Enrofloxacin-associated ocular disease in cats: A scoping review

Abstract

Background

Ocular effects, most notably retinal degeneration, have been linked to enrofloxacin use in cats. However, data have been limited and there is a need for formal evidence synthesis to better understand and characterize these potentially life-altering adverse events.

Objective

The objective was to describe data regarding ocular adverse events associated with enrofloxacin administration in cats and to identify information gaps.

Procedure

A scoping review was conducted, searching the MEDLINE (via the Ovid platform), Web of Science, and CAB Abstracts bibliographic databases.

Results

Six references passed title and abstract screening, then full text screening: 1 case series (n = 17 cases), 3 single case reports, 1 conference research abstract (n = 2 cases), and 1 pharmacovigilance summary publication. There were reports of enrofloxacin-associated adverse ocular events in 163 cats. Loss of vision, mydriasis, and altered pupillary light responses were most commonly reported. Increased tapetal reflectivity, retinal vessel attenuation, and retinal degeneration were the most common abnormalities on ophthalmological examination. Most cats had permanent blindness or altered vision. Most cats had received doses well in excess of the current label recommendation, but 15/103 (14%) cats for which dosing data were available were reported to have received ≤ 5 mg/kg per day.

Conclusion and clinical relevance

Suspected enrofloxacin-associated ocular disease in cats typically resulted in long-term blindness or visual deficits, but full recovery occurred in a subset. Inadequate data were available to assess potential risk factors (e.g., age, dose). Although a dose of ≤ 5 mg/kg per day will likely reduce the likelihood of adverse events, these data indicated that ocular disease was still possible at that dosage.

INTRODUCTION

Enrofloxacin is a fluoroquinolone antimicrobial that is licensed for use in various animal species, including cats. Although fluoroquinolones are higher-tier drugs (1) and not typically recommended as 1st-line treatments, their relatively broad spectrum activity, limited risk of adverse gastrointestinal effects, and once-daily dosing led to widespread use in companion animals (2–4). Originally licensed for use at a dosage of 2.5 mg/kg q12h, label recommendations were changed in July 1997, allowing for flexible dosing at 5 to 20 mg/kg q24h. However, adverse ocular effects, most notably retinal degeneration, were recognized soon thereafter (5,6). Blindness is also a recognized, but rare, adverse event in humans treated with fluoroquinolones (7).

Understanding adverse-event risks is an important component of effective antimicrobial stewardship, maximizing the benefits of antimicrobials while minimizing adverse events. Although adverse ocular effects have been sporadically described in cats, there is a need for formal evidence synthesis to better understand and characterize these potentially life-altering adverse events. This scoping review aimed to describe the current literature and knowledge gaps pertaining to ocular adverse events related to enrofloxacin treatment.

MATERIALS AND METHODS

A scoping review was designed based on the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) extension for scoping reviews (8). The full protocol is available at the University of Guelph’s (Guelph, Ontario) institutional repository, The Atrium (https://hdl.handle.net/10214/28718). The PRISMA Extension for Scoping Reviews Checklist is available in Appendix 1 (available online from: Supplementary Materials).

The scoping review was based on a population, concept, and context framework, including studies of cats in which adverse ocular events that occurred in response to enrofloxacin treatment were reported, in all settings. Controlled trials, case control studies, cohort studies, case series, and case reports were eligible for inclusion. Non-research conference outputs (e.g., proceedings) were excluded. Review articles in peer-reviewed journals were included for reference-list screening. No language or time limits were set.

Searches were conducted for CAB Abstracts, MEDLINE (via the Ovid platform), and Web of Science bibliographic databases. Search strategies are presented in Tables 1, 2, and 3. Reference lists of reviews and included manuscripts were searched manually to identify additional relevant citations.

TABLE 1.

Search strategy: Ovid MEDLINE.

All Ovid MEDLINE(R) <1946 to Present> Search date: December 9, 2024
Search #	Search query	Number of results
1	Cats/	181 281
2	Cat OR cats OR feline OR feline.mp	142 484
3	1 OR 2	224 447
4	Enrofloxacin.mp	3913
5	Enrofloxacin/ae, to [adverse effects, toxicity]	34
6	4 OR 5	3914
7	Retina OR retinal OR retinopathy OR blind* OR ocular OR eye OR visual.mp	1 604 909
8	Vision, ocular/	27 364
9	Retina/	84 993
10	7 OR 8 OR 9	1 276 682
11	3 AND 6 AND 10	23

TABLE 2.

Search strategy: CAB Abstracts, December 9, 2024. There were 40 references identified.

Search line	Search string/field
1	Cat OR cats OR feline OR felines OR felid/all fields
2	Enrofloxacin/all fields
3	Retina OR retinal OR retinopathy OR blind* OR ocular OR eye OR visual OR vision/all fields

TABLE 3.

Search strategy: Web of Science, December 9, 2024. There were 40 results.

Search line	Search string/field
1	Cat OR cats OR feline OR felines OR felid/all fields
2	Enrofloxacin/all fields
3	Retina OR retinal OR retinopathy OR blind* OR ocular OR eye OR visual OR vision/all fields

Results were imported into Covidence systematic review software (Veritas Health Innovation, Melbourne, Australia; www.covidence.org) and de-duplicated. Level 1 (title and abstract) and Level 2 (full text) screening were conducted by both authors.

Level 1 (abstract) screening criteria were as follows:

1. Does the title and/or abstract describe the occurrence, incidence, risk factors, or outcome for known or suspected ocular adverse events following enrofloxacin administration in cats?

2. Is the reference a controlled trial or observational study, conference research abstract, case report or case series, or review article in a peer-reviewed journal?

Level 2 (full text) screening criteria were as follows:

1. Is the full text available?

2. Is the reference a controlled trial or observational study, conference research abstract, case report or case series, or review article in a peer-reviewed journal?

3. Does the manuscript include one or more of enrofloxacin dosing, retinopathy incidence, risk factors, or outcome for known or suspected ocular adverse events following enrofloxacin administration in cats?

In the event of conflicts, the authors discussed the citations to come to consensus. Authors were blinded to their original assessment. A calibration step was undertaken after ~10% of the title/abstract search was completed by the 2 authors. If < 90% agreement had been achieved, the inclusion and exclusion criteria would have been evaluated by the authors.

Data extraction was done by 1 author using a pre-established collection form within Covidence. Collected data included (where available) age, sex, breed, enrofloxacin dose, frequency of administration, route of administration, total daily dose, comorbidities, reason for treatment, clinical and ophthalmological signs, and outcomes. Because this was a scoping review, critical appraisal of the individual sources of evidence was not done. The data were summarized and described.

RESULTS

After de-duplication, 68 references were screened. A total of 6 references passed Level 1 and Level 2 screening (Figure 1). These consisted of 1 case series (n = 17 cases), 3 single case reports, 1 conference research abstract (n = 2 cases), and 1 pharmacovigilance summary publication (Table 4).

FIGURE 1.

Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) diagram describing the selection of articles, inclusion and exclusion, and the final number of studies from which data were extracted.

TABLE 4.

Studies from which data were extracted for a scoping review of enrofloxacinassociated ocular disease in cats.

First author and year (reference)	Type	Number of cats	Country
Gelatt et al 2001 (6)	Case series	17	USA
Saroglu et al 2008 (18)	Case report	1	Türkiye
Sandmeyer et al 2008 (19)	Case report	1	Canada
Abrams-Ogg et al 2002 (11)	Case report	1	Canada
Kindler et al 2019 (20)	Abstract, case series	2	Germany
Hampshire et al 2004 (5)	Pharmacovigilance	141	USA

In total, there were reports of enrofloxacin-associated adverse ocular events in 163 cats. Most were from a pharmacovigilance study (5) that presented adverse drug events reported to the U.S. Food and Drug Administration, providing summary information for 141 cats affected between 1992 and 2002. The data bridged the period of label dose change in the United States. Reports before April 2000 covered a period in which the United States enrofloxacin label dose was 5 to 20 mg/kg. In July 2000, the manufacturer advised veterinarians of concerns about ocular disease at the upper end of the dosing range and recommended doses of no more than 5 mg/kg per day. After further U.S. Food and Drug Administration investigation, the manufacturer further advised veterinarians in March 2001 of the risks and new label dose.

Treatment information and dosing

Treatment regimen data for the 22 cats from case reports and case series are presented in Table 5. Dosing data were also provided for 71 cats in the pharmacovigilance study, resulting in dosing data for 93 cats. Fifteen of 93 (16%) cats had received ≤ 5 mg/kg per day, 50 (54%) had received 5.1 to 20 mg/kg per day, and 10 (11%) had received > 20 mg/kg per day. An additional 18 (19%) cats were reported to have received 5.6 to 34 mg/kg per day, but more refined dosing data were not provided. Additional details were available for 9/15 (69%) cats that had received ≤ 5 mg/kg per day. Six (67%) had concurrent diseases (e.g., diabetes mellitus, suspected toxoplasmosis) or risk factors (e.g., anesthesia) that could also have potentially caused vision abnormalities.

TABLE 5.

Metadata from cats with suspected enrofloxacinassociated ocular disease for which specific data were available (n = 22).

Variable	Result
Age	Median: 8 y (range: 2 to 16 y)
Sex	Male neutered: 9/20 (45%)
	Male intact: 4/20 (20%)
	Female spayed: 1/20 (5%)
	Female unspecified: 6/20 (30%)
Route of administration	IV: 3/22 (14%)
	IV then PO: 4/22 (18%)
	PO: 13/22 (59%)
	SC: 2/22 (9.1%)
Frequency of administration	Once daily: 4/22 (18%)
	Twice daily: 12/22 (55%)
	Variable, once to twice daily: 2/22 (9%)
	Twice daily, then unclear: 1 (4.5%)
	Once to twice daily, then twice daily: 1 (4.5%)
	Single dose: 1/22 (4.5%)
	Two doses, interval unclear: 1 (4.5%)
Total daily dosea	Median: 14 mg/kg per day (range: 4.6 to 54 mg/kg per day)
Duration of treatment	Median: 10 d (range: single dose to 120 d)
Maximum administered dose ≤ 5 mg/kg per day	4/22 (18%)
Maximum administered dose ≤ 10 mg/kg per day	8/22 (36%)
Outcomes Blindness:	9/22 (41%)
	Residual visual deficits: 8/22 (36%)
	Resolution: 3/22 (14%)
	Unknown/lost to follow-up: 2/22 (9%)

^aIf dosing regimens were changed during treatment, the largest daily dose was reported.

A total of 112/163 (69%) cats had received only oral enrofloxacin. Ten (6.1%) had received only injectable enrofloxacin and 11 (6.7%) had received both oral and injectable enrofloxacin. Whether the remaining 30 cats had received injectable only versus oral and injectable enrofloxacin was not reported.

Duration of treatment was not reported in the pharmacovigilance study. Duration data from the case reports and case series are presented in Table 5 and ranged from a single dose to 120 d. Fifteen (69%) cats had been treated for > 7 d.

Reasons for treatment

Reported reasons for enrofloxacin treatment were urinary tract disease (12/22, 55%), including urinary obstruction (n = 3, 14%); cystitis (n = 2, 9.0%); unspecified “genitourinary disease” (n = 2, 9.0%); otitis (n = 2, 9.0%); upper respiratory tract infection (n = 2, 9.0%); cystic calculi (n = 1, 4.5%); unspecified urinary disease (n = 1, 4.5%); unspecified renal disease (n = 1, 4.5%); renal failure (n = 1, 4.5%); suspected pyelonephritis (n = 1, 4.5%); and 1 (4.5%) each of dermatitis, diarrhea, pancreatitis, and pleural effusion. Reason for treatment was not reported for 2 (9.0%) cases.

Clinical signs

Individual cat-level data were retrieved from the case series, case reports, and research abstracts. Characteristics of the affected cats and treatment regimens are presented in Table 5. Loss of vision (complete or partial; 19/22, 86%), mydriasis (16/22, 73%) and reduced or absent pupillary light responses (12/22, 55%) were the most commonly described abnormalities. Descriptions of ophthalmological examination abnormalities included increased tapetal reflectivity (16/22, 73%), retinal vessel attenuation (16/22, 73%), and retinal degeneration (8/22, 36%). Because the level of detail for descriptions of clinical signs and lesions was variable, an absence of reporting of specific abnormalities did not necessarily mean they were not present.

Outcomes

Outcomes for cats described in case reports and case series are presented in Table 5. Three (14%) cats were reported to have had complete vision recovery. One of those had received a single dose of 5 mg/kg, SC; 1 received 2 doses of 5 mg/kg, SC; and 1 received doses of 10 mg/kg, IV, “once to twice daily for 5 d, then 5 mg/kg, PO, twice daily for 14 d.” Two of the cats were reported to have been blind at the time of initial ophthalmologic examination. Vision was not described for the other, but it had had slow pupillary light responses, increased tapetal reflectivity, and retinal vessel attenuation. The small number of cats with visual recovery precluded statistical analysis of factors associated with outcomes.

Outcome data were limited in the pharmacovigilance study. Two cats that each received 5 mg/kg per day regained vision within days of cessation of enrofloxacin treatment. One other cat that received 3 mg/kg per day and had no other identified cause of ocular disease was not completely blind but had residual visual deficits 1 mo later. No other outcome data were reported.

DISCUSSION

Retinopathy is a known complication of enrofloxacin treatment in cats, but its incidence and pathophysiology are unclear. Published data identified in this review do not provide any assessment of incidence, an important gap that hampers risk assessment. The pharmacovigilance study provided the largest number of cases, but data were inadequate to estimate either the numerators (not all adverse events are diagnosed and reported) or denominator (the number of cats treated with enrofloxacin is unknown). A review article (9) indicated an incidence of 1 per 122 414 cats treated but provided no citation. The same rate was stated in a letter to the Editor written by an employee of the manufacturer (10) in response to 1 of the papers cited here (11), also with no citation. This estimate was likely provided by the manufacturer based on the volume of drug that was sold; however, for the reasons stated herein, it was not likely based on adequate data to make a reasonable estimate of the incidence of disease. Volume of drug cannot be used to accurately estimate number of cats treated because of the wide ranges in dosing and treatment durations, as well as the availability of products used for both cats and dogs. The true incidence is presumably very low, but even a rough estimate is not possible with available data.

Risk factors for adverse retinal effects are similarly inadequately understood. A review (9) that evaluated the large case series reported here suggested that age may play a role, based on a statement (lacking supporting data or analysis) that older cats seemed to develop blindness at lower doses. A mechanism related to age-related alterations in drug elimination or metabolism (and therefore, higher drug levels) was suggested. Although this is plausible, the lack of controls in a case series or in the data from this scoping review precluded assessment of risk factors, and the age distribution in cases identified in this review did not suggest a prominent effect of age. This does not rule out an effect of age but provides no support for the assumption that age is an important risk factor. Limited age data were identified herein, but 6/20 (30%) cases with ages reported were in cats ≤ 5 y of age, and 11/20 (55%) were ≤ 8 y of age. No conclusions can be made without controls, but these data do not suggest a strong effect of age on retinopathy risk.

It has been suggested that “urinary tract infections with concomitant renal impairment” could predispose to adverse events through reduced elimination of the drug (9). However, a subsequent pharmacokinetic study of cats with varying degrees of renal function did not identify a significant effect of renal function on enrofloxacin clearance and suggested that dose adjustments are not indicated for azotemic cats (12). It has also been suggested that IV dosing might pose a higher risk because of rapid development of high systemic and retinal drug levels (9).

Inadequate data were available to assess the relative risk of oral versus parenteral administration. Most affected animals identified here had received oral enrofloxacin, but oral administration presumably accounts for the vast majority of enrofloxacin use. It is plausible that parenteral administration increases the risk, but this could not be investigated with the available data. Parenteral use of enrofloxacin is extra label in cats and the relative use of oral versus parenteral enrofloxacin in cats has not been reported. These data highlight that parenteral administration may be a risk factor but is not required for retinal toxicity to develop. The relative risk from different routes of administration and the risk associated with kidney disease and other comorbidities have not been adequately assessed. Such assessment is challenging for rare outcomes such as this, but multicenter case-control studies might help address these gaps.

Dose has been the main area of focus, on the assumption that complications are dose-dependent, not idiosyncratic. The incidence of adverse effects over time is impossible to assess with confidence because of issues with reporting bias, but there was an apparent increase in reported adverse events after flexible dosing was introduced that allowed for up to 20 mg/kg per day. Subsequent manufacturer safety data published in a freedom of information summary reported abnormal electroretinogram findings in 2/8 (25%) cats that received doses of 20 mg/kg, and severe fundic lesions and retinal degeneration in 5/8 (63%) cats that received 50 mg/kg (13). Because testing only included doses of 5, 20, and 50 mg/kg, potential effects within the flexible (5 to 20 mg/kg) range were not assessed.

An experimental study reported fundoscopic and electroretinogram abnormalities in cats that received doses of 50 mg/kg for 3, 5, or 7 d (14). Concerns about a dose-dependent effect led to changes in label recommendations to limit doses to 5 mg/kg per day. Herein, most affected cats had received doses > 10 mg/kg per day. Yet, although a dose of 5 mg/kg per day is currently recommended, this cannot guarantee no risk of retinal effects, since 16% of cats had received ≤ 5 mg/kg. This included 2 cats that had received only 1 or 2 doses of 5 mg/kg. An experimental study also reported “moderate retinal alterations” in 2/8 (25%) cats treated with 5 mg/kg per day for 10 d (15). This is in contrast to the statement on the commercial product label that no drug-related adverse events were identified in 124 cats that received 5 mg/kg per day as part of a clinical trial (16). The type (if any) of ocular monitoring was not noted and a citation was not provided for this statement. Thus, although it is likely that restricting dosing to 5 mg/kg per day will reduce the risk of retinal effects, this should be considered risk reduction, not risk elimination. Further study of the role of dosing is required to properly assess this area, given the data limitations identified in this scoping review.

Prognosis was variable. Full recovery occurred in a minority of cats. The number was too small for statistical analysis, but it is noteworthy that 2 of the 3 that recovered fully received 5 mg/kg — a single dose in one and 2 doses in the other. It is plausible that lower doses and shorter durations could be associated with a greater likelihood of recovery, but more data are needed to properly assess this. In a retrospective study of retinal degeneration, some cats regained vision but often had residual damage (5). However, the cited reference (9) did not discuss this.

This review focused only on enrofloxacin, not on other fluoroquinolones. A potential genetic basis for susceptibility to retinopathy has been identified in cats (17), and other veterinary fluoroquinolones are generally considered low-risk for retinal adverse effects in cats. The pharmacovigilance report cited herein (5) indicated that there were a small number of reports of blindness in animals treated with marbofloxacin or orbifloxacin, but that these ultimately were not attributed to the antimicrobials (5). Evaluation of other antimicrobials was not done as part of this scoping review, so the risks, if any, from alternative fluoroquinolones could not be assessed. Data were also limited to what was published. This may be of particular relevance for national adverse drug event reporting, in which data are collected by pharmacovigilance systems but not often published or readily available with enough detail to assess risk. It should also be recognized that dosing information was as reported in the studies. It is possible that misreporting or unidentified mis-dosing occurred.

Assessment of the appropriateness of enrofloxacin or need for any antimicrobial therapy was not possible given the limited case-level data. However, conditions that are often noninfectious (e.g., urinary obstruction, cystitis) were often reported. It was also unclear how often (if ever) enrofloxacin was indicated over lower-tier or 1st-line drug options. Further, although duration of treatment has not been well scrutinized in cats, the durations reported herein were often much longer than typically recommended. Although the roles of these individual factors have not been adequately investigated, long duration of treatment, higher-than-recommended dose, and unnecessary use of enrofloxacin all presumably increase the overall risk while conferring few or no health benefits. This highlights the importance of overall antimicrobial stewardship as a means of reducing all types of adverse drug event.

As discussed above, there were numerous gaps in the data. These included a small number of cases for which adequate details were present, variable information about dosing regimens, and limited information about potentially relevant comorbidities and risk factors. These cases represent only a fraction of the true events, which cannot be accurately quantified. Even if the true number could be identified, determination of rates would be challenging since there is no clear way to obtain denominator data. There is also not a standard case definition for enrofloxacin-associated retinopathy. Other limitations should be considered. This review did not include government pharmacovigilance reports. Government reports and websites were scanned but none was identified that reported any information apart from case numbers. These sources would not have provided any extra context and so were not included. There are presumably abundant data in government databases that could provide important information for assessment of the risks from enrofloxacin, but only if granular data are accessible or summaries are published. Since most of the reports included herein were case reports or case series, it is also expected that reporting would decrease over time (i.e., as the complication becomes better known and identified cases are not considered novel and therefore not prioritized for publication). This would also complicate consideration of temporal changes.

Adverse events are inherent risks of virtually any drug treatment, and risks versus benefits must be considered for any case. This review was not intended to indicate when, or if, enrofloxacin should be used in cats. It describes various components of a known adverse event, information that is important for clinicians who will ascertain their own cost-benefit assessments. It is reasonable to assume that the incidence of adverse ocular events is very low, and that the risk can potentially be mitigated by ensuring that the lower label dosing recommendations are followed and that short durations of treatment are used. However, the present data indicated that using a lower (5 mg/kg) dose should be considered a risk mitigation, not risk elimination, approach. The use of enrofloxacin must be assessed on a case-by-case basis, considering both the need for a fluoroquinolone and the risks and benefits of using enrofloxacin versus other licensed veterinary fluoroquinolones or other antimicrobials. CVJ