Blood Propofol Research Articles

The first fatal 2,6-di-isopropylphenol (propofol) poisoning in Singapore: a case report

The propofol levels in the blood and tissues of a 37-year-old Chinese male suspected of having fatally self-administered an intravenous dose of 1600 mg of propofol (12.3–15.4 times the dose required for the induction of anaesthesia) were determined by headspace gas chromatography. The blood (femoral), liver, kidney and brain propofol levels were 2.5 μg/ml, 22 μg/g, 3.6 μg/g and 11.3 μg/g, respectively. The blood propofol level in the present case is 11.4 times and the liver propofol level is 15.7 times that of the first propofol overdose reported in the literature.

Propofol infusion and the suppression of consciousness: the EEG and dose requirements

We have used Median Power Frequency (MPF) to study changes in the electroencephalogram during propofol infusions in 52 women about to undergo gynaecological surgery. Patients were allocated to receive propofol by one of nine different manually-controlled infusion schemes designed to achieve and maintain a stable blood propofol concentration between 1.0 and 6.0 micrograms ml-1, covering a range of states between conscious sedation and full anaesthesia. We recorded the changes in MPF and the response to clinical signs of loss of consciousness at these different doses and concentrations of propofol. Using probit analysis, we derived MPF values corresponding to 50% and 95% suppression of response to verbal (9.3 Hz and 6.8 Hz), eyelash (8.9 Hz and 6.7 Hz) and venepuncture (5.7 Hz and 3.0 Hz) stimuli. Likewise, we obtained dose and concentration requirements for propofol to suppress these stimuli. The mean (95% confidence intervals) ED50 (5.8 (3.5-6.8) mg kg-1 h-1) and ED95 (8.3 (7.1-16.9) mg kg-1 h-1) propofol doses for suppression of consciousness were similar to the values for suppression of the eyelash reflex (6.2 (5.3-6.8) mg kg-1 h-1 and 8.6 (7.8-10.8) mg kg-1 h-1, respectively). The EC50 for loss of consciousness was a propofol concentration of 2.3 (1.8-2.7) micrograms ml-1 and for 50% suppression of MPF was 3.1 (2.7-3.5) micrograms ml-1. The dose required for 50% suppression of MPF was 7.1 (6.2-8.0) mg kg-1 h-1. After 30 min, at blood propofol concentrations > 4.0 micrograms ml-1, consistent with stable anaesthesia, the mean MPF was 5.6 (4.5-6.3) Hz.

Durée de l'anesthésie et mode d'administration du Diprivan ®

The dosage regimen to maintain propofol anaesthesia should ensure adequate depth of anaesthesia for the procedure, and allow a swift recovery. If the dosage is not adapted to the various phases of surgery, there is a risk of awareness especially after painful stimuli. This awareness is not always announced by signs which would allow its prevention. It is characterized by patient's movements but does not result in a memorization of the episode, except if the patient opens his eyes at that moment. Conversely, if propofol blood concentration is maintained at a sufficiently high level to prevent such unexpected movements, it will be necessary to reduce this concentration in the last stage of surgery to avoid a delayed recovery. Intermittent bolus injections and not controlled infusions are gradually abandoned. The optimal use of propofol should rely on the: 1. Use of an infusion pump (volumetric pump) which ensures regular administration: 2. Correct choice of administration technique, that ensures the most stable anaesthesia regardless of the type of operation, using the least possible amount of anaesthetic agent; 3. Combination of points 1 and 2, the first intellectual step towards progressively acquainting anaesthetists with the simultaneous use of an automated administration associated with a computer, which is the most reliable and effective technique today.

Propofol as an induction agent in the goat: a pharmacokinetic study

The pharmacokinetics of propofol, 4 mg/kg, administered as a bolus dose intravenously (i.v.) prior to the maintenance of anaesthesia with halothane in oxygen, were determined in five goats, and a clinical impression of its use as an induction agent was made. Induction of anaesthesia was rapid and smooth, providing satisfactory conditions for intubation in all animals. Post-induction apnoea occurred in one goat and minimal regurgitation of ruminal contents was recorded in two animals. Recovery times were rapid with a mean time to standing after halothane inhalation ceased of 13.7 min. The blood propofol concentration time profile was best described by a bi-exponential decline in all five goats. The mean elimination half-life was short (15.5 min), the volume of distribution at steady state large (2.56 l/kg) and the clearance rapid (275 ml/min.kg). Propofol was shown to be a very satisfactory induction agent in the goat.

Effects of medetomidine premedication on propofol infusion anaesthesia in dogs

Observations of cardiovascular and respiratory parameters were made on six dogs anaesthetized on two separate occasions for 120 minutes with a propofol infusion, once without premedication and once following premedication with 10 μg kg −1 of intramuscular medetomidine. During anaesthesia the heart rate and cardiac index tended to be lower following medetomidine premedication, while the mean arterial pressure was significantly greater (p<0.05). Although the differences were not statistically significant, the systemic vascular resistance, pulmonary vascular resistance and stroke volume index were also greater in dogs given medetomidine. The mean arterial oxygen and carbon dioxide tensions were similar under both regimens, but in 2 dogs supplementary oxygen had to be administered during anaesthesia to alleviate severe hypoxaemia on both occasions they were anaesthetized. Minute and tidal volumes of respiration tended to be greater in dogs not given medetomidine but medetomidine premedication appeared to have no effect on venous admixture. Dogs given medetomidine received intramuscular atipamezole at the end of the 120 min. propofol infusion; the mean time from induction of anaesthesia to walking without ataxia was 174. min in the unpremedicated dogs and 160 min. in the dogs given atipamezole. The mean blood propofol concentration at which the dogs walked without ataxia was higher in the unpremedicated animals (2.12 ± 0.077 μg. ml −1 compared with 1.27 ± 0.518 μg. ml −1 in the premedicated dogs). The oxygen delivery to the tissues was lower after medetomidine premedication (p = 0.03) and the oxygen consumption was generally lower after medetomidine premedication but the difference did not achieve statistical significance. No correlation could be demonstrated between blood propofol concentration and cardiac index, systemic or pulmonary vascular resistance indices, systolic, diastolic or mean arterial blood pressures.

Concentration-Effect Relationships of Propofol After Total Intravenous Anesthesia

To evaluate the concentration-effect relationships of propofol during recovery after total intravenous anesthesia, 20 female patients undergoing lower abdominal surgery were studied. In 10 patients (Group B) the propofol infusion was supplemented with an epidural block with bupivacaine to evaluate the relation between the blood concentration of propofol and various pharmacodynamic end-points. The remaining 10 patients (Group A) received an alfentanil infusion intravenously instead of the epidural block to assess the dynamic interactions of alfentanil and propofol. Postoperative performances (drowsiness, amnesia, cooperation, and orientation) were evaluated by means of scoring scales. Critical flicker fusion threshold (CFF) also was used to assess the level of postoperative alertness. A propofol blood concentration of 2.5 micrograms/mL was required for satisfactory hypnosis during surgery and at 0.8 +/- 0.4 microgram/mL, the patients were considered fully awake. A concomitant alfentanil infusion reduced the propofol concentration required by 0.2-0.4 microgram/mL for the same degree of effect. Rapid recovery was seen in all patients, but in the group receiving alfentanil infusion there was a shift to the left of the concentration-effect curve in regard to drowsiness and a statistically significant prolongation of recovery by CFF-measurement which suggests a possible dynamic interaction with alfentanil. We conclude that there is a good correlation between the blood concentration of propofol and the pharmacodynamic responses during recovery.

Power spectral analysis of the electroencephalographic and hemodynamic correlates of propofol anesthesia in the rat: Intravenous infusion

Based on the tail-flick response to noxious thermal stimuli, we determined in the present study that effective antinociception could be achieved in adult male Sprague-Dawley rats 15 min after intravenous infusion of propofol at 60 mg/kg/h. Simultaneous power spectral analysis of the electroencephalographic (EEG) and systemic arterial pressure signals further revealed a concomitant depression of the activity of all EEG frequency bands (δ, θ, α, β), alongside hypotension, negative inotropic and chronotropic actions, and attenuated baroreceptor reflex and vasomotor activity. These effects were congruent with a plasma concentration of propofol in the arterial blood of 1.70 ± 0.13 μg/ml, as determined by high-performance liquid chromatography.

The effects of halothane and nitrous oxide on the pharmacokinetics of propofol in dogs.

The pharmacokinetics of propofol, 6.5 mg/kg, administered as a bolus dose intravenously (i.v.) were studied in six dogs (group 1). The effect of maintaining anaesthesia with halothane and nitrous oxide in oxygen on propofol pharmacokinetics was also investigated in six dogs undergoing routine anaesthesia (group 2). Induction of anaesthesia was rapid in all animals. Post-induction apnoea was a feature in three of the 12 dogs. The blood propofol concentration-time profile was best described by a bi-exponential decline in two dogs in group 1 and in 3 dogs in group 2, and by a tri-exponential decline in four dogs in group 1 and 3 dogs in group 2. The elimination half-life was long in both groups (90.9 min and 75.2 min, respectively), the volume of distribution at steady state large (4889 and 4863 ml/kg) and the clearance rapid (58.6 and 56.3 ml/kg.min). There were no significant differences between the groups, thus indicating that maintenance of anaesthesia with halothane and nitrous oxide had no effect on the pharmacokinetics of propofol in the dog.

Chronic Alcoholism Increases the Induction Dose of Propofol in Humans

The doses of propofol that produce loss of consciousness were investigated in 26 patients with chronic alcoholism and in 20 patients with a history of small alcoholic intake undergoing ear, nose, and throat surgery under general anesthesia. Last ethanol consumption by the alcoholics was 24 h preoperatively, as they had no access to alcohol when admitted to the hospital. Propofol was infused at a rate of 1200 mL/h (200 mg/min). The doses required to produce (a) loss of verbal contact and (b) loss of ability to grasp a 20-mL syringe filled with water were recorded. At this time a 2-mL venous blood sample was collected to detect propofol blood concentrations. A painful stimulus was applied to the abdomen, and a positive or negative response was noted. The mean +/- SD dose of propofol required for loss of verbal contact was 2.7 +/- 0.42 mg/kg in the alcoholic group and 2.2 +/- 0.43 mg/kg in the control group (P < 0.001). The dose of propofol required for dropping the syringe was significantly higher in the alcoholic group, 4.2 +/- 1.02 mg/kg versus 3.2 +/- 0.75 mg/kg in the control group (P < 0.01). The two groups did not differ significantly regarding the propofol blood concentrations at loss of consciousness, or the frequency of response or no response to painful stimulus. These findings suggest that the doses of propofol required to induce anesthesia in chronic alcoholic patients are more than in patients who drink socially.

Pharmacokinetics of propofol in mixed-breed dogs and Greyhounds

Summary Pharmacokinetics and recovery characteristics of propofol in Greyhounds and mixed-breed dogs were compared. In all dogs, disposition of propofol was adequately described by a 2-compartment open model, with a rapid distribution phase followed by a slower elimination phase. When findings in Greyhounds were compared with those in mixed-breed dogs, significant differences were observed in mean concentrations of propofol in blood, recovery characteristics, and estimates for apparent volume of distribution, volume of distribution at steady state, and total body clearance. In addition, Greyhounds recovered from anesthesia at higher concentrations of propofol than did mixed-breed dogs. A secondary peak in blood propofol concentration was observed in 8 of 10 Greyhounds and in 5 of 8 mixed-breed dogs. This peak corresponded to the time of return of the righting reflex.

PHARMACOKINETICS OF PROPOFOL ADMINISTERED BY INFUSION IN DOGS UNDERGOING SURGERY

We have investigated the pharmacokinetics of propofol in seven beagle dogs undergoing surgery for implantation of s.c. tissue pouches. Blood concentrations of propofol were measured using high pressure liquid chromatography with fluorescence detection. After premedication with acepromazine and papaveretum and induction of anaesthesia with propofol 4 mg kg-1 i.v., anaesthesia was maintained with propofol 0.4 mg kg-1 min-1 and 67% nitrous oxide in six dogs. In the seventh dog, the infusion rate was increased to 0.6 mg kg-1 min-1 in order to achieve surgical anaesthesia. The mean elimination half-life was 322.3 (sem 27.0) min, mean volume of distribution at steady state 6.510 (0.524) litre kg-1, mean body clearance 50.1 (3.9) ml kg-1 min-1 and mean residence time 131.6 (8.9) min. The blood propofol concentrations achieved using this standard dosing regimen showed wide variation among individuals. All dogs recovered rapidly from anaesthesia. The mean time to extubation was 7.6 min and to standing unaided was 30.7 min from the end of the infusion.

Power spectral analysis of the electroencephalographic and hemodynamic correlates of propofol anesthesia in the rat: Intravenous bolus administration

Based on the tail-flick response to noxious thermal stimuli, we determined in the present study that the minimal effective antinociceptive dose of propofol in adult, male Sprague-Dawley rats, given in an intravenous bolus manner, was 10 mg/kg. Simultaneous power spectral analysis of the electroencephalographic (EEG) and systemic arterial pressure signals further revealed a concomitant transient suppression of the EEG activity, primarily in the ϑ and σ bands, alongside minor hypotensive and negative inotropic and chronotropic actions, but with maintained vasomotor tone. These alterations followed a time course that paralleled the plasma concentration of propofol in the arterial blood, as detected by high-performance liquid chromatography.

Sedation with propofol during surgery under local blockade

The ability of a target-controlled propofol infusion system to provide sedation for 40 patients undergoing surgery under regional blockade was assessed. Eighty-eight per cent of the total infusion time was at the desired sedation level with little oversedation. This was achieved with a median blood propofol concentration of 0.93 micrograms.ml-1. The pharmacokinetic algorithm performed as well when used for sedation as for total intravenous anaesthesia. The predicted and measured blood propofol concentrations showed a bias of -12% and a precision of 34%.

Pharmacodynamics of propofol in female patients.

Although the clinical properties of propofol have been studied extensively, the pharmacodynamics have not yet been described fully. We studied the propofol concentration-effect relationships for loss of eyelash reflex, loss of consciousness, and hemodynamic changes in 18 female patients, ASA physical status 1, aged 20-49 yr. Propofol was given by computer-controlled infusion. The initial target concentration of 0.5-1 microgram/ml was increased every 12 min by 0.5-1 microgram/ml until the patients lost consciousness. Every 3 min, loss of eyelash reflex and loss of consciousness were tested and an arterial blood sample was taken for analysis of the blood propofol concentration. The concentration-response relationships for loss of eyelash reflex and loss of consciousness were defined by fitting a sigmoid Emax function (where Emax = the maximum effect that can be reached; i.e., 100% of the patients showing loss of eyelash reflex or loss of consciousness) to the response/no response data versus the propofol concentration, using nonlinear regression. The effect of propofol on hemodynamic parameters was analyzed by linear regression. The propofol concentrations at which 50% and 90% of the patients showed loss of eyelash reflex were 2.07 and 2.78 micrograms/ml, respectively. The corresponding values for loss of consciousness were 3.40 and 4.34 micrograms/ml. The systolic and diastolic blood pressure decreased with increasing blood propofol concentration. The correlation coefficients for the decrease in systolic and diastolic blood pressure versus the blood propofol concentration were r2 = -0.663 and r2 = -0.243, but heart rate did not change. In conclusion, propofol concentrations inducing loss of eyelash reflex are less than those inducing loss of consciousness.

PHARMACOKINETICS OF PROPOFOL DURING AND AFTER LONG TERM CONTINUOUS INFUSION FOR MAINTENANCE OF SEDATION IN ICU PATIENTS

The pharmacokinetics of propofol administered as long term infusions were determined in 12 intensive care unit patients (two female; mean age 58 yr, mean weight 66.9 kg) requiring sedation during mechanical ventilation. Patients were recruited after having been administered propofol for 24 h. Blood samples for analysis of propofol were taken during the infusion (mean duration 85.6 h; mean rate 2.58 mg kg-1 h-1) and for up to about 42 h after its termination. The median propofol total body clearance, derived from the apparent steady state propofol blood concentrations during infusion, was 2.11 litre min-1. One patient died during the infusion, from multi-organ failure secondary to a pre-existing septicaemia, and in one other patient no sampling was possible during the first 30 min after infusion; full elimination data were obtained for 10 patients. After termination of the infusion, propofol blood concentrations declined rapidly, with an overall mean decrease of 50% over the first 10 min; thereafter the decline was more gradual. The elimination profile was triphasic in seven patients and biphasic in three patients. Mean half-lives for the three phases were 1.81 (n = 10) min, 70.9 (n = 7) min and 1411 (n = 11) min. There was no apparent trend in the terminal phase half-life with the duration of sampling after infusion.

A Dose–Response Study of the Influence of Propofol on Cerebral Blood Flow, Metabolism and the Electroencephalogram in the Rabbit

This experiment was designed to study the effect of progressively increasing blood propofol concentrations on cerebral blood flow (CBF), cerebral metabolic rate for oxygen (CMRO2), and the electroencephalogram (EEG). Nine New Zealand White rabbits were anesthetized with morphine (10 mg kg bolus and 2 mg/kg/h infusion) and 70% N2O. Both normothermia and normocarbia were maintained throughout the study. A 300 mum diameter platinum electrode and a 20 gauge sampling needle were inserted into the confluence of venous sinuses to permit the measurement of forebrain CBF and CMRO2. Cerebral blood flow was determined using the H2 clearance method and CMRO2 was calculated as CBF x arteriovenous O2 content difference. A single bifrontal EEG signal was also recorded. After baseline data were collected, a propofol infusion was begun, and the dose increased in a stepwise fashion from 0.28 to 1.11 mg/kg/min over 90 min (total dose 62.6 mg/kg). Every 22.5 min CBF, CMRO2, and EEG were recorded and arterial blood was sampled for the determination of propofol concentrations (by high-performance liquid chromatography). Angiotensin II was used to maintain mean arterial pressure >/=80 mm Hg. Eight animals completed the protocol. Blood propofol concentrations rose progressively in all animals, reaching a mean of 34 +/- 12 microg/ml (+/-SD) at the end of 90 min. Electroencephalogram changes during the early stages of the infusion were extremely variable. However, concentrations above approximately 20 microg/ml were associated with progressive EEG suppression, and isoelectricity developed in two animals, at blood concentrations of 41 and 52 microg/ml. There was a progressive dose-related decrease in CBF, which reached a value of approximately 62% of baseline at a concentration of 40 microg/ml (as predicted by a polynomial curve fitted to a plot of CBF versus blood concentration). The CMRO2 also decreased progressively, reaching approximately 57% of baseline at 40 microg/ml. A plot of CMRO2 versus EEG total power suggests that isoelectricity should be associated with a CMRO2 approximately equal to 53% of baseline. We conclude that propofol produces a dose-related decrease in CBF and CMRO2. The relationship between EEG suppression and CMRO2 is qualitatively similar to that seen with barbiturates.

Sédation avec propofol et fentanyl en soins intensifs

This study investigated the efficacy of a constant rate infusion of propofol and fentanyl in thirty patients requiring artificial ventilation for more than 24 h. A loading dose, which differed according to the patient's age, was administered over a 30 min period : 2.5 mg · kg −1 for patients less than 50 (G1) (n = 9), 2 mg · kg −1 for patients between 50 and 60 years old (G2) (n = 9), and 1.5 mg · kg −1 for patients over 60 (G3) (n = 12). This was followed by an infusion of 3 mg · kg −1, · h −1 in G1 and G2, and 2 mg · kg −1 · h −1 in G3. A 1 μg · kg −1 · h −1 infusion of fentanyl was also given. The degree of sedation was assessed with the Ramsay scale before starting, after induction, and every four hours thereafter. When this proved to be insufficient, the dose of propofol was increased by 0.5 mg · kg −1 · h −1 as well as that of fentanyl by 0.5 μg · kg −1 · h −1. Heart rate, mean arterial blood pressure, blood propofol, creatinine, transaminase and lipid levels, and urine output were measured before, during, and after the infusion. The blood propofol level increased during the infusion, being correlated to the doses given (r = 0.64, p < 0.001). Sedation lasted 91.7 ± 57.7 h. After stopping the infusion of propofol, mean recovery times were 7.5 ± 5.9 min (G1), 11.4 ± 11.4 min, and 14.4 ± 13.5 min (G3) (p < 0.05). In this last group, 4 patients took more than 30 min to recover, even though blood propofol levels at the end of the infusion were lower than in the other group members. There was a significant decrease in heart rate in G3, as well as a significant fall in mean arterial blood pressure in G2 and G3, these not being correlated with propofol levels. Fluid loading was required in two patients in G2 and in six in G3 as a result of the fall in blood pressure. Urine volume did not change significantly. Cholesterol level increased in G1 and G3, as well as phospholipids in G1. Combined continuous infusions of propofol and fentanyl was a practical way of sedating, on the long-term, intensive care patients being mechanically ventilated. However, doses should be reduced in elderly patients so as to avoid delaying recovery, as well as the haemodynamic alterations.

EEG-Assisted Titration of Propofol Infusion During Neuroanesthesia

Total intravenous anesthesia (TIVA) with propofol is an alternative to standard techniques for neuroanesthesia. The present study compared the hemodynamic and recovery profiles of 46 neurosurgical patients randomly assigned to one of three different anesthetic treatment groups. Group 1 was anesthetized with a TIVA technique in which propofol was titrated using an EEG-assisted quantification method. Group 2 received a similar propofol-based infusion technique in combination with nitrous oxide. Group 3 (control) received a standard anesthetic technique consisting of thiopental, nitrous oxide, fentanyl, and isoflurane. Significantly less propofol was required in group 2 than in group 1 (7.4 +/- 1.9 vs. 9.0 +/- 1.0 mg/kg/h, respectively). The propofol blood concentration at the first appearance of EEG burst suppression was also higher in group 1 compared to group 2 (5.8 +/- 1.1 vs. 4.8 +/- 0.8 microg/ml). However, 25% of the patients in group 2 were treated for hypotension after induction, compared to none in groups 1 and 3. Hypertensive episodes, on the other hand, were more frequent in groups 1 (43%) and 3 (31%) than in group 2 (12%). Time to awakening was significantly shorter in the control group (6 +/- 6 min) than in groups 1 (14 +/- 10 min) or 2 (12 +/- 16 min). In conclusion, titration of propofol to achieve a burst suppressive EEG pattern resulted in a slower emergence from anesthesia than a standard "balanced" technique. Use of nitrous oxide with propofol produced more hypotension during induction; however, its use improved hemodynamic stability during the maintenance period.

Pharmacodynamic properties of propofol during recovery from anaesthesia.

Propofol was used to induce and maintain anaesthesia in 20 healthy adult ASA physical status 1 patients undergoing dental surgery. Recovery was studied using psychometric tests and clinical assessment while propofol blood levels were obtained at the same study times. Recovery was rapid and there were no significant differences 60 min after anaesthesia (t60) for the tracing test and 90 min for the choice reaction time. All patients were orientated in time at t30 and in space at t45; Stewart's score was maximal in all patients at t60; recovery of normal walking was slower, i.e. t120. Return to baseline for psychometric function was linked to a decrease in propofol blood level, for both the tracing test (P less than 10(-6)) and the choice reaction time (P less than 10(-3)). We conclude that propofol is a suitable anaesthetic agent for induction and maintenance of short duration dental anaesthesia.

Enhancement by Propofol of Epinephrine-induced Arrhythmias in Dogs

Although propofol is a widely used intravenous anesthetic, its effect on epinephrine-induced arrhythmias remains unknown. This study examined the possible interaction between propofol and epinephrine that might affect the induction of ventricular arrhythmias in dogs. The arrhythmogenic threshold of epinephrine was determined during anesthesia with halothane alone, propofol alone, etomidate alone, or etomidate plus varying doses of propofol. The arrhythmogenic dose and the corresponding plasma concentration of epinephrine during propofol anesthesia (blood propofol concentration 18.0 +/- 0.98 micrograms/ml) were 2.52 +/- 0.43 micrograms.kg-1.min-1 and 23.6 +/- 8.5 ng/ml, respectively. During halothane anesthesia (end-tidal 1.3 MAC), they were 2.66 +/- 0.21 micrograms.kg-1.min-1 and 35.7 +/- 1.9 ng/ml, respectively. During etomidate anesthesia, they were 9.67 +/- 1.06 micrograms.kg-1.min-1 and 205 +/- 27.5 ng/ml, respectively. The dose-effect relationship for propofol was examined during etomidate plus propofol anesthesia. Propofol reduced the arrhythmogenic plasma concentration of epinephrine in a concentration-dependent manner: at blood propofol concentrations of 2.33 +/- 0.46, 5.46 +/- 0.71, and 11.2 +/- 0.81 micrograms/ml, the corresponding plasma epinephrine concentrations were 182.6 +/- 52.5, 89.0 +/- 28.8, and 26.6 +/- 6.9 ng/ml, respectively. These results suggest that propofol enhances epinephrine-induced arrhythmias in a dose-dependent manner in dogs.