Materials and Methods

Animals—Medical records from April 2006 through March 2008 were reviewed to identify all dogs that underwent anesthesia and phacoemulsification for treatment of cataracts (treatment dogs). All dogs received topical application of ophthalmic 1% atropine sulfate ointment^a the day before surgery. Then every 30 minutes during the 2-hour period preceding anesthetic induction, 1 drop each of 10% phenylephrine hydrochloride^b and 0.3% scopolamine hydrobromide^c solutions was applied, for a total of 4 applications/drug. Because diabetes requires specific treatment (eg, insulin, glucose, and dietary changes) in the period prior to anesthesia and because diabetes can affect autonomic function and hence might contribute to perturbing blood pressure,^4,5 dogs in the treatment group were further classified as nondiabetic and diabetic.

For comparison purposes, data were also collected on dogs anesthetized during the same period for surgical treatment of corneal ulcers via conjunctival grafting (control dogs). The only ophthalmic treatment these dogs received was atropine before surgery and, in some situations, ophthalmic application of antimicrobials (bacitracin-neomycin-polymyxin,^d ciprofloxacin, or chloramphenicol); otherwise, the control dogs were treated similarly to those undergoing phacoemulsification except they did not receive topically applied ophthalmic drugs in the 2 hours preceding anesthetic induction.

Measurement of MAP—Anesthesia was induced. Arterial blood pressure was measured from a catheter inserted in the dorsal pedal artery with the transducer zeroed at the level of the heart^e or by use of an oscillometric monitor.^f Values of MAP > 131 mm Hg were considered indicative of hypertension.⁶ Additional monitoring included continuous ECG, pulse oximetry, capnography, and neuromuscular blockade as assessed with a peripheral nerve stimulator. Measurements and events were generally recorded at 5-minute intervals on the anesthetic record. All patients received atracurium^g (0.1 to 0.4 mg/kg, IV; median, 0.2 mg/kg) during the period of interest (ie, after a dog was positioned in the operating room and before the start of surgery). Most anesthetic records had discontinuities in MAP and HR data. However, many were complete over the 20-minute period around the time atracurium was administered, presumably because the anesthetist was concerned about the potential for atracurium to cause hypotension.^7,8 To standardize the measurement points at which data were collected, values for MAP and HR were obtained for 10 (t–10) and 5 (t–5) minutes before atracurium was given, at the time atracurium was administered (t0), and 5 (t5) and 10 (t10) minutes later.

To exclude the possible effects of surgical stimulation on MAP data, dogs were not included in the study if surgery began 10 minutes before or after the administration of atracurium. Also excluded were dogs that received cardioactive drugs within 10 minutes before or after atracurium administration (eg, anticholinergics or α₂-adrenoreceptor agonists but not IV administered crystalloid fluids). When > 1 dose of atracurium was given to a dog, only measurements recorded within 10 minutes before or after the first dose were considered. In all cases, patients were in dorsal recumbency. Other data collected from eligible records included dog sex, age, and body weight; drugs used for sedation, induction, and maintenance of anesthesia; whether anticholinergics were administered parenterally before anesthetic induction; and dose and time of atracurium administration.

Statistical analysis—Data were analyzed by use of computer software.^h Use of the Shapiro-Wilk test revealed the data were not normally distributed; therefore, Kruskal-Wallis 1-way ANOVA was used to determine whether MAP and HR differed among all 3 groups (diabetic and nondiabetic treated dogs and control dogs) at the measurement point. The Wilcoxon signed rank test was used as a post hoc method to determine whether MAP and HR values were significantly greater at t–5 and t5 than at t0. A value of P < 0.05 was considered significant unless Bonferroni corrections were applicable, in which situation the corrected value of P was used. Because the data were not normally distributed, summary data are reported as median (minimum and maximum) or as percentages.

Results

Animals—One hundred forty-three dogs that underwent phacoemulsification were identified, of which 79 were not diabetic and 64 were diabetic. Only 37 nondiabetic dogs and 29 diabetic dogs met the criteria for inclusion. Forty-five dogs were identified for the control group, none of which were diabetic and 15 of which met the criteria for inclusion. Data for all dog groups were summarized (Table 1). In all groups, the most common cause for exclusion was that surgery started sooner than 10 minutes after atracurium administration. Bilateral cataracts were present in 89% (33/37) of the nondiabetic dogs in the treatment (phacoemulsification) group and 100% of the diabetic dogs in the same group; these dogs received ophthalmic medication (ie, 10% phenylephrine hydrochloride and 0.3% scopolamine hydrobromide) in both eyes.

Anesthesia—All study dogs recovered from anesthesia and were discharged from the hospital. Drugs used during anesthesia were summarized (Table 1). Because acepromazine or medetomidine were used prior to surgery in many dogs and because both drugs can alter arterial blood pressure, a hypothesis was tested a posteriori (χ² test with Bonferroni correction) that the proportions of dogs receiving either drug were not different among groups. Test results confirmed that the proportions of dogs receiving acepromazine or medetomidine were not significantly (P = 0.34 and P = 0.027, respectively) different among groups.

Dog groups did not differ significantly with respect to mean body weight. However, the interval from induction of anesthesia to atracurium administration was longer and the dose of atracurium was greater in diabetic dogs than values for the other 2 groups (Table 1).

MAP and HR—Results of a post hoc χ² test indicated a significantly (P = 0.006) greater proportion of dogs had MAP measured directly (via arterial catheter), compared with the other 2 groups. Analysis of available data revealed that MAP values were greater in the 2 treatment groups than in the control group at t0 (P = 0.006) and t5 (P = 0.0012; Table 2). At t10, MAP was also higher than the control MAP for the nondiabetic treatment group but not for the diabetic treatment group. When compared with values at t0, both MAP and HR were significantly higher at t5 in the nondiabetic and diabetic treatment groups (P < 0.025). In the control group, however, MAP and HR were the same at t0 and t5. At t–10, t–5, t0, t5, and t10, 6% (1/16), 8% (2/24), 3% (1/32), 17% (5/30), and 22% (8/36), respectively, of nondiabetic treatment dogs had an MAP value > 131 mm Hg; in all, 9 dogs in this group were hypertensive at least once. One diabetic treatment dog had an MAP value > 131 mm Hg (at t0 and t5). Seventy-three percent of MAP values > 131 mm Hg were detected within 10 minutes after atracurium administration in the combined nondiabetic and diabetic treatment groups. At no time did a control dog have an MAP value > 131 mm Hg.

Five dogs in the treatment group received drugs (ie, esmolol, acepromazine, fentanyl, hydromorphone, or atipamezole) to treat hypertension detected during the observation period, and the vaporizer setting was also increased in several dogs at the same time. With the exception of 1 dog, all treatments for hypertension were given after administration of atracurium. All dogs treated for hypertension were nondiabetic and had received the phenylephrine and scopolamine eyedrops prior to anesthesia. Heart rate did not differ among the 3 groups at any time.

Values of MAP and HR were stable over the 5 minutes prior to atracurium administration (P ≥ 0.11 for all comparisons) and then appeared to increase over the 5 minutes after atracurium administration in both treatment groups (Figure 1). Changes were not apparent before or after atracurium administration in the control dogs (P > 0.17 for all comparisons). Among the 2 treatment groups, 27% (7/26) of nondiabetic and 4% (1/25) of diabetic dogs had an increase in MAP of ≥ 20% during the first 5 minutes after atracurium administration. Over the same period, an increase in HR of ≥ 20% was noticed in 21% (5/24) of nondiabetic and 17% (4/24) of diabetic treatment dogs. None of the control dogs had such increases.

View Full Size

Percentage change (Δ) in MAP and HR values in the 5 minutes before (t0 minus t–5 value; A–C) and after (t5 minus t0 value; D–F) atracurium administration in dogs anesthetized to undergo phacoemulsification for treatment of cataracts (treatment group; 37 nondiabetic dogs [A and D] and 39 diabetic dogs [B and E]) or conjunctival grafting for treatment of corneal ulcers (control dogs [C and F]; 15 dogs).

Citation: American Journal of Veterinary Research 71, 12; 10.2460/ajvr.71.12.1407

• Download Figure
• Download figure as PowerPoint slide

Because increases in MAP of > 20% developed after atracurium administration and because the dose of atracurium was significantly higher in the diabetic treatment dogs than in the other 2 groups, a Wilcoxon rank sum test was performed to determine whether a relationship existed between the dose of atracurium and an increase of MAP of at least 20% after atracurium administration. Such a relationship could not be detected (P = 0.74).

Discussion

In the present study, the effect of administration of eye drops containing 10% phenylephrine and 0.3% scopolamine on MAP in dogs anesthetized for phacoemulsification (treatment dogs) was examined. The MAP was measured either directly from an arterial catheter or indirectly by use of the oscillometric method. Because a greater proportion of diabetic dogs that underwent phacoemulsification had MAP measured directly, compared with the proportion of dogs in the other 2 groups (diabetic dogs that underwent phacoemulsification and control dogs that underwent conjunctival grafting), this was a potential source of bias in our data.

When MAP is measured over a wide range of pressures in anesthetized dogs, the oscillometric method has a bias of between 1 mm Hg and −11 mm Hg with respect to the direct method.^1,9 Reanalysis of the data in which a best-case correction (1 mm Hg) was applied revealed trivial differences from the results obtained in our study (Table 2). Reanalysis in which a worst-case correction (−11 mm Hg) was applied revealed differences from our results: at t0, MAP values were not significantly different among the 3 groups; at t5 and t10, the MAP values of nondiabetic dogs anesthetized for phacoemulsification were still greater than those of control dogs (P = 0.01 and P = 0.009, respectively), whereas the MAP values of diabetic dogs no longer differed significantly from those of control dogs. Because no rationale could be found for preferring one correction method to the other, we chose to report and discuss uncorrected results.

The anesthetized dogs that were about to undergo phacoemulsification at our institution had higher blood pressure values than did the anesthetized control dogs. Mean arterial blood pressure does not ordinarily exceed 131 mm Hg in healthy dogs⁶; in 10 of the treatment dogs in our study, the MAP exceeded this value. During the period of recording, the highest MAP value in the nondiabetic treatment dogs was 155 mm Hg and the highest MAP value in the diabetic treatment dogs was 132 mm Hg. On the other hand, the highest MAP value recorded in the control group was 122 mm Hg, indicating none of the control dogs were hypertensive. In the opinion of the anesthesiologists responsible for the study dogs, the hypertension was deemed serious enough to warrant specific corrective action via drug administration in at least 14% (5/37) of the nondiabetic treatment dogs. In several dogs, the increase in blood pressure was perceived as inadequate depth of anesthesia, which triggered administration of additional anesthetics and analgesics. However, if increases in arterial blood pressure or HR are solely due to a pharmacological interaction and not as a response to noxious stimulation, additional administration of anesthetics or analgesics might lead to an overdose. Thus, this complication is clinically important.

The lower MAP and absence of hypertension in the control group suggested that the eyedrops given to both treatment groups contributed to the higher MAP and hypertension detected in those dogs. The eyedrops contained antimicrobials (neomycin and polymixin), dexamethasone, furbiprofen, phenylephrine, and scopolamine. It is unlikely that the higher MAP would be associated with administration of the antimicrobials, dexamethasone, or furbiprofen because these drugs do not usually influence important autonomic activity. Scopolamine is a parasympatholytic and would therefore be expected to increase HR but not necessarily increase MAP. Phenylephrine is an α₁-adrenoceptor agonist that causes vasoconstriction and, thereby, increases blood pressure. The higher blood pressure that was evident in both groups that received preoperative eyedrops suggested that the phenylephrine component of the drops achieved a sufficient plasma concentration to have had a systemic effect. The higher MAP that was detected is in accordance with findings of another study² that demonstrated an increase in systemic arterial blood pressure after administration of 10% phenylephrine ophthalmic drops in nonanesthetized dogs. In that study,² dogs allocated to the high-dose group received approximately 75% of the dose used in the treatment dogs in our study, and that dose was given over a much shorter period (5 minutes). That treatment caused MAP values to increase from a mean ± SD value of 91 ± 10 mm Hg to 107 ± 14 mm Hg. Additionally, another research group¹⁰ reported 3 anesthetized dogs that had arterial hypertension and had been treated with ophthalmic phenylephrine prior to induction of anesthesia. These findings all support the hypothesis that ophthalmic administration of phenylephrine causes phenylephrine to reach plasma concentrations high enough to yield systemic vasoconstriction.

Although the control group was younger than the nondiabetic treatment group, the youngest dog in the control group was 6 months old, which is an age at which dogs can be considered adults from a physiologic and pharmacokinetic standpoint.¹¹ The increment in MAP with age in healthy dogs is trivial after 6 months of age, as it is reportedly between 1 and 3 mm Hg/y of life or it may not change at all.⁶ However, it cannot be discounted that the higher MAP values detected in the dogs anesthetized for phacoemulsification could be attributable to their greater age being associated with lower sensitivity to the hypotensive effects of the anesthetics. Because MAP and HR data were obtained from anesthetic records, preoperative MAP values were not available; therefore, the presence of arterial hypertension prior to ophthalmic drug administration and anesthetic induction could not be ruled out. However, all but 2 instances of arterial hypertension (MAP > 131 mm Hg) were recorded after atracurium administration. Prior to those instances, the affected dogs had been normotensive or hypotensive when anesthetized.

After atracurium was administered, MAP increased; in fact, most of the hypertensive events were recorded after atracurium administration. Heart rate also increased after atracurium administration in the dogs anesthetized for phacoemulsification; this simultaneous increase of HR with MAP was unexpected by the investigators. Such increases could have been signs of increased sympathetic nervous activity or a reduction in parasympathetic nervous activity; the exclusion criteria ensured that these signs of autonomic perturbation were not attributable to surgical intervention, a change in posture, or acute administration of other drugs.

In some study dogs, MAP and HR increased considerably after atracurium administration (Figure 1). The temporal relationship of these increases with atracurium administration suggested that the drug could be responsible. Concomitant decreases in vaporizer settings that could have accounted for the increases in HR and MAP were not evident in the anesthetic records. Additionally, surgical stimulation did not begin for at least 10 minutes after atracurium was administered, therefore ruling out the possibility of noxious stimuli triggering the recorded responses. Ordinarily, atracurium is regarded as having minimal cardiovascular effects when administered to healthy patients at conventional doses¹²; however, the findings of the present study raised the possibility of an interaction between the eye-drops and atracurium influencing MAP and HR. Findings in humans also suggest that atracurium augments the effects of previously high catecholamine concentrations. For example, atracurium administration was associated with spontaneous increases in systemic arterial blood pressure in 3 people with pheochromocytoma, and investigators in that study¹³ concluded that atracurium had circulatory effects of its own or caused catecholamine release in those patients. Hypertension was also detected in a woman receiving monoamine oxidase inhibitors in whom anesthesia was induced with etomidate and atracurium.¹⁴ Prior to orotracheal intubation, systolic arterial pressure reached 350 mm Hg. Those findings and ours suggest that in some circumstances, atracurium acts synergistically with high circulating catecholamine concentrations. However, it can neither be confirmed nor excluded that the rise in MAP and HR after atracurium administration in all situations would have happened without the atracurium. Additional investigation in this regard is warranted.

The proportion of nondiabetic treatment dogs that developed hypertension during the study period (27%) did not differ significantly from the proportion of diabetic treatment dogs (4%). Dogs that had the greatest increases in MAP in both groups anesthetized for phacoemulsification, and thus receiving the 10% phenylephrine and 0.3% scopolamine eye drops, also had the largest increases in HR. The ophthalmic drops were administered at a dose of 1 drop/affected eye at each of the 4 administration points. Because of this, dogs in the treatment group that underwent bilateral phacoemulsification received more phenylephrine and scopolamine than did dogs that underwent a unilateral procedure. In addition, the dose was not normalized to body weight; hence, small dogs received a larger dosage of phenylephrine and scopolamine than did larger dogs. This raised the possibility that plasma concentrations of the drugs were greater in small dogs and suggested that smaller dogs would more likely be hypertensive. To test these possibilities, a Spearman rank test was conducted after the study concluded to evaluate the correlation between the dosage of phenylephrine plus scopolamine (in drops/kg) and the MAP just prior to atracurium administration. The result was a weak (ρ = 0.31) but significant correlation between dosage and MAP (P = 0.007).