- Top of page
- Materials and Methods
Objective To evaluate the effectiveness of topical nalbuphine or oral tramadol in the treatment of corneal pain in dogs.
Animals studied Fourteen male Beagle dogs.
Procedures Dogs were divided into three treatment groups and sedated with dexmedetomidine (5 μ/kg IV). A 4 mm corneal epithelial wound was created in the right eye (OD) of all dogs. Sedation was reversed with atipamazole IM. All dogs received pre/post ophthalmic examinations. Post operatively, Group NB (n = 5) received topical 1% preservative-free nalbuphine OD q8 h and an oral placebo PO q8 h. Group TR (n = 5) received tramadol (4 mg/kg) PO q8 h and topical sterile saline OD q8 h. Group CNTRL (n = 4) received topical sterile saline OD q8 h and an oral placebo q8 h. All dogs received topical 0.3% gentamicin OD TID until healed. Dogs were pain scored using a pain scoring system modified from the University of Melbourne pain scale at 0, 1, 2, 4, and 6 h, then every 6 h by observers masked to treatment, until corneal wounds were healed. Treatment failure was recorded if cumulative pain scores were above a minimum threshold of acceptable pain and rescue analgesia of morphine (1.0 mg/kg IM) was administered subsequently.
Result Four dogs in Group NB, one dog in Group TR, and two dogs in Group CNTRL required rescue analgesia. There was no significant difference in the incidence of treatment failure between groups (P = 0.184). Mean time to rescue was 9.16 h. All corneal wounds were healed by 84 h.
Conclusions The results of this study suggest tramadol rather than nalbuphine should be further investigated for the treatment of corneal pain.
- Top of page
- Materials and Methods
Pain is commonly associated with a variety of ocular conditions. Pain may be one of the first symptoms noted by owners and may provide the impetus for their visit to the veterinarian. A recent survey of over 450 small animal owners indicated that the experience of pain was one of the highest concerns they had for their pet.1 The superficial cornea in the dog is richly innervated by sensory nerves and corneal damage typically results in signs associated with pain.2,3 These signs may include epiphora, conjunctival hyperemia, blepharospasm, reflex uveitis, self-trauma, and other signs consistent with discomfort. Corneal conditions such as spontaneous chronic corneal epithelial defects (SCCED; also referred to as indolent erosions and recurrent erosions) can cause significant pain for the patient. Treatment of SCCEDs includes epithelial debridement, linear grid keratotomy, multiple punctate keratotomy, and superficial keratectomy among others.4 These treatment modalities may also cause pain to the patient. Consideration of pain and its control is an important part of the overall therapy of ocular disease in many species.
Limited options have been available in the past for treatment of pain associated with corneal wounds. Treatment has included the use of topical anesthetics, topical or systemic non-steroidal anti-inflammatory drugs (NSAIDs) and analgesics. Topical anesthetics, such as proparacaine, provide good short acting corneal anesthesia by blocking action potentials in the sensory nerve membrane in dogs,5,6 cats,7 and horses.8 Chronic use of proparacaine, however, has been shown to cause delayed corneal healing and ulceration due to its epitheliotoxic effects, therefore it should be used infrequently for short duration examination and diagnostic procedures.6,9,10 NSAIDs have been used to treat ocular pain in dogs,11 although no controlled pain studies have been reported. Applied topically, NSAIDs have toxic effects on the corneal epithelial cells12 and have been sporadically associated with corneal melting.12,13 Systemic NSAIDs can provide control of ocular inflammation but are not without side effects such as gastric ulceration and renal toxicity which could limit their use in certain patients.11 Use of opioids is an option for treatment of pain due to corneal wounds. One of these, morphine sulfate, is often used systemically to control pain but has the potential for undesirable side effects. Topical use of morphine has also been shown to control pain associated with corneal wounds in rabbits,14 rats,15 and dogs16 without systemic side effects. Additionally, morphine did not cause a delay in corneal wound healing when used topically.16 Recently, oral tramadol and topical nalbuphine have seen increased use as alternatives for pain control in dogs. These drugs offer an interesting alternative to morphine due to their analgesic effects and ease of use as nonscheduled drugs.
Tramadol is a synthetic codeine analog that produces analgesia via both opioid and nonopioid mechanisms.17 Tramadol is a pro drug and is metabolized in the liver to produce an active metabolite, o-desmethyltramadol (M1).18,19 Tramadol has opiate activity on μ-, κ-, and δ-receptors, causes α-2 adrenergic stimulation, inhibits the reuptake of norepinephrine and serotonin, and enhances serotonin release.20 Oral tramadol has been shown to have good efficacy in humans21 and cats,22 however, controlled studies in dogs are lacking. There have been several studies that have described the use and efficacy of tramadol in its injectable form in the treatment of pain in humans,23,24 dogs,25,26 and horses.27 The oral dosage of tramadol in dogs has become accepted as 2–4 mg/kg BID–TID.28
Nalbuphine is a potent synthetic mixed opiate having agonist activity (analgesia) at κ-receptors and antagonist activity at μ-receptors.29–31 Nalbuphine, like tramadol, has been shown to have low side effects and dependence profile in humans.32,33 Nalbuphine was recently evaluated as a preoperative pain medication in dogs undergoing ovariohysterectomy25 or castration and found to provide improved pain control when compared to controls but less pain control compared to preoperative morphine and postoperative ketoprofen.34 Additionally, recent studies have investigated the effects of topically applied nalbuphine on corneal sensitivity, when measured with a Cochet-Bonnet esthesiometer, in normal dogs35 and horses.36 In dogs, topical nalbuphine was found to decrease corneal sensitivity, however, in horses, it had no significant effect.35,36
The purpose of this study was to evaluate the efficacy of topical nalbuphine and oral tramadol in the treatment of corneal pain in dogs. This is, to our knowledge, the first study to evaluate the effects of oral tramadol and topical nalbuphine on pain in dogs with experimental corneal wounds.
Materials and Methods
- Top of page
- Materials and Methods
Fourteen intact male Beagle dogs, weighing 11.42 ± 1.34 kg (mean ± SD) and aged 6.0 ± 0.1 years, were used in this study. All dogs underwent preoperative ophthalmic examinations including slit-lamp biomicroscopy and indirect ophthalmoscopy. Applanation tonometry (Medtronics, Minneapolis, MN, USA) and Schirmer tear tests (Alcon, Chicago, IL, USA) were within normal reference ranges on all dogs. No fluorescein dye was retained in either eye of all animals. Dogs were housed in individual kennel runs, given free access to water, and provided food twice daily. The pain scoring system was based on six different categories (Table 1) and was performed by the same trained observer (JSC) throughout the study period. The study protocol was approved by the University of Wisconsin Animal Care and Use Committee and conformed to the Association of Research in Vision and Ophthalmology’s Statement for the Use of Animals in Ophthalmic and Vision Research.
Table 1. Subjective pain score system (modified from the University of Melbourne Pain Scale) used to assess the analgesic effects of topical nalbuphine and oral tramadol in dogs with experimentally created corneal wounds
|Comfort||0||Patient asleep, calm, or awake, interested in surroundings|
|1||Mild agitation or depressed, uninterested in surroundings|
|2||Moderate agitation, restless, and uncomfortable|
|3||Extremely agitated, thrashing|
|1||1–2 position changes/min|
|2||2–6 position changes/min|
|3||Continuous position changes|
|Appearance/blepharospasm||0||Lids are completely open, in normal position|
|1||Lids are partially closed so that the palpebral opening is decreased by ∼25%|
|2||Lids are partially closed so that the palpebral opening is decreased by ∼50%, mild tearing|
|3||Lids are partially closed so that the palpebral opening is decreased by ∼75%, moderate tearing|
|4||Lids are completely closed, marked tearing|
|Behavior (unprovoked)||0||Too sedate to evaluate|
|3||Moderately abnormal (less mobile or alert than normal, unaware of surroundings or very restless)|
|4||Markedly abnormal (very restless, vocalizing, self-mutilation, grunting, facing back of cage)|
|Interactive behaviors||0||Too sedate to evaluate|
|2||Pulls away when eye touched|
|3||Vocalizes when eye touched, reluctant to move but will if coaxed|
|4||Violent reaction to touching of eye, snapping, growling when approached, will not move when coaxed|
|1||Crying but responds to quiet voice and stroking|
|2||Intermittent crying, no response to quiet voice and stroking|
|3||Constant crying (unusual for this individual animal), no response to stroking or voice|
On the morning of the study, all dogs were pain scored in order to establish baseline pain values. Each dog was sedated with dexmedetomidine (Dexdomitor®; Pfizer, New York, NY, USA) IV (5 μg/kg) prior to corneal wounding. After placement of an eyelid speculum, a 4 mm diameter area of the central cornea of the right eye (OD) was outlined with a corneal trephine. An excimer laser spatula (BD Ophthalmic Systems, Bidford on the Avon, Warks, UK) was used to remove the epithelium. The right cornea of the OD was then stained with fluorescein dye and photographed. Sedation was then reversed using atipamazole (Antisedan®; Pfizer, New York, NY, USA) IM in an equal volume to the dexmedetomidine dose that the dog had received.
Following wounding, the dogs were randomly divided into three treatment groups. All dogs received a topical gentamicin 0.3% (Major Pharmaceuticals, Livonia, MI, USA) in the right eye every 8 h until healing as evidenced by the lack of fluorescein stain retention. Treatment medications consisted of either topical nalbuphine solution (Prescription Center, Fayetteville, NC, USA), oral tramadol (Amneal Pharmaceuticals of NY, Hauppauge, NY, USA) PO (4 mg/kg) administered in pill pockets (Pill Pockets®, Mars Petcare, Franklin, TN, USA), and saline eye drops (Baxter, Deerfield, IL, USA). Empty pill pockets served as an oral placebo.
The three treatment groups were as follows:
Group NB (n = 5) – one drop 1% preservative-free nalbuphine placed OD every 8 h and an oral placebo every 8 h.
Group TR (n = 5) – one drop sterile saline solution placed OD every 8 h and oral tramadol (4 mg/kg) every 8 h.
Group CNTRL (n = 4) – one drop sterile saline solution placed OD every 8 h and an oral placebo every 8 h.
Nalbuphine solution or saline solution was placed in the eye at least 5 min after administration of gentamicin. Topical and oral treatments were begun when the dog was sternal and able to swallow. Initiation of treatment was defined as t = 0. Treatments were administered by a person not involved in pain scoring.
Pain scoring and examination
All dogs were assessed using a subjective pain scoring system (Table 1) by the same trained observer who was unaware of the grouping or the treatments being administered. Pain scoring categories included comfort, movement, degree of blepharospasm, unprovoked behavior, interactive behaviors, and vocalization. Pain scores were recorded at t = 0, 1, 2, 4, and 6 h and then at every 6 h until wounds were healed in each dog. Additionally, each dog underwent a daily ophthalmic exam including slit lamp biomicroscopy, indirect ophthalmoscopy, applanation tonometry, and corneal fluorescein staining. Ocular exam findings were scored using a modified Hackett–McDonald scoring system.37 Exam scores were recorded for conjunctival congestion/hyperemia (0–4+), conjunctival chemosis/swelling (0–4+), conjunctival/ocular discharge (0–3+), corneal edema (0–4+), and corneal vascularization (0–2+). Wounds were defined as healed when fluorescein dye was no longer retained, at which time treatments and pain scoring were discontinued for that dog. Any dog having a total subjective pain score of ≥9 (maximum possible = 21) or that scored ≥3 in any one category was given rescue analgesia (morphine, Baxter, Deerfield, IL, USA; 1.0 mg/kg IM). Saline, tramadol, nalbuphine, and placebo were discontinued in rescued dogs and pain scoring and gentamicin were continued until corneal wounds were healed. Additional rescue analgesia was administered based on the pain scoring criteria until corneal wounds were healed. Treatment failure was defined as the need for morphine administration as a rescue analgesic.
Differences in pain scores were analyzed using the Wilcoxon’s rank-sum test. Differences in incidence of treatment failure were analyzed using a Fisher’s exact test. Differences in healing times were analyzed using a logrank test. Differences in conjunctival congestion, conjunctival chemosis, and ocular discharge were analyzed using a repeated measure anova. Significance was inferred at P < 0.05.
- Top of page
- Materials and Methods
Median (range) overall baseline pain scores were 2 (2) for all dogs. There was no significant difference in total subjective pain scores between the groups at any time point during the study (Table 2). Mean pain scores in nonrescued dogs tended to peak in all groups between 6 and 18 h after t = 0 (Fig. 1). At no time point was a dog’s total pain score ≥9. Dogs that required rescue analgesia did so due to a blepharospasm score ≥3. Mean pain scores for all dogs returned to baseline value by 48 h.
Table 2. Median (range) pain scores before (baseline) and at all time points after wounding in dogs treated with oral tramadol, topical nalbuphine, or saline/placebo. Baseline or normal pain score for all patients was 2
|Pain score time||Tramadol||Nalbuphine||Saline/placebo|
|No. of dogs*||Score (range)||No. of dogs*||Score (range)||No. of dogs*||Score (range)|
|Baseline||5||2 (2)||5||2 (2)||4||2 (2)|
| 0 h||5||2 (2)||5||2 (2)||4||2 (2)|
| 1 h||5||2 (2–4)||5||2 (2)||4||2 (2–3)|
| 2 h||5||2 (2–5)||5||2 (2–3)||4||2.5 (2–8)|
| 4 h||4||2 (2)||5||3 (2–4)||3||2 (2)|
| 6 h||4||2 (2–3)||5||6 (3–6)||3||4 (2–4)|
|12 h||4||2.5 (2–5)||4||5.5 (3–7)||3||2 (2–6)|
|18 h||4||4 (2–6)||2||3 (3–6)||2||2 (2)|
|24 h||4||4 (2–5)||1||3 (3)||2||2 (2)|
|30 h||4||2.5 (2–4)||1||4 (4)||2||2 (2)|
|36 h||4||2.5 (2–3)||1||2 (2)||2||2 (2)|
|42 h||4||3 (2–3)||1||2 (2)||2||2 (2)|
|48 h||4||2 (2)||1||2 (2)||2||2 (2)|
|54 h||2||2 (2)||1||2 (2)||1||2 (2)|
|60 h||2||2 (2)||0|| ||1||2 (2)|
|66 h||1||2 (2)||0|| ||1||2 (2)|
|72 h||1||2 (2)||0|| ||1||2 (2)|
|78 h||0|| ||0|| ||1||2 (2)|
|84 h||0|| ||0|| ||1||2 (2)|
Figure 1. Mean pain score trends in dogs receiving treatment for unilateral corneal wounds. Treatment groups were: NB – topical nalbuphine, TR – oral tramadol, and CNTRL – saline/placebo. Trends illustrated are for each treatment group from time 0 to 54 h. Each time point reflects only dogs still receiving study medication. Mean pain scores returned to the baseline value by 48 h for all dogs that had not been rescued.
Download figure to PowerPoint
Treatment failure occurred in a total of seven (50%) of the 14 dogs in the study (Table 3). In the NB group 4/5 dogs (80%) required a rescue dose of morphine, while 1/5 (20%) in the TR group and 2/4 (50%) in the CNTRL group required rescue analgesia. Of the seven dogs that received rescue analgesia, two dogs (one in the NB group and one in the CNTRL group) required a second dose prior to wound healing. Mean time until rescue analgesia was needed was 9.5 ± 7 h. There was no significant difference in incidence of treatment failure between groups (P = 0.184).
Table 3. Incidence of treatment failure and time-to-treatment failure (TTF)
|% Treatment failure||80||20||50|
|TTF||6 h||2 h||2 h|
|12 h|| ||12 h|
|12 h|| || |
|18 h|| || |
|TTF mean ± SD (h)||12 ± 4.9||2 ± 0||7 ± 7.07|
Mean corneal healing time, as noted by negative fluorescein uptake, was 54 h for the nalbuphine group, 56.4 h for the tramadol group, and 57 h for the control group (Fig. 2). All wounds were healed by 84 h (3.5 days) with a mean healing time for all wounds of 55.8 h (2.3 days). There was no significant difference between healing times for each group (P = 0.95).
Figure 2. Mean ± SD corneal wound healing times for dogs treated for unilateral corneal wounds. Treatment groups were: NB – topical nalbuphine, TR – oral tramadol, and CNTRL – saline/placebo. There was no significant difference in time to heal between groups (P = 0.95).
Download figure to PowerPoint
Mild to moderate (1–2+) conjunctival hyperemia was noted in all dogs. Minimal corneal edema (1+) at the site of the corneal wound was noted in all dogs beginning at approximately 12 h after wounding and resolved in all dogs by 42 h. Mild conjunctival discharge (1+) was observed in eight dogs and mild chemosis (1+) in four dogs. No signs of reflex uveitis (aqueous flare or cells and miosis) were noted and no corneal vascularization was noted during any examination. There was no significant difference in the observed ocular findings between treatment groups (P = 0.245). Intraocular pressure values OU were within the accepted normal limits of 19.2 ± 5.9 mmHg for the duration of the study.38
- Top of page
- Materials and Methods
The results of this study are disappointing as neither oral tramadol nor topical nalbuphine showed a significant difference in pain relief from the control animals treated with topical saline and placebo. Although our results suggest that corneal pain in four of five dogs (80%) treated with oral tramadol was adequately controlled, and the percentage of dogs requiring rescue analgesia in the tramadol group was lower, this result was not statistically significant when compared to the results of the other treatment groups. This is most likely due to the small sample size used in this pilot study. Power calculations based results of this study indicate that a sample size of 11 dogs/group would find a statistically significant difference between tramadol and nalbuphine, if one exists, assuming an α of 0.5 and a power of 0.7.
Interestingly, several human studies have shown that low doses of nalbuphine did not control pain in male patients.39–41 One of the studies found that nalbuphine not only did not control pain but caused increased pain in men compared to those receiving placebo, while a similar anti-analgesic effect was not observed in women.40 Studies involving rats and primates have reported a different outcome in which females showed a poorer response to nalbuphine than males.42–44 To date, the cause of these gender and species differences has not been elucidated. Without investigating nalbuphine’s effects in female dogs, and because there was no statistically significant difference, we cannot definitively confirm a gender difference, or anti-analgesic effect, in response to nalbuphine, although our results suggest such an effect might be present.
Another possible reason for the poor pain control with nalbuphine reported in the present study can be argued based in nalbuphine’s known receptor activity. It has been shown that nalbuphine has weak κ-receptor agonist effects and antagonist effects at the μ- and δ-opioid receptors.31 This κ-agonism along with μ-antagonism may not provide corneal pain relief due to the known receptors present in the canine cornea. It has been shown that dogs possess primarily δ-receptors with infrequent μ-receptors in the corneal epithelium and anterior stroma.16 The presence or absence of κ-receptors in the cornea of dogs is unknown and must be further investigated.
All corneal wounds in this study healed within 84 h of wounding, with the average healing time for all dogs being 55.8 h (2.3 days), which corresponds with normal healing intervals observed elsewhere.45 These normal healing times suggest that neither oral tramadol nor topical nalbuphine have inhibitory effects on corneal wound healing, although we could not fully compare healing rates between groups as study medications were discontinued in rescued dogs. Furthermore, a recent in vitro study reported that topical 1% nalbuphine has the potential to be toxic to corneal epithelial cells.46 This same study also found the nalbuphine solution to be unstable, with precipitation occurring as temperature and pH approached physiologic levels.46 The pH of the nalbuphine solution used in this study was 6.2 and no precipitation was noted following application. Mild to moderate conjunctival hyperemia was noted in all dogs within several hours of the wounding procedure. There were no significant differences in conjunctival hyperemia between treatment groups further suggesting that pH was within tolerable limits.
An additional limitation of this study is that pain assessment in animals is highly subjective and response to pain among animals is variable. This was demonstrated by the unexpectedly low 50% rescue rate noted in the control group of this study as well as in other canine pain studies.47,48 Multiple studies demonstrate that even with appropriate scoring systems, control groups may not always have a 100% rescue rate.34,47,48 The pain scoring system used in this study was modified from the University of Melbourne Pain Scale49 and has been validated in several previous studies, including ocular models, where control animals were rescued at significantly higher rates than other groups.48,50,51 In the study reported here, blepharospasm proved to be the most consistent indication of pain, since no dog ever had a total subjective pain score ≥9. This finding also points to the need for pain scoring systems that are tailored to meet criteria that are appropriate for the type of pain incurred and the expected clinical signs given the injury. Additionally, the low rescue rate in the control group, and the need for only one rescue dose of analgesia in five of seven dogs rescued, suggests that superficial corneal wounds are not as painful as we expected, which fits with the lower corneal sensitivity previously documented in dogs.3 Unfortunately, this low rate of rescue in the control group was not accounted for in pre-study sample size calculations as there were no other studies from which to derive data. The information derived from this study will be crucial for researchers planning future studies on corneal pain in dogs.
Pain scores in nonrescued dogs peaked between 6 and 18 h and then returned to normal levels by 48 h, which was an unexpected result. This variation in onset and decline of pain may explain the large variation seen in the pain levels in clinical patients, as the time of presentation to a veterinarian after the original injury is highly variable. Although only 50% of corneal wounds achieved complete closure at 48 h, subsequent pain scores in nonrescued dogs remained at baseline values. This further demonstrates the variability in the expression of pain by patients with corneal wounds and the need to tailor pain medication toward the specific patient.
Results of this pilot study suggest that topical nalbuphine is not effective for treating corneal pain, at least with this model of corneal wounding. The purpose of disseminating data from this pilot study is to encourage further studies with larger sample sizes to fully validate the effectiveness and safety oral tramadol, but probably not topical nalbuphine, as therapy for corneal pain in dogs.