Abstract
The current recommended treatment for feline chlamydophilosis involves daily oral administration of antimicrobials to all cats within an affected group for a prolonged period of time (4–6 weeks). Not surprisingly, owner compliance can be poor resulting in apparent treatment failure. Recent anecdotal evidence, supported by its efficacy in the treatment of Chlamydia trachomatis infection in humans, has suggested that azithromycin may offer an alternative by allowing less frequent dosing for a shorter duration. A clinical trial was designed to evaluate the efficacy of azithromycin for the treatment of chlamydia (Chlamydophila felis) infection in cats. Whilst azithromycin, given at 10–15 mg/kg daily for 3 days and then twice weekly, provided a similar, rapid resolution of clinical signs and negative isolation scores as doxycycline, C felis was re-isolated in four out of the five cats treated. Furthermore, even daily administration of azithromycin to chronically infected cats was ineffective in clearing infection. The azithromycin protocols used here were therefore found to be unsuccessful in eliminating the carriage of this strain of C felis.
Introduction
The genus Chlamydia contains Gram-negative, obligate intracellular, bacteria with rigid cell walls. These bacteria have a tendency to produce chronic or relapsing infections, suggesting that the host immune response is only partially effective. Chlamydia psittaci has a number of different strains, infects a wide range of veterinary species and has a worldwide distribution. A single strain is thought to infect cats and has recently been renamed Chlamydophila felis (Everett et al 1999). C felis is recognised as primarily an ocular pathogen being responsible for up to 30% of cases of conjunctivitis in the U.K. (Wills et al 1988) and 16.5% of cases in Australia (Sykes et al 1999b). Serological surveys have suggested that between 5 and 10% of cats have been exposed to C felis (Lazarowicz 1977, Studdert et al 1981, Wills et al 1988, Gunn-Moore et al 1995). One Japanese study found prevalence rates of C felis antibodies as high as 45.5% in stray cats and 17.3% in pet cats (Yan et al 2000).
In untreated cats, excretion of C felis elementary bodies may persist for 8 months following infection despite signs of chronic conjunctivitis only being apparent for approximately 2 months (Wills 1986). Systemic therapy is usually required to eliminate shedding (Sparkes et al 1999). The simultaneous treatment of all cats within an infected group is advised as high rates of asymptomatic infection have been reported (Pointonet al 1991) resulting in cyclical re-infection in a group. Elimination of infection may also beimportant regarding human health, as it has been suggested that C felis could be transmitted to people (Yan et al 2000). Tetracyclines are the current treatment of choice and a prolonged course (4–6 weeks) is generally recommended to prevent recrudescence and subsequent shedding of infectious elementary bodies (Wills and Gaskell 1994, Greene 1998). In recent studies we have successfully eliminated C felis from experimentally infected cats using 3–4 weeks of doxycycline treatment at 10 mg/kg once daily (Sparkes et al 1999, Sturgess et al 2001). Sykes et al (1999a) have suggested that shorter treatment regimes than those currently recommended may be effective and found that 3 weeks of twice daily doxycycline at 5 mg/kg cleared infection from experimental cats challenged with an Australian strain of C felis. However, a number of side effects in cats treated with doxycycline have been reported. These include the potential for permanent discolouration of the teeth when used in young kittens and pregnant queens (Majcherczyk and Szymanska-Jachimczak 1965); hepatopathies (Kaufman and Greene 1993); and oesophagitis with possible subsequent oesophageal stricture. Oesophageal ulceration following oral doxycycline therapy is a recognised problem in man (Carlborg et al 1983) and has been reproduced experimentally in cats (Lanza 1988) and suspected in four clinical cases (Melendez et al 2000, McGrotty and Knottenbelt 2002). Furthermore, for clients with a colony of cats or individual cats that are difficult tomedicate, daily administration of medication for several weeks can be a problem leading to poor compliance and treatment failure.
Azithromycin is a macrolide antibiotic that is under investigation for use in veterinary medicine (Jordan 2001, Westfall et al 2001). In humans, azithromycin is used to treat Chlamydia trachomatis genital infections. It has a similar efficacy to doxycycline, but requires only a single dose rather than a 7-day course (Weber and Johnson 1995, Hillis et al 1998). Similarly, in cats, azithromycin has a long tissue half-life (Hunter et al 1995) suggesting that relatively infrequent dosing may be sufficient to maintain adequate plasma levels. For these reasons, azithromycin has been suggested as an alternative treatment for feline chlamydiosis (Ramsey 2000). There are anecdotal reports of its efficacy mainly from the U.S.A., and this information is readily available via the Internet.
The possibility of being able to effectively treat feline chlamydophilosis using a dosing regime that requires less frequent administration for a shorter duration offers significant advantages. A study was therefore designed to determine the efficacy of azithromycin for the treatment of C felis infection in cats.
Materials and methods
Nine barrier-maintained specific-pathogen-free cats (three entire females and six neutered males), 5–6 months of age, were housed in two groups. Prior to experimental challenge the cats were screened for evidence of C felis infection and shown to be seronegative (titre≤1:16) by indirectimmunofluorescence (Wills et al 1988). All cats were clinically examined immediately prior to challenge and were considered normal. Experimental challenge was performed on Day 0 by bilateral instillation of 2.3×103 infectious units of a field isolate of C felis (K2487) made up to a volume of 25 μl in phosphate buffered saline (PBS, 0.01 M, pH 7.4).
By Day 4, all cats had developed significant clinical signs and were randomly allocated to three treatment groups. Two cats served as untreated controls (group U) whilst two treated control cats (group D) received doxycycline tablets (Ronaxan; Merial) at 10–15 mg/kg once daily. The remaining five cats (group A) were given azithromycin suspension (Zithromax; Pfizer), at 10–15 mg/kg once daily for 3 days and thereafter twice weekly. The cats were housed in the two rooms available such that each room contained one cat from group U, one from group D and either two or three cats from group A.
Medications were stored at room temperature (below 25°C) and azithromycin was used within the recommended time once reconstituted. All cats were starved prior to treatment and then fed shortly afterwards. Treatment was administered over a period of 25 days. The dose rate for both doxycycline and azithromycin was calculated to the nearest 10 mg (half a tablet of doxycyclineor 0.25 ml of azithromycin suspension) above 10 mg/kg. Cats were weighed once weekly and treatment dosage was altered as necessary.
The day after the initial treatment had ceased (Day 29), the two group U cats were isolated and the other cats were mixed. The untreated control cats were then given azithromycin at 10–15 mg/kg once daily for 25 days as a model for the treatment of chronic chlamydiosis (Table 1). Four of the five group A cats (C5, C6, C7, C8) also received daily azithromycin for a 25 day period following a return of clinical disease from Day 38. All cats in groups A and U received a 4 week course of doxycycline at 10–15 mg/kg once daily at the end of the trial in order to eradicate infection. For logistical reasons, medication was only given for five consecutive days each week.
Days when Chlamydophila felis was isolated from each individual cat and when treatments were given
00072-X/unzip/10.1016_S1098-612X(03)00072-X-table1.png)
■ Positive C felis isolation
▩ Daily azithromycin (10–15 mg/kg)
▥ Daily doxycycline (10–15 mg/kg)
▧ Twice weekly azithromycin (10–15 mg/kg)
Clinical scoring and collection of conjunctival swabs for C felis isolation was performed on Day 2, Days 4–8 inclusive then twice weekly for 5 weeks and once weekly for the remaining 3 weeks of the trial, which ended on Day 69(Table 1). Clinical examination and scoring was carried out by the same investigator (DAH) who was blinded to the cat's treatment group. Fifteen specific ocular and systemic clinical signs were evaluated using a system described by Sparkeset al (1999). Each sign was given a score from 0 (normal) to 3 (very severe) in half units and these scores were summed to give an overall clinical score between 0 and 45. Cats were clinically monitored for 6 months after the end of the trial (Day 69) for any recurrence of signs. Since swabbing may affect the clinical score of the eye used (Wills et al 1987), the left eye was clinically evaluated whilst the right eye was swabbed. Conjunctival swabs were placed in virus/Chlamydia transport medium (Spencer and Johnson 1983) prior to being frozen at −80°C until isolation was performed.
The technique for chlamydial isolation has been previously described (Wills et al 1987, O'Dair et al 1994). Absolute numbers of infected cells per coverslip were not determined but a score of 0–4 was assigned giving an indication of the number of organisms cultured (Sturgess et al 2001).
Results
Clinical assessment
No clinical evidence of adverse reactions associated with either doxycycline or azithromycin was identified in any of the cats at the dose rates given. The individual clinical scores of the cats ranged from 0.5 to 2.0 on Day 0, prior to challenge. Following infection, clinical signs primarily consisting of conjunctival hyperaemia, chemosis, mucopurulent ocular discharge and submandibular lymphadenopathy, typical of feline chlamydiosis (Wills et al 1987, O'Dair et al 1994), were observed in all cats. Enlargement of other peripheral lymph nodes and occasional nasal discharge also occurred in some cats during the acute stage of infection. Respiratory signs other than nasal discharge were not observed. Cats were not found to be depressed or lethargic during the acute phase but depression or lethargy was recorded in both cats in group U on Day 14 and two cats in group A on Day 35. The two cats in group A became pyrexic on Day 35 with rectal temperatures of 40.2°C (C6) and 40.8°C (C7). For cats in groups U and D, changes in clinical scores over time followed the previously reported patterns (Sparkes et al 1999, Sturgess et al 2001) (Fig 1). Group U scores remained high until Day 14 from which time a gradual decline was seen whilst group D cats showed a rapid response to treatment with scores decreasing sharply from Day 6 and remaining low for the duration of the trial. Although the size of groups does not allow statistical evaluation, the general trend for group A was initially similar to that for group D, with a rapid response to treatment and fall in clinical scores from Day 5. However, from Day 31, the clinical scores for group A tended to be higher than those for group D. Treatment of group U with daily azithromycin from Day 29 led to a clinical improvement but this was not maintained as the scores started to gradually increase onDay 38. Clinical signs were not reported in groups U and A following doxycycline treatment at the end of the trial, except in one group A cat (C5). This cat displayed a third episode of clinical signs 12 days after finishing the doxycycline course. C5 was then isolated and treated with daily doxycycline for 5 weeks, resulting in a sustained recovery from infection. None of the cats displayed further clinical signs during the 6-month observation period following the trial.
00072-X/unzip/10.1016_S1098-612X(03)00072-X-fig1.png)
Mean clinical scores of each treatment group over time. The treatments indicated in the legend were given up to and including Day 28; for further treatments see Table 1.
Isolation of C felis
C felis was isolated from two cats on Day 2 post-infection and from all cats (with similar mean isolation scores between groups) on Day 4, when treatment was commenced.
In common with the clinical scoring results, the pattern of isolation scores for groups U and D followed the expected trend (Sparkes et al 1999, Sturgess et al 2001) (Fig 2). C felis was isolated on each occasion from cats in group U until Day 31 (Table 1) but positive isolation ceased following daily azithromycin treatment. However, this initial response to treatment was not maintained and C felis was isolated from both cats 14 days into treatment. Response to doxycycline was rapid and group D scores fell to zero by Day 7 and swabs remained negative for the duration of the study. Isolation scores in the azithromycin-treated group initially followed the same trend with negative swabs from all cats on Day 6. Swabs from only one cat (C9) in group A remained negative on isolation from Day 6. C felis was repeatedly isolated from one cat (C5) in group A from Day 11 (Table 1).
00072-X/unzip/10.1016_S1098-612X(03)00072-X-fig2.png)
Mean isolation scores of each treatment group over time. The treatments indicated in the legend were given up to and including Day 28; for further treatments see Table 1.
Discussion
C felis infection is more common in cats in the age group 5 weeks to 9 months (Wills et al 1988, Sykes et al 1999b); hence young cats (5–6 months old) were used in this study. Due to logistical reasons, the cats were not swabbed prior to challenge and therefore their seronegative results and specific-pathogen-free status were assumed to be sufficient evidence of absence of C felis infection before the trial commenced. Active infection was confirmed in all cats by observation of typical clinical signs, positive isolation scores and persistence of infection in the untreated controls. Treatment was instituted on Day 4 compared to Day 7 in previous studies (Sparkes et al 1999, Sturgess et al 2001), as all cats showed significant clinical signs by this time. This difference is most likely to reflect a variation in the pathogenicity of the field isolate used. However, the use of different C felis strains for the induction of experimental infection has not affected the overall pattern of clinical and isolation scores in previous untreated cats studies (O'Dair et al 1994, Sturgess et al 1995, 2001, Sparkes et al 1999).
We have conducted a number of studies using the model described above that have produced a predictable and reproducible infection. The number of control cats was therefore kept to the minimum required to establish that this study followed a similar pattern of infection and response to treatment. In the first phase of the study, cats were not isolated in order to mimic a colony situation with the risk of cyclic infection. Isolation of young cats for a prolonged period also has significant welfare implications. Despite being housed together, shedding of C felis by cats in the azithromycin-treated group was not associated with recrudescence of clinical signs in the two doxycycline-treated cats. This is an interesting observation as it implies that once an individual has eliminated the infection, either immunity to re-infection (in the short-term) is complete or spread between in-contact cats is inefficient. However, further studies involving larger numbers of in-contact cats would be required forconfirmation of this finding.
C felis shedding recurred in one group A cat whilst receiving twice weekly azithromycintreatment. Two other cats in group A developed lethargy and pyrexia on Day 35. These signs preceded increases in clinical scores in these cats and were probably an indication of relapse. C felis was also isolated from conjunctival swabs taken from group U and A cats during daily azithromycin treatment in the chronic stages of infection. As azithromycin treatment, even on a daily basis, was proving ineffective, a course of daily doxycycline was required to clear infection at the end of the trial. However, the doxycycline was only administered on 5 days of the week and this may have contributed to one of the group A cats showing a further recurrence of disease. This case suggests that owner compliance may be a significant factor in treatment failure in cats infected with C felis.
It has been previously suggested that the correlation between clinical signs and C felis isolation in individual cats is poor, especially in chronic infection (Sturgess et al 2001). This observation is also supported by this study (Figs 1 and 2) but it is still unclear whether this represents a failure to identify the presence of infection or an immune-mediated component to the clinical signs that is not dependent on active infection. A more sensitive test, such as the polymerase chain reaction (PCR), for detection of C felis, may be useful to identify cats that are shedding low levels of the organism (Sykes et al 1999a). However, non-viable agents may also yield a positive PCR result (McDonald et al 1998).
A variety of therapeutic regimes have been suggested for azithromycin treatment of feline infections. Ramsey (2000) has suggested a dose rate of 7–10 mg/kg once daily for 14 days, then 5 mg/kg once daily for 7 days, followed by 5 mg/kg every other day for 14 days for the treatment of chlamydophilosis. Higher oral doses (15 mg/kg twice daily for 7 days) have been used in the management of Haemobartonella felis (Mycoplasma hemofelis) infection (Westfall et al 2001). In this study, azithromycin was given every 24 h for the initial 3 days to act as a loading dose increasing the tissue:plasma ratio and half-life (Jordan 2001). Due to the long tissue half-life, twice weekly medication was then felt to be appropriate (Hunter et al 1995). The possibility of infrequent dosing was considered to be the main potential benefit of azithromycin treatment, facilitating improved owner compliance.
Azithromycin is generally considered to bebacteriostatic but may be bactericidal at higher concentrations against certain pathogens (Jordan 2001). The inability of the azithromycin treatment used in this study to eliminate shedding of C felis might have been due to insufficient concentrations of the drug in the tissue of the eye or possibly a bacteriostatic action associated with a partially effective immune response. Hunter et al (1995) detected a concentration of <1 μg/g in ocular tissue 24–72 h after a single oral administration of 5 mg/kg azithromycin to cats. A minimum inhibitory concentration of around 4 μg/ml has been suggested to be necessary for Gram-negative pathogens such as C felis. Whether this was achieved by the higher dose rate and repeated administration used in this study is unknown.
When interpreting the results of this study, the effect of variation in C felis strain pathogenicity should be considered. A difference in strain virulence may explain why Sparkes et al (1999) found that cats remained positive by isolation for a median of 3 days after starting doxycycline treatment, whilst Sykes et al (1999a) obtained positive isolation results, using an Australian challenge strain, for only 1 day after treatment had commenced. The results from this study shouldideally be confirmed using other strains and with a larger number of untreated and doxycycline-treated control cats.
The azithromycin treatment protocols used were only successful in eliminating this strain of C felis from one out of five cats. Shedding of C felis following treatment is a major concern when treating an outbreak in a multi-cat establishment, such as a breeding colony, in order to avoid cycles of re-infection. Therefore, we suggest that azithromycin (an unlicensed product) may not offer an advantage over doxycycline in the management of feline chlamydophilosis.
Acknowledgements
The authors would like to thank the feline diagnostic service for performing the C felis isolations, the small animal technicians for looking after the cats, and Fort Dodge Animal Health for providing a grant to support WMA Owen.