Unmitigated pain can result in adverse physiologic effects that increase patient morbidity, as well as negatively impact welfare (1, 2). Opioids are at the core of effective acute pain management strategies in dogs, with hydromorphone being used by 72% of veterinary practitioners for this purpose (3). Hydromorphone has a large volume of distribution and is rapidly cleared in dogs, meaning frequent bolus dosing is necessary to achieve sustained therapeutically effective plasma concentrations (4). Hydromorphone administered at a dose of 0.1 mg/kg IV every 2 h provides an adequate antinociceptive effect in dogs (4–6). While intravenous (IV) bolus administration of hydromorphone is cost-effective and reliable, repeated bolus administration is labor intensive, risks peaks and troughs in analgesic effect, and is associated with adverse effects including nausea, emesis, vocalization, and panting (7). An IV continuous rate infusion (CRI) of hydromorphone could overcome some, if not all, of these issues. There is also an emerging need to develop novel opioid dosing regimens for small animals. Some opioids are prohibitively expensive in large dogs, or there have been recent supply chain issues leading to inconsistent opioid availability, and lastly, there are serious concerns over the use of certain opioids due to risk of human diversion (3).

The pharmacokinetics (PK) of several opioid IV CRI protocols have been reported in dogs, but the PK of hydromorphone administered by this route is yet to be investigated, despite clinical descriptions of its use in dogs (7). Previous PK studies suggest IV CRI rates of 0.03 mg/kg/h to target and maintain plasma concentrations of 4 ng/mL, but these recommendations are based on single IV bolus dosing studies where clearance was found to be very high following a single dose (5, 6). These suggested rates might be too high based on pilot work performed by our group in a study evaluating single IV bolus of hydromorphone and using a continuous subcutaneous administration device (8). Fentanyl infusions have variable accumulation in dogs depending on the duration of infusion, a phenomenon described as the context-sensitive half-time, such that the longer the infusion, the longer the elimination half-life and thus time to offset of effect (9). It is plausible, but unknown, if hydromorphone administered by CRI could behave similarly.

The primary goal of this study, therefore, was to describe the PK parameters of hydromorphone following a 48-h IV CRI administered to healthy dogs. The secondary aims of the study were to evaluate serial nausea and sedation scores in the dogs administered hydromorphone.

All dogs completed the study without any serious adverse effects. Heart rate and respiratory rate over the course of the study are shown in Figures 1, 2. The median (range) heart rate during hydromorphone administration was 88 (50–126) beats/min, while the median (range) respiratory rate was 19 (12–52) breaths/min. Panting was noted after initial hydromorphone bolus in each dog but subsided without further panting episodes during the remaining study period.

Sedation scores are shown in Figure 3 and Supplementary Table S2. The median (range) sedation score at baseline was 1 (1–2). Sedation scores increased after hydromorphone administration and remained higher than the baseline values for the duration of the hydromorphone CRI administration. The median (range) sedation score during the hydromorphone CRI was 6 (2.5–11). The median (range) sedation score after hydromorphone discontinuation was 0.8 (0–6).

Nausea scores are shown in Figure 4 and Supplementary Table S3. All dogs had nausea scores of 0 (no nausea) at baseline, which increased after hydromorphone bolus administration. The median (range) nausea scores were 4.9 (1.2–8.7) for the first 4 h after hydromorphone administration and 0.5 (0–3.5) between the 8–48 h timepoints. Nausea scores of 0 were recorded after hydromorphone infusion discontinuation.

Our prospective experimental study was designed to characterize the PK profiles of hydromorphone administered by IV CRI over a 48-h period in healthy dogs. Our data identified that an initial IV bolus of 0.1 mg/kg hydromorphone, immediately followed by an IV CRI of 0.01 mg/kg/h, maintained steady state plasma concentrations between 8 and 12 ng/mL. This is above the minimum effective plasma concentration associated with an antinociceptive effect in people (4 ng/mL) (14, 15). Minimal adverse effects were seen in the dogs administered the hydromorphone IV CRI in this study. Dogs were mildly sedated but with minimal impact on respiratory rate. A short duration of nausea was observed following hydromorphone administration. There was no apparent drug accumulation during the infusion and plasma concentrations decreased quickly once the infusion was discontinued. While the terminal elimination half-life (T1/2_β) was calculated to be nearly 12 h, this finding is likely due to the sensitivity of the assay used in this study (LOQ of 0.1 ng/mL compared with higher concentrations in older reports) and these plasma concentrations are unlikely to provide any clinically relevant analgesia (<1 ng/mL). The results of our study suggest that previous recommendations for IV CRI doses of hydromorphone (0.03 mg/kg/h) are higher than necessary to achieve human therapeutic targets for hydromorphone. Further work is necessary to investigate whether the human target is also appropriate for dogs.

Constant rate infusions are used to achieve a steady plasma concentration for a drug and might minimize the adverse effects of medications compared to their use with bolus dosing. Intravenous hydromorphone exhibits a rapid clearance and short half-life in dogs, making its administration ideal for CRI. Previously the pharmacokinetics of hydromorphone IV bolus dosing were evaluated (4, 6). This data was used to extrapolate that a CRI of 0.03 mg/kg/h would reach 4 ng/mL steady state in 2 h (6). Preliminary data based on a subcutaneous infusion study postulated the 0.03 mg/kg/h CRI may be higher than necessary to achieve therapeutic levels (8). No previous CRI studies on hydromorphone in dogs exist but studies reporting single IV bolus data were compared to our results. Some parameter estimates from our study are similar to previous IV bolus studies in dogs that utilized either compartmental or noncompartmental analyses and Beagle dogs (4, 6). For example, the volume of distribution at steady state (Vdss) in our study was 4.5 L/kg. Previously reported Vdss values following a bolus of 0.1 mg/kg were 4.2 L/kg (6), 4.5 L/kg (4), and 7.2 L/kg for a bolus of 0.2 mg/kg (4). There were interesting differences in the total body clearance compared to those values previously reported. Our results revealed a clearance of 30.4 mL/min/kg, which is nearly identical to hepatic blood flow in the dog (32 mL/min/kg) (16, 17). Previous studies reported 106 mL/min/kg (6) and 68 mL/min/kg (4), which was far greater than hepatic blood flow and while the authors speculated that extrahepatic metabolism could account for these values, this has never been investigated. Hydromorphone is metabolized by the liver and dogs are well known for their capacity to metabolize opioid drugs (18). Differences in analytical assays, pharmacokinetic methods, and dosing regimens may all account for the variation reported in clearance, and perhaps additional studies could help further characterize the clearance of hydromorphone.

Hydromorphone is a mu opioid agonist that is seven times more potent than morphine. The quality and efficacy of analgesia provided to dogs by hydromorphone compared to fentanyl, butorphanol, oxymorphone have been described (7, 19, 20). However, numerous adverse effects are reported with hydromorphone boluses including nausea, vomiting, defecation, excitation, dysphoria, panting, bradycardia, and respiratory depression (18). In our study, nausea scores were increased during the first 4-h of hydromorphone infusion, which corresponded to plasma concentrations between 9 and 41 ng/mL. Interestingly, the nausea scores returned to nearly normal baseline values quickly, while dogs had higher sedation scores for the duration of the infusion. The sedation scores were consistent with observations of dogs that were subjectively quiet, alert, and responsive. The sedative effect of hydromorphone CRI in systemically sick dogs or when combined with other sedative drugs is not yet known. There is potential for increased risk of adverse effects in these instances (e.g., aspiration risk).

There are several limitations of this study, including the small number of dogs investigated and the uniformity of this cohort of the dogs (Beagle dogs of a similar bodyweight). While this homogeneity is preferred for pharmacokinetic evaluation as it limits variability in data, it also excludes differences that could be seen in other breeds, conformations, or genetic make-up of dogs. While we anticipate the results of this study can be applied to populations of animals experiencing pain, further work is necessary to evaluate the efficacy of hydromorphone CRIs in clinical cases.

Hydromorphone administered as an IV bolus and CRI for 48 h at a dose of 0.01 mg/kg/h resulted in steady state plasma concentrations above the minimum effective plasma concentration in people and was cleared rapidly upon discontinuation of the infusion. Minimal effects on nausea scores and heart and respiratory rates were observed, while sedation was observed for the duration of the infusion in dogs.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was approved by North Carolina State University. The study was conducted in accordance with the local legislation and institutional requirements.

CW: Writing – original draft, Writing – review & editing. AL: Writing – original draft, Writing – review & editing. HK: Writing – original draft, Writing – review & editing. YU: Writing – original draft, Writing – review & editing. KM: Writing – original draft, Writing – review & editing.