. Significance was defined as P < 0.05.

Of the 145 cats in Group 1, 63 cats had seizures and 82 cats had other neurological signs. There was no difference in B. henselae seroprevalence rates between cats with seizures (37 of 63 cats; 59%) and cats with other neurological diseases (35 of 82 cats; 43%). When the analysis was expanded to include age and risk of flea exposure by logistic regression, neither of these additional factors was significant and the effect of seropositivity was still not significant. Antibodies against B. henselae were detected in serum of 104 of 163 Group 2 cats (63.8%) and 68 of 97 of Group 3 cats (70.1%). In the first evaluation, Group 2 cats (P = 0.0153) and Group 3 cats (P = 0.0022) had a greater seropositive rate than Group 1 cats. When the analyses were expanded to include age and risk of flea exposure by logistic regression, neither of these additional factors influenced the seropositive rate within group and the effect of group was still significant.

In this study, cats with neurological disease were not more likely than other cats to have B. henselae antibodies. Presence of B. henselae antibodies in serum of individual cats with neurological disease does not prove the clinical signs are related to B. henselae.

Effects of a Single Dose of an Intranasal Feline Herpesvirus 1, Calicivirus, and Panleukopenia Vaccine on Clinical Signs and Virus Shedding After Challenge with Virulent Feline Herpesvirus 1

ACVIM 2005

M.R. Lappin¹; R.W. Sebring²; M. Porter²; S.J. Radecki²; J. Veir^1
1Colorado State University, Fort Collins, CO; ²Heska Corporation, Fort Collins, CO

Feline herpesvirus 1 (FHV-1) infection is very common in cats, is extremely contagious between cats, and frequently results in severe clinical disease. In a recent prevalence study at a humane society in north central Colorado, FHV-1 DNA was amplified from throat swabs or nasal discharges collected from 52 of the 61 cats (85.2%) tested. Thus, it is important to induce immunity in kittens by vaccination quickly, particularly in populations at high risk for exposure. The purpose of this study was to determine whether a single administration of commercially available intranasal FVRCP vaccine (Feline UltraNasal^TM FVRCP Vaccine, Heska Corporation, Fort Collins, CO) to kittens lessened clinical signs and FHV-1 viral shedding when compared to unvaccinated control kittens after FHV-1 challenge.

Three groups of 10 unvaccinated kittens were administered one dose of the vaccine two, four, or six days before challenge, respectively and one group was maintained as unvaccinated controls. FHV-1 challenge was then induced following USDA protocols and the kittens were observed for clinical signs of disease for 14 days. Throat swabs were collected from the group of kittens vaccinated on day -6 and the group of kittens used as controls on days -6, -3, 0, 4, 6, 8, and 12. Swabs were placed in sterile saline, incubated for two hours at room temperature, and frozen at -70°C until transported on dry ice for assay. FHV-1 and feline GAPDH DNA were amplified from each sample by use of a previously described fluorgenic PCR. Results for the fluorgenic assay for FHV-1 DNA were compared to cell count derived from a standard curve for GAPDH/cell to ensure an adequate cell count was present on the swab for analysis.

When vaccinated kitten results were compared to control kitten results, cats vaccinated six or four days prior to challenge had significantly lower clinical scores (P < 0.05) than control cats. FHV-1 shedding was lower in kittens vaccinated six days prior to challenge than in control cats on day 6 after challenge (P < 0.05).

Administration of this vaccine within several days prior to exposure lessened clinical signs of disease and FHV-1 shedding compared to unvaccinated cats.

Pharmacodynamic Data of Five Fluorinated Quinolones toward Canine and Feline Pathogens from Jan 1999 through June 2001

ACVIM 2005

D.M. Boothe¹; B.R. Simpson^2
1Auburn University, Auburn AL (Boothe); ²Texas A&M University, College Station, TX

This study prospectively compared in vitro resistance and potential efficacy of five fluorinated quinolones (FQ) toward aerobic canine or feline isolates (n=246) submitted to diagnostic laboratories. Microtube dilution was implemented at concentrations ranging from 0.03125 to 32 µg/ml. Drugs studied were (dose mg/kg) ciprofloxacin (CIP; 5-20), enrofloxacin (ENR; 5-20), difloxacin (DIF; 5-10), marbofloxacin (MAR; 2.5-5) and orbifloxacin (ORB; 2.5-7.5). Percent resistance of each organism to each FQ and pharmacodynamic indices were determined. Potency was measured as MIC₅₀ or MIC₉₀ and susceptibility as inhibitory quotient (IQ), defined as peak plasma drug concentration at the low or high dose of each FQ to MIC_mean (mean is the geometric mean). The organisms most commonly cultured and % resistant to all FQ were, for Gram negative (GN; n = 180; 20 ± 3%), Escherichia coli (n=61; 39±5%; at least 55% from the urine), Pseudomonas aeruginosa (n=58; 16 ±3%), and Gram positive (GP; n=66; 17 ±3%,), Staphylococcus sp (n=17; 2 ± 1%) including Staphylococcus intermedius (n=19; 6 ± 5%) and Streptococcocus (n=20; 17 ±10%). Percent resistance did not differ among the FQ for any organism. For GN, the MIC_mean and MIC_50, but not MIC₉₀, were lower than the breakpoint (MIC_BP) of each drug for all organisms. In general, potency for GN was ranked as CIP > ENR > MAR > DIF> ORB. However, the susceptibility (magnitude of IQ) was ENR > CIP > MAR > ORB > DIF. An IQ of 10 was achieved at the low dose only for ENR and MAR and only against Proteus. At the high dose, an IQ of 10 was achieved by ENR for all drugs and by DIF and ORB for none. For GP, potency was CIP > ENR=MAR > DIF > ORB. However, the magnitude of IQ were ENR > CIP > MAR >ORB> DIF. Only ENR was able to achieve an IQ of 10 for GP organisms although CIP could for S. intermedius. This study of a limited number of organisms demonstrates ENR is the FQ most able to achieve targeted IQ for concentration dependent drugs and that the high dose of any FQ may be the more prudent dose for susceptible organisms. The use of generic CIP may not meet judicious antimicrobial use guidelines.

Plasma Concentrations of Enrofloxacin in Cats after Transdermal Administration of a PLO Gel Formulation

ACVIM 2005

M. Karriker; V. Wiebe; K. Parsons; S. Stanley
Veterinary Medical Teaching Hospital, University of California, Davis, Davis, CA

Medications formulated in PLO (pluronic lecithin organogel) transdermal gels are compounded and marketed by pharmacies as alternatives to commercially available veterinary products for difficult to dose feline patients. As the popularity of this dosage form increases with both veterinarians and owners, there is growing concern regarding the efficacy of drugs administered transdermally. Previously published studies have shown that medications formulated in a PLO vehicle produce highly variable and often minimal drug levels in veterinary patients. This prospective study was designed to evaluate the serum enrofloxacin (ENRO) levels achieved after dosing healthy cats with ENRO formulated in a transdermal PLO gel.

Four healthy cats (mean age, 7.25 years) were dosed with 0.1ml of ENRO PLO gel 227mg/ml that was rubbed into the hairless, inner skin of the right ear of each cat once daily. In a crossover design, each cat received both a one dose and three dose regimen with a 3-month washout period. For both dosing regimens, a 1 ml whole blood sample was collected via venipuncture at times 0, 30, 60, and 120 minutes around the third or only dose depending on the regimen. Whole blood was centrifuged and the plasma was separated and frozen until analysis.

Plasma samples were assayed for enrofloxacin and ciprofloxacin simultaneously via reverse-phase high-pressure liquid chromatography with florescence detection. ENRO and ciprofloxacin (CIPRO) were extracted from plasma by use of a solid-phase extraction cartridge. Fifty microliters of each sample was separated using a 5µm, C18 reverse phase column with mobile phase of 90:10 water:acetonitrile with a column flow rate of 0.8ml/min. Plasma standard curves for both ENRO and CIPRO (concentrations 0.005, 0.05, 0.1, 0.5, 1, 10 ug/ml, correlation coefficient 0.998) were made for each assay using pooled feline plasma. The limit of quantification (LOQ) was 0.05µg/ml for both ENRO and CIPRO.

None of the samples analyzed from either dosing regimen showed ENRO or CIPRO levels above the limit of quantification. An independent analytical laboratory further verified the absence of quantifiable drug levels. It can be concluded from these results that neither ENRO nor CIPRO reached detectable levels (>0.05µg/ml) in plasma following transdermal administration in our sample population at the time points tested. As the LOQ was below the minimum inhibitory concentration (MIC) for most pathogens treated with ENRO, the results of this study suggest that a PLO formulation of may not adequately deliver ENRO across the membrane of the feline ear with enough efficiency to achieve plasma levels appropriate to treat any systemic pathogen. This study was not designed to explore alternate dosages or drug concentrations in other body fluids such as urine. Further controlled trials should be performed before administering ENRO, with the intent of achieving appropriate plasma levels, via this route.

Detection of Cryptosporidium Spp. in Feces of Cats and Dogs in the United States by PCR Assay and IFA

ACVIM 2005

A.V. Scorza; M.R. Lappin
From the Department of Clinical Sciences, Colorado State University, Ft. Collins, CO

In experimentally infected cats, we previously showed a polymerase chain reaction assay (PCR) to be more sensitive than a commercially available, monoclonal antibody-based immunofluorescence assay (IFA; Meridian Diagnostics, Cincinnati, Ohio) for the detection of Cryptosporidium spp. infection. The objective of this study was to compare results of the PCR assay and the IFA using feces from client-owned cats and dogs presented for evaluation of diarrhea.

The samples tested in this study are part of a ongoing study of feline and canine cryptosporidiosis in the United States and were collected between October 1999 and November 2004. Referring veterinarians were asked to send a representative fecal sample from cats and dogs with diarrhea to Colorado State University packaged with a cold pack by overnight express. Samples were refrigerated until processed for analysis in the Cryptosporidium PCR assay as previously reported and the IFA following manufacturer's recommendations. In this PCR assay, the primers used (awaF-995=5'-T A Γ A Γ A T T Γ Γ A Γ Γ T T Γ T T X X T-3' and awaR-1206=5' X T T X X A X X A A X T A A Γ A A X Γ Γ X X-3') amplify C. parvum and most of the C. felis and C. canis strains reported in Genbank.

A total of 292 samples were tested for Cryptosporidium spp. by PCR assay and IFA. The data outside and within parentheses are the number of positive samples or the percentage positive, respectively.

Total Samples (n = 292)		Canine Samples (n = 112)		Feline Samples (n = 180)
PCR	IFA	PCR	IFA	PCR	IFA
71 (24.3%)	8 (2.7%)	18 (15.1%)	2 (1.8%)	53 (29.4%)	6 (3.3%)

All IFA positive samples were concurrently positive by PCR assay. When compared to the results of the PCR assay, the sensitivity and specificity of the IFA were 11.3% and 100%, respectively.

These results document that Cryptosporidium spp. infection is common in client-owned cats and dogs with diarrhea in the United States and that the PCR assay utilized here is more sensitive than a commercially available IFA.

Systemic Endocrine Effects of an Inhalant Glucocorticoid in Healthy Cats

ACVIM 2005

L. Brownlee¹; B. Seguin¹; K. Decile¹; R.W. Nelson¹; L.J. Gershwin¹; C.R. Reinero^2
1University of California, Davis, CA; ²University of Missouri, Columbia

The mainstay of therapy for inflammatory bronchial disease in cats has traditionally been oral or injectable glucocorticoids. While as a species, cats tend to be comparatively resistant to developing side effects with chronic glucocorticoid therapy some cats do not tolerate this drug. Delivery of metered dose inhalant glucocorticoids (IGC) has revolutionized the treatment of human asthma by maximizing local efficacy and minimizing systemic bioavailability. The purpose of this study was to evaluate systemic endocrine effects of an IGC, an oral glucocorticoid (OGC), and placebo in healthy cats. We hypothesized that IGC would have minimal systemic effects on the hypothalamic-pituitary-adrenal axis (HPAA) in healthy pet cats.

A randomized crossover design was used in six clinically normal cats. Drugs administered included inhaled flunisolide (250mcg/puff administered BID), oral prednisone (10 mg/day) and placebo (an empty metered dose inhalant, 1 puff BID). All cats received each treatment for two weeks followed by a one-month washout period. Endocrine effects were evaluated using single early morning cortisol levels, urine cortisol:creatinine ratios (UC:Cr), and ACTH stimulation tests performed pre and post-drug administration for each treatment period.

The mean early morning serum cortisol concentration was lowest for the OGC treatment, but means were not significantly different between treatments: (placebo, 2.8±0.8 ug/dl; IGC, 2.4±0.7 ug/dl; OGC, 1.1±0.9 ug/dl). No significant differences for UC:Cr were noted across treatments. Responses to ACTH stimulation were below the normal range in all cats receiving IGC and in 2 cats receiving OGC. Mean ACTH stimulation peak cortisol levels were significantly lower post-treatment compared to pre-treatment levels in cats receiving IGC (pre, 8.5±2.2 ug/dl; post, 2.9±1.4 ug/dl, p=0.0004) but not OGC (pre, 8.0±2.5 ug/dl; post, 6.0±1.9 ug/dl, p=0.07). The pre-treatment ACTH stimulation cortisol levels did not significantly differ between IGC, OGC and placebo (p=0.59), indicating that the washout period was adequate.

In conclusion, suppression of the HPAA did occur with the relatively high dose of IGC chosen in this study. Surprisingly, the OGC dose used in this study did not cause significant endocrine suppression, although a Type II statistical error may have been committed because of the small sample size.

Semi-Quantitative Evaluation of Protein in Feline Urine

ACVIM 2005

H.M. Syme; J. Elliott
Royal Veterinary College, London, UK

The importance of protein in feline urine as evidence of significant kidney and other systemic diseases has been highlighted recently with survival of both azotemic and non-azotemic cats being inversely related to urine protein to creatinine ratio (UPC) at initial presentation. The aim of this study was to evaluate a semiquantitative highly sensitive assay of urine albumin concentration (E.R.D.-HealthScreen^TM Feline Urine Test, Heska Corporation.) and assess its performance against more quantitative measures of urine protein.

Urine samples were collected from 69 cats by cystocentesis during routine clinical evaluation. A complete routine urinalysis was undertaken. Samples were then diluted to a urine specific gravity of 1.010 with distilled water and urine albumin concentration was measured semiquantitatively by the E.R.D.-HealthScreen test. After screening by this method, 10 urine samples in the negative, low, moderate and high/very high positive categories were selected and aliquots were submitted to a clinical laboratory for UPC measurement. The same samples were also assayed for urinary albumin using a polyclonal ELISA previously validated for feline urine. A further 38 urine samples, where the UPC had been measured previously and the samples stored at -80°C, were selected to give samples with a wide range of protein concentrations. Aliquots of these samples were thawed and subjected to the E.R.D.-HealthScreen test. The median (25^th and 75^th percentile) UPC and urine albumin to creatinine ratios (UACs) for the four categories of proteinuria in the E.R.D.-HealthScreen test were calculated and statistical comparisons were made between the four categories using Kruskal Wallis with Dunn's Multiple comparison test.

The UPC and UAC values respectively found for each category of the E.R.D.-HealthScreen test were: 0.14 (0.10 to 0.21) and 19 (11-30) mg/g for negative tests (n=19); 0.23 (0.17 to 0.32) and 56 (33-81) mg/g for the low category (n=16); 0.23 (0.21 to 0.40) and 88 (70-180) mg/g for the moderate category (n=19) and 0.66 (0.44 to 1.11) and 260 (138-415) mg/g for the high category (n=22). Statistically, the negative category had significantly lower UPC and UAC results when compared to the moderate and high categories.

This study has defined the range of UPC and UAC values encompassed by the different categories on the E.R.D.-HealthScreen test. A negative result using the E.R.D.-HealthScreen test consistently gave UPCs below 0.25 and UACs below 58 mg/g. High positive results using the E.R.D.-HealthScreen test usually yielded UPC and UAC results outside of our laboratory normal reference ranges (>0.4 and >82 mg/g respectively).

Comparison of Urinary Albumin Excretion Normalized by Creatinine Concentration or Urine Specific Gravity

ACVIM 2005

H.M. Syme; J. Elliott
Royal Veterinary College, London, UK

Survival of both azotemic and non-azotemic cats is inversely related to urine albumin to creatinine ratio (UAC). Semi-quantitative test kits for the measurement of urine albumin concentration (E.R.D.-HealthScreen^TM Feline Urine Test, Heska Corporation) normalize the urine albumin concentration with urine specific gravity (USG) instead of urine creatinine concentration (UC). This approach has potential advantages in that it reduces expense, the volume of urine required and the sample turn-around time (if, as a result, samples do not need to be sent to an outside laboratory). The purpose of the present study was to determine if urinary albumin normalized to USG (UA/USG) predicts survival of cats as well as UAC does, and investigate the correlation between the two methods for normalization.

Cats were included in the study if they had been diagnosed with renal failure or were apparently healthy (on the basis of history, physical examination and routine biochemical analysis). In addition the following data had to be available from the medical record: the age of the cat, urine albumin and creatinine concentrations, USG, plasma creatinine concentration and survival time. Urinary albumin concentration was measured using a sandwich ELISA assay previously validated for feline urine. Survival analysis (time to death due to any cause) was performed by Cox's regression. Cases were censored if they were alive at the conclusion of the study or lost to follow up (LTFU). Variables that were included in the analyses were age, plasma creatinine concentration and either log UAC or log UA/USG (in separate models due to co-linearity of the variables). The measures of protein excretion were log transformed for statistical analysis. Pearson's correlation coefficient was calculated to describe the relation between the two measurements of albumin excretion.

Data were available from 124 cats, 54 of which died during the period of follow-up. Age (p=0.03) and plasma creatinine concentration (p<0.001) were predictive of survival in both models. Log UAC (p<0.001) and log UA/USG (p<0.001) were both highly predictive of reduced survival time. Correlation between UAC and UA/USG was strong (Pearson correlation coefficient 0.97, p<0.001).

Normalization of urine protein measurements to USG or to urine creatinine concentration yields similar prognostic information. The use of either method is acceptable for clinical practice.

Clinical Evaluation of Effects of Dietary Modification in Cats with Spontaneous Chronic Renal Failure

ACVIM 2005

S. Ross¹; C. Osborne¹; D. Polzin¹; S. Lowry²; C. Kirk³; L. Koehler^1
1College of Veterinary Medicine, University of Minnesota, St. Paul, MN;²Hill's Science and Technology Center, Topeka, KS; ³ College of Veterinary Medicine, The University of Tennessee, Knoxville, TN

A double-masked, controlled, randomized, clinical trial was designed to determine if a renal diet (modified in protein, phosphorous, sodium, and lipid composition) was superior to an adult maintenance diet in minimizing uremic episodes and mortality rate in cats with mild to moderate chronic renal failure.

Forty-five client owned cats were randomly assigned to a maintenance diet or a renal diet and evaluated tri-monthly for up to 24 months. Kaplan-Meier survival analyses were used to evaluate efficacy of the renal diet compared to the maintenance diet in minimizing uremia, renal-related mortality, and all causes of mortality.

Events	Renal diet (%)	Maintenance diet (%)	P value
Uremic crises	0/22 (0)	5/23 (22)	0.02
Renal cause mortality	0/22 (0)	4/23 (17)	0.03
All causes of mortality	3/22 (14)	9/23 (39)	0.06

Serum urea nitrogen concentrations were significantly lower and blood bicarbonate concentrations were significantly higher in the group fed the renal diet at baseline and during the 12- and 24-month intervals. Cats fed the maintenance diet had a significantly greater number of uremic episodes (22%) compared to cats fed the renal diet (0%). A significant reduction in renal-related mortality was observed in cats fed the renal diet.

The renal diet evaluated in this study was superior to an adult maintenance diet in minimizing uremic episodes and mortality rate in cats with mild to moderate spontaneous chronic renal failure.