Patient population and source of injury
Seventy-seven patients (10 cats and 67 dogs) were divided into two categories and the end-results are summarized in Table 1. The source of the injury and which eye was injured are shown in Table 2. A cat claw was known or strongly suspected to be causative in 52 of the 77 patients (68%). In cat claw injury patients, right and left eyes were almost equal in number (29 vs. 33, respectively). The average dog’s age was 3.1 years and the median age was 1.3 years (range: 2 months–14.6 years) compared with the average and median age of cats, which were 5.1 and 5.4 years, respectively (range: 2 months–15.8 years). Thirty-five of the 77 patients (45%) were 12 months of age or younger, while 26 of the 77 (34%) were 5 months of age or younger at the time of injury. Two-thirds (45 of 67) of the dogs in this study were small breeds (weighing <20 pounds when full-grown), and 29 of the 67 (42%) patients were considered to be brachycephalic and/or exophthalmic breeds. The most common dog breeds were Chihuahua (6), Shih Tzu (6), Dachshund (5), Pitt Bull (4), Yorkshire Terrier (3), Boston Terrier (3), Lhasa Apso (3), Labrador Retriever (3), Pug (3), Terrier mix (3), Australian Shepherd (2), Toy Fox Terrier (2), and Chow Chow (2). Females were over-represented, as 56 of the 77 patients (73%) in this study were sexually intact or spayed females (P < 0.0001), and among the cat claw injury patients, 44 of the 62 (71%) were females (P = 0.0013).
Table 2. Source of injury and eye injured
|Source of injury||Canine||Feline||Total|
|Presumed cat claw||6||7||1|| ||14|
|Squirrel claw||1||1|| || ||2|
|Cactus needle|| || || ||2||2|
|Thorn||2|| || || ||2|
|Glass pane||1|| || || ||1|
|Rawhide chew toy|| ||1|| || ||1|
|Dog bite||1|| || || ||1|
Rationale for patient treatment
Thirteen of the 47 patients with a favorable prognosis were examined prior to July 2003, and 12 of the 13 had emergent, prophylactic lens removal based on previously described clinical recommendations for patients with this injury.4 The remaining 34 patients were examined after July 2003 and only three had lens removal surgery. The lens was considered irreversibly damaged in two of these three patients and the third developed a sterile lenticular abscess. Vision was lost in two of the three patients (endophthalmitis in one and retinal detachment/secondary glaucoma in the other) and both were enucleated.
In all patients in Treatment Group A, surgical repair of the corneal laceration was performed immediately prior to lens removal and was performed in all other patients (Treatment Group B) with poor apposition of the wound margins or when accompanied by iris prolapse and/or continued aqueous leakage (Fig. 1). Twenty-three of the 38 (61%) patients in this study examined acutely (during the first 72 h following injury) underwent corneal repair.
Figure 1. 3-month-old Dachshund Fe, squirrel claw injury OS; Group B. (a) T(time after injury) = <24 h. Following pharmacologic dilation dyscoria due to anterior synechiation was evident at the temporal margin of the 10 mm corneal laceration and the anterior chamber was shallow. Corneal wound apposition was poor and surgical repair of the cornea was performed. The lens capsule tear was estimated to be 10 mm in length. The lens underwent complete resorption within 1 month of surgery and glaucoma blinded the eye sometime between 3 and 26 months post-injury.
Download figure to PowerPoint
Patient categories and treatment groups
The 47 patients in Category 1 were subdivided into three treatment groups for comparison of the success or failure of prescribed surgical and/or medical treatment protocols. Group A consisted of 15 patients treated by corneal repair and lens removal. Group B contained nine patients treated by corneal repair without lens removal. Group C contained 23 patients treated by medical management. Table 3 shows a comparison of the average age, length of corneal laceration, length of lens capsule tear, and time from injury to referral for 47 patients in Category 1, subdivided according to last known success or failure. The average age of patients in Treatment Group A was 2.6 years (range 2 months–7.6 years), Group B was 5.9 years (range 2 months–15.8 years), and Group C was 2.4 years (range 2 months–11 years).
Table 3. Category 1 was composed of those patients deemed to have potential for vision and it was further subdivided into three treatment groups as follows: (A) corneal repair with lens removal, (B) corneal repair without lens removal, and (C) medical management. Successful treatment outcome was a patient with functional vision restored and having (i) a normal sized globe (relative to the fellow eye), (ii) a positive menace reflex, and (iii) a clear visual axis that allowed detailed funduscopic examination of >50% of the retina. Those patients suffering vision-threatening complications that resulted in blindness or enucleation were considered failures. The average patient’s age, the average length of the corneal laceration, the average length of lens capsule tear, and the average time from injury to referral were calculated for each group. Avg. = average (range; median) n = number of patients with data available
|Patients in treatment Groups A–C||Age||Length of corneal laceration||Length of lens capsule tear||Time from injury to referral (days)|
| n ||Avg. (months)|| n ||Avg. (mm)|| n ||Avg. (mm)|| n ||Avg. (days)|
|Success||31||38.2 (2–189; 10)||30||4.3 (2–10; 4)||27||6.5 (2–12; 7)||31||2.8 (0.5–14; 1)|
|Failure||16||36.8 (2–84; 30)||12||5.5 (2–12; 6)||8||8.6 (5–12; 9)||15||2.1 (0.5–10; 0.5)|
The 30 patients in Category 2 consisted of three types, which are as follows: 20 patients with an initial assessment of poor prognosis to restore vision (17 dogs and three cats), five patients with a traumatic cataract diagnosed as an old injury or an incidental finding in an eye with functional vision (four dogs and one cat), and five patients lost to follow up after the initial examination or surgery (four dogs and one cat).
Follow-up time periods of treatment groups
Postinjury follow-up time periods for Treatment Groups A–C were as follows: 1 month, 2–6 months, >6–12 months, >12–18 months, and >18 months; results are summarized in Table 4. Fifteen of the 15 patients in Group A (corneal repair/lens removal) were followed for 1 month, and 13 of the 15 (87%) had functional vision at the time of last examination. Six of 10 patients (60%) followed for 2–6 months had vision at the last examination. Three of the four patients had vision for period >6–12 months, three of the three patients had vision for the period >12 months–18 months, and one of the three patients (33%) followed longer than 18 months had functional vision at the last examination 90 months postinjury (average follow-up 45 months; range 18–90 months). Two patients had functional vision at 18 and 20 months postinjury, but failed to return for follow-up until glaucoma occurred at 72 and 90 months postinjury, respectively. Because these recheck times were clearly outliers among the study population and it was not possible to accurately determine the time for onset of glaucoma, both of these patients were censored at 18 months, the longest follow-up period considered in the analysis. Complications leading to blindness or enucleation included phthisis bulbi (4), secondary glaucoma (2), retinal detachment (2), and endophthalmitis (1). Two patients underwent lens replacement surgery several months after lens removal. Failure occurred in both patients because of phthisis bulbi in one patient (6 months postoperative) and glaucoma in the other patient (54 months postoperative; 72 months postinjury).
Table 4. Comparison of results in patients with differing follow-up periods. The number of patients in each successive follow-up time period declined because of blindness or patient lost to follow up. The proportion of patients experiencing vision loss at times >18 months in Groups A (67%) and B (100%) was greater than the proportion in Treatment Group C (0%) (P = 0.029 and P = 0.077, respectively)
|Treatment group (n = no. of patients)||Successful 1 month||Successful 2–6 months||Successful >6–12 months||Successful >12–18 months||Successful >18 months|
|(A) Corneal repair/Lens removal (n = 15)||13/15 (87%)||6/10 (60%)||3/4 (75%)||3/3 (100%)||1/3 (33%)|
|(B) Corneal repair (n = 9)||8/9 (89%)||1/2 (50%)||None examined||None examined||0/1 (0%)|
|(C) Medical management (n = 23)||23/23 (100%)||17/21 (81%)||15/15(100%)||12/12 (100%)||12/12 (100%)|
In Treatment Group B (corneal repair without lens removal), eight of the nine patients were followed for 1 month, and all had functional vision. Only one of the two patients followed for 2–6 months maintained functional vision at the last examination and the only patient followed longer than 18 months was blind at 26 months postinjury owing to glaucoma. The onset of glaucoma was not possible to accurately determine in this patient as examinations were performed at 3 months and 26 months following injury. None of the nine patients in this group were examined during the >6–12-month and >12–18-month follow-up time periods. The six patients in Group B judged to be successful at the last examination prior to discharge were all censored early (average follow-up 0.8 month; range 1 week–1 month). Three of the six patients censored in Group B were cats, and all feline patients in Category 1 were censored early (average censoring time 1 month; range 0.25–2 months). Two of the three owners of the three canine patients failed to return for re-examination and were censored early.
All 23 patients (100%) in Treatment Group C (medical management) followed for 1 month had functional vision at that time. Twenty-one patients were followed for 2–6 months and 17 (81%) were judged successful during this time period. Fifteen of the 15 patients (100%) followed for >6–12 months had functional vision at last examination, and 12 of the 12 patients followed for the >12–18-month and >18-month time periods maintained vision (average follow-up 31.8 months; range 18–63 months). Seven of the 12 patients followed >18 months were 5 months of age or less. Five of the 12 patients were over 5 months of age (average age 5.4 years; range 0.8–11 years). Two patients in Group C had extensive lens damage, and lens removal was recommended but not elected by the owner. Both patients were successfully managed with medical treatment. (Figs 2a–e and 4a–e). Three patients in this group did not receive the recommended aftercare in the 3-month period following injury and did not return for recommended follow-up examinations until vision-threatening complications developed 2–3 months after the initial examination. Complications leading to blindness or enucleation in the four failures included secondary glaucoma (2), endophthalmitis (1), and phthisis bulbi (1).
Figure 7 shows the Kaplan-Meier plot comparing the probability of maintaining functional vision in different treatment groups as a function of time following treatment. Treatment Groups A and B showed a greater rate of vision loss compared with Treatment Group C (P = 0.029 and P = 0.0097, respectively) that was most obvious >18 months postinjury.
Figure 7. Kaplan-Meier plot comparing the probability of patients maintaining functional vision in the three groups. Treatment Group A (corneal repair with lens removal) and Group B (corneal repair without lens removal) showed a significantly greater rate of vision loss compared to Group C (medical management) that was most obvious >18 months post-injury (P = 0.029 and P = 0.0097, respectively).
Download figure to PowerPoint
Across the 18-month postsurgical period in both dogs and cats, there was an average overall higher but nonsignificant rate of vision loss in the corneal repair group compared with the medically treated group (HR = 3.0, 95% CI = 0.5–17.9, P = 0.23) and a higher significant rate of vision loss in the lens removal group compared with the medically treated group (HR = 4.5, 95% CI = 1.2–16.7, P = 0.023). These average effects over time were more pronounced in dogs when comparing the corneal repair group with the medically treated group (HR = 13.9, 95% CI = 0.7–277.1, P = 0.085) and the lens removal group compared with the medically treated group (HR = 32.2, 95% CI = 3.7–280.4, P = 0.0017). In addition, however, there was statistical evidence that relative treatment effects on the rate of vision loss were not constant (nonproportional) over the 18-month follow-up period. In dogs and cats together, as well as dogs alone, the effects were most pronounced in the time early following surgery and gradually declined over the ensuing 18 months.
In the aforementioned multivariable models of dogs alone, corneal lacerations because of confirmed or suspected cat claw injuries were associated with a lower (but not significant) rate of vision loss compared with other causes of injuries (HR < 0.01, 95% CI = 0.0–4.2, P = 0.10). Increasing length of the corneal laceration was significantly associated with a lower rate of vision loss (HR = 0.12, 95% CI = 0.012–0.96, P = 0.046). An increasing time interval between injury and referral was significantly associated with a higher rate of vision loss (HR = 24.64, 95% CI = 1.16–524.4, P = 0.040). None of the interactions between treatment group and the other model variables (age, cat claw injury, length of corneal laceration, referral time) were significant. When cats were included in the analysis, findings were qualitatively similar but not significant for cat claw injuries (HR = 0.54, 95% CI = 0.056–5.22, P = 0.60, length of corneal laceration (HR = 0.70, 95% CI = 0.47–1.07, P = 0.098), and referral time (HR = 1.46, 95% CI = 0.90–2.38, P = 0.13).
Complications leading to failure
Complications resulting in blindness or enucleation were similar with or without lens removal. Seventy-two of the 77 patients were followed (five lost to follow up), and 36 (50%) patients were categorized as failures. The causes for failure were as follows: secondary glaucoma (11), endophthalmitis (9), phthisis bulbi (6), chronic uveitis/pupillary occlusion (6), retinal detachment (3), and mature cataract (1). These results are summarized in Table 6. The 20 patients that were determined to have a poor prognosis at the initial examination are included in this group. Thirty-six of the 72 (50%) patients had functional vision at last examination (29 phakic and seven aphakic; six lens removal; one complete lens resorption).
Table 6. Cause of failure
|Cause vision loss/eye removal||Corneal repair/lens removal||Corneal repair only||Medical management||Poor prognosis at initial examination||Total|
|Phthisis bulbi||4|| || ||2||6|
|Chronic uveitis/pupillary occlusion|| ||1|| ||5||6|
|Retinal detachment||2|| || ||1||3|
|Mature cataract|| ||1|| || ||1|