Cardiopulmonary and sedative effects of intravenous or epidural methadone in conscious dogs

Epidural (EP) administration of opioids (principally morphine) provides better antinociception, with minimal cardiorespiratory effects compared with other routes of administration (Pascoe, 2000). Furthermore, lower doses are required and the duration of antinociception is longer (Valverde, Conlon, Dyson, & Burger, 1992). Methadone, a synthetic opioid with additional antinociceptive properties, results in fewer adverse effects than morphine when administered EP to women undergoing obstetrical procedures (Beeby, MacIntosh, Bailey, & Welch, 1984). Nevertheless, due to its lipophilicity, there is little advantage of methadone administered epidurally as there is significant systemic uptake (Davis & Walsh, 2001). This study compared sedative and cardiopulmonary effects of methadone given IV or EP to healthy, conscious dogs, during 8 hr of assessment.

The Institutional Ethics Committee on Animal Care approved this research (protocol number 032/13). Six healthy 2-year-old beagle dogs (four male; two female) weighing 15.3 ± 2.2 kg were used. Animals were conditioned for 30 days prior to the study and fasted for 8 hr and water was restricted for 4 hr immediately before the study.

Dogs received one of two treatments in a randomized crossover design with a minimum interval of 7 days between treatments. Group MIV received 0.5 mg/kg methadone (Metadon, Critstália, Itapira, Brazil) IV and normal saline (Fisiológico, JP Indústria Farmacêutica S.A., Ribeirão Preto, Brazil) epidurally; group MEP received 0.5 mg/kg methadone epidurally and normal saline IV. Final volumes were adjusted to 0.4 ml/kg with normal saline and administration was performed simultaneously. Sedation scores, cardiopulmonary and blood gas measurements and body temperature, were recorded before administration (T0), and 15, 30, 60, 120, 240 and 480 min following administration (T15–T480).

An epidural catheter (Becton Dickinson, BD™ Catheter Epidural, São Paulo, SP, Brazil) was placed at the lumbosacral junction using a technique described by Valverde (2008). Dogs were anaesthetized using IV propofol (Propovan, Cristália, Brazil), and maintained with increments. Pulse rate, respiratory rate (f_R) and haemoglobin oxygen saturation using pulse oximetry (SpO₂) were monitored and animals allowed to recover completely.

Heart rate (HR) and cardiac rhythm were assessed using electrocardiography (TEB^® ECGPC software v. 1.1, São Paulo, Brazil). A 22G × 1″ catheter was placed in a dorsal metatarsal artery and connected to a calibrated manometer (Bic Med, Itupeva, Brazil) to measure mean arterial blood pressure (MAP). The system was zeroed at the level of the right atrium. Rectal temperature (RT—Celsius degrees °C) was assessed using a clinical thermometer (Flexterm, Incoterm, Porto Alegre, Brazil). Respiratory rate was evaluated by observation. Arterial blood samples were analysed immediately (pH, partial pressure of carbon dioxide (PaCO₂), partial pressure of oxygen (SaO₂), arterial haemoglobin oxygen saturation, base excess (BE), bicarbonate ([HCO₃⁻]), sodium, potassium and chloride ion concentrations), using an automatic analyzer (Cobas b 121 Roche®, Basel, Switzerland). Three evaluators, blinded to the route of methadone administration, scored sedation according to a scale proposed by Kuusela et al. (2000) (Appendix 1). The final score was the mean of the three evaluators and used for statistical analysis.

Normality was assessed (Shapiro-Wilk's test) and data analysed using GraphPad PRISM v. 5 (GraphPad Software, Inc., La Jolla, CA, USA). Repeated measures two-way ANOVA and Bonferroni's post-test were used for comparison of cardiopulmonary variables, RT and blood gas measurements among time points and Student's t test for between treatment comparisons. Sedation scores were compared using Friedman's test and Dunn's post-test, if necessary. Significance was set at 5%.

Results are shown as mean ± standard deviation (SD), for parametric data and median (range) for nonparametric data.

Cardiopulmonary data are shown in Table 1. There were no significant differences between treatments (p > .05). HR was decreased from T15–T120 in MIV (p < .001) and T15–T120 (p < .001) and T240 (p < .05) in MEP, when compared to baseline. No changes in cardiac rhythm occurred. In group MEP, a significant decrease in MAP occurred at T15–T60 (p < .01). Compared with baseline, f_R was significantly increased at T15 and T30 in group MIV (p < .01), and at T15 and T30 in group MEP (p < .05). Compared with baseline, RT was significantly decreased in group MIV at T30–T120 (p < .05) and in group MEP at T15 (p < .05), T30 (p < .01) and, T60–T240 (p < .001).

Compared with baseline, mean pH values were significantly decreased at T15–T60 (p < .001) and T120 (p < .01) in group MIV, and at T15, T30 and T120 in group MEP (p < .05). PaCO₂ was significantly increased at T15, T30 (p < .01) and T60 (p < .001) in group MIV, and at T15, T30 and T120 in group MEP (p < .01), compared with baseline. There were no significant differences in PaO₂ and SaO₂. There were no other significant differences between treatments in arterial blood gas and electrolyte measurements at any time point (p > .05) (Table 2).

Sedation scores were higher in comparison with baseline at T15 and T30 in group MIV, and at T15–T60 in group MEP (p < .05). At T120 and T240, sedation score was higher in group MEP compared with group MIV (p < .01) (Table 3). One animal in group MIV demonstrated clinical signs of CNS excitation (dysphoria) up to 60 min following drug administration. This behaviour was not evident prior to methadone administration.

This study demonstrates that cardiopulmonary and sedative effects are similar when identical doses of methadone are administered IV or EP to healthy, conscious dogs.

Alterations in HR induced by methadone are usually attributed to direct action on vagal tone (Stanley, Liu, Webster, & Johansen, 1980). Additionally, a baroreceptor-mediated fall in HR has been described in response to vasoconstriction and increased systemic vascular resistance (Hellebrekers, Van Den Brom, & Mol, 1989; Maiante, Teixeira Neto, Beier, Corrente, & Pedroso, 2009). This increased blood pressure may be a consequence of CNS excitation (Maiante et al., 2009) or the secretion of arginine vasopressin (AVP) (Hellebrekers et al., 1989; Maiante et al., 2009). We did not observe an increase in MAP following methadone administration and cannot corroborate this peripheral effect on HR or on the secretion of AVP; this may be a reflection of the dose of methadone used in our study. In a study in isoflurane-anaesthetized dogs, a fall in HR occurred almost immediately following IV injection of methadone; such an effect was observed only after 10–50 min following epidural administration (Campagnol et al., 2012).

In our study, panting occurred for up to 30 min following administration of methadone in both groups. This systemic effect of methadone has been described in conscious dogs (Maiante et al., 2009). Panting observed following administration of μ agonist opioids, may be secondary to changes in the thermoregulatory centre and the animal attempting to cool itself down (Pascoe, 2000). The RT of dogs in both groups fell significantly over time (but was not clinically significant), in agreement with Maiante et al. (2009), which may be attributable to this effect on the thermoregulatory centre. However, sedation also induces reduced muscular function, which may lead to RT depression.

There were minimal changes in acid base status, blood gas values and electrolyte concentrations in our study. Oxygenation was unaffected despite a reduction in ventilation evidenced by an increase in PaCO_2.. This increase in PaCO₂ was similar between groups and clinically insignificant but was suggestive of systemic uptake of EP methadone. Arterial pH decreased following methadone administration in both groups but was attributed to increases in PaCO₂ and was not clinically significant. Other studies have produced similar findings (Garofalo et al., 2012), although alternative speculative explanations have included the development of a mild metabolic acidosis as a result of reduced tissue perfusion secondary to elevations in AVP concentrations (Garofalo et al., 2012; Hellebrekers et al., 1989). As we did not identify significant changes in [HCO₃⁻] or BE, we cannot corroborate this theory.

Systemic absorption of drugs from the EP space is determined in part by lipophilicity (Valverde, 2008). The lipophilic nature of methadone may explain the similarity of sedation scores observed in this study between treatments. In light of the results presented in this study, it is difficult to recommend the administration of methadone epidurally, as sedation was similar and of longer duration following administration via this route.

A limitation of this study was that analgesic potency of the treatments was not assessed. Although identical doses of methadone were administered by both routes, it might be expected that analgesia in group MEP would be superior compared to group MIV. As the drug is administered close to the effect site a smaller dose of methadone may produce similar analgesic effects. Furthermore, as doses were identical, the sedative and cardiopulmonary effects are likely to be similar between groups due to systemic absorption of methadone administered epidurally. However, plasma concentrations were not measured and we cannot confirm that this is true. Further studies are warranted to assess the analgesic potency of lower doses of methadone administered by the EP route.

In conclusion, identical doses of methadone administered IV or EP lead to similar cardiopulmonary changes in healthy, conscious dogs with longer effects when administered epidurally.