Anesthetics are commonly used in fish to reduce handling stress, prevent injuries, provide immobilization, and allow humane surgical interventions. Traditionally, MS-222 has been used as a primary anesthetic agent in US aquaculture because it is currently the only FDA-approved anesthetic in food fish.1 As a local anesthetic, MS-222 blocks sodium channels at peripheral nerves and causes decreased neural activity2; however, the mechanism of action for how it induces general anesthesia in fish remains unknown.3,4 In contrast to MS-222, propofol is a widely used anesthetic in mammalian species. It is known to induce anesthesia in these species by activating GABA receptors in the brain following intravascular administration.5 Despite the popularity of propofol as an anesthetic in mammals, little is known regarding the use of propofol in fish, even though previous studies6,7 have shown that teleost fish have GABA receptors in their spinal cord and brain. In one of a few studies, Fleming et al8 reported that a light plane of anesthesia can be induced in Gulf of Mexico sturgeon (Acipenser oxyrinchus de soti) by IV administration of propofol at a dose ≥ 3.5 mg/kg (1.59 mg/lb).
Unlike in mammalian species, in fish, the gills provide a unique route of drug absorption and elimination. Drugs that are lipid soluble and not ionized can easily diffuse through the gills, and MS-222 is known to reach the bloodstream of fish via the gills and is excreted through the gills rather than the kidneys.9,10 It is therefore possible that propofol may be absorbed via the gills owing to its high lipid solubility.5 At the time this study was performed, we were unaware of any studies that reported immersion with propofol in fish species. However, since that time, propofol has been successfully used to anesthetize catfish by immersion.11
The objective of the study reported here was to determine the efficacy of propofol as an immersion anesthetic agent in koi (Cyprinus carpio). We hypothesized that immersion of koi in propofol would induce general anesthesia.
Materials and Methods
Animals and husbandry—Ten adult koi were used in the study. The koi were housed in a fiberglass circular tank holding approximately 1,500 L of dechlorinated municipal water. The water of the tank equilibrated with climate-controlled room temperature (23.1° to 23.3°C [73.6° to 74.0°F]), and a 50% water change was performed every 2 weeks or more frequently to maintain the concentration of ammonia < 0.5 mg/L. Temperature, pH, and ammonia and nitrate concentrations of the water were measured and recorded daily. A 12-hour light-dark cycle was maintained. Fish were monitored on a daily basis and fed a pelleted dieta formulated for ornamental koi on Monday, Wednesday, and Friday. Fish had no known history of illness and were considered healthy on the basis of results of visual examination and examination of skin and gill scrapings within 1 week of the start of the study. All procedures were reviewed and approved by the Institutional Animal Care and Use Committee of North Carolina State University.
Induction of anesthesia—Each koi (n = 10) was immersed in 1 of 4 concentrations of propofolb (1, 2.5, 5, and 10 mg/L) in a crossover design, with at least a 1-week washout period between immersion at each concentration. The concentrations were chosen on the basis of data from a pilot study.
Each fish was caught with a fishnet and moved from the housing tank into an anesthetic induction container that contained the appropriate concentration of propofol in 6 L of water for the 1, 2.5, and 5 mg/L concentrations and in 8 L of water for the 10 mg/L concentration. Water in the anesthetic induction containers was obtained from the housing tank, and the pH of the anesthetic induction solution containing propofol was recorded during each trial. Immediately after the transfer, a stopwatch was started and initial opercular rate was recorded. Fish were considered anesthetized when they lost voluntary movement and the righting reflex or when their opercular rate decreased to approximately a quarter of the initial opercular rate and they did not respond to manual stimulation. Once these criteria were met, anesthetic induction time was recorded. After anesthetic induction, fish were removed from the water and weighed on a gram scale.c Heart rate was measured by placing a Dopplerd crystal on the ventrum at the heart level. A 25-gauge needle was inserted into the caudal epaxial muscle between scales to assess the response to a noxious stimulus. Response to the noxious stimulus was judged as either positive or negative. A positive response included gross purposeful movement but did not include muscle fasciculation or shuddering. Fish were then placed into a recovery container that contained approximately 10 L of fresh dechlorinated water, and recovery opercular rate was recorded. Recovery was characterized as regaining of the righting reflex when fish were intentionally pushed into lateral recumbency. Once the fish recovered successfully, time of recovery and postanesthesia opercular rate were recorded. Following each experiment, the fish were returned to the housing tank and monitored routinely. Throughout the induction and recovery period, the koi were visually monitored for the physiologic parameters measured (eg, opercular rate) as well as for any unusual unexpected parameters.
Maintenance of anesthesia—Fish (n = 9; 1 fish was lost to the study for reasons not associated with the study) were removed from their housing tank and transferred into a bucket with fresh dechlorinated water to obtain initial opercular rate. Fish were then moved into an anesthetic induction container containing propofol (5 mg/L) and 6 L of water from the housing tank. The anesthetic induction dose was chosen on the basis of favorable characteristics (anesthetic induction time and physiologic parameters) from the anesthetic induction–only experiments in this study. Induction time was recorded, and once the fish was anesthetized, it was transferred onto a recirculating water system (FADS) that contained 10 L of water with propofol (3 mg/L). The recirculating fish anesthesia system continually delivered anesthetic water at a flow rate of approximately 2.0 L/min through the buccal cavity and across the gills.12 Each fish was maintained on the FADS for 20 minutes. Immediately after the fish was placed on the FADS, the opercular rate and heart rate as measured with the Doppler crystal were recorded and approximately 0.3 to 0.4 mL of blood was collected from the hemal arch for blood gas analysis with 2 analyzers.e,f After 10 minutes on the recirculating water system, heart rate and opercular rate were again recorded and response to a noxious stimulus was assessed by inserting a 25-gauge needle into the caudal epaxial muscle. Fish were kept on the FADS for another 10 minutes, and heart rate, opercular rate, blood gas analysis, and weight were obtained at the end of the 20-minute anesthesia period. Fish were then placed into a recovery container. Time to recovery and opercular rate following recovery were recorded. Following recovery from anesthesia, fish were returned to the housing tank and routinely monitored. Throughout the induction, maintenance, and recovery periods, koi were visually monitored for the physiologic parameters measured (eg, opercular rate) as well as for any unusual unexpected parameters.
pH measurement—The pH of propofol combined with fresh dechlorinated water at concentrations of 1, 2.5, 3, 5, and 10 mg/L was measured with a benchtop pH meter.g The pH meter was calibrated with buffer solutions at a pH of 4 and 7. The 1 mg/L solution was made in a 1-L sample, and the rest of the samples were tested in 100-mL samples. Propofol was measured in a 1-mL syringe for the 1-L sample and in a 0.3-mL syringe for the remaining samples. Five measurements were taken from each sample after the pH meter was equilibrated for approximately 9 minutes.
Statistical analysis—Single-dose trial data were analyzed with statistical software.h All of the variables were tested for normality of distribution by the Kolmogorov-Smirnov test. Differences in anesthetic induction time, heart rate, opercular rate, and weight were assessed on the basis of linear mixed models with statistical software.h When significance was detected, a post hoc pairwise comparison was conducted by use of t tests of least squares means with a Bonferroni correction applied. Differences in opercular rate and heart rate at various time points were assessed on the basis of paired t tests and pairwise comparisons with a Bonferroni correction applied. Response to the noxious stimulus was compared by means of the Fisher exact test. Significance was set as P < 0.05. Data are reported as mean ± SD.
Results
Anesthetic induction trial—Initial mean ± SD weight of the 10 koi was 325 ± 81 g and did not change significantly (P = 0.84) over the duration of the experiment (ending weight, 333 ± 80 g). There were no changes noted on physical examination (ie, eyes, gills, or scales) during the course of the experiment.
At a propofol concentration of 1 mg/L, none of the koi were anesthetized, although they subjectively appeared sedated (slowed swimming and loss of equilibrium but responsive to handling). Anesthetic induction time and recovery time at propofol concentrations of 2.5, 3, 5, and 10 mg/L were reported (Figure 1). All of the koi recovered uneventfully. Heart rate and opercular rates were reported (Figure 2). At a propofol concentration of 10 mg/L, the water in the anesthetic induction container was no longer clear owing to the addition of a greater amount of the opaque-white propofol solution, and the preanesthetic opercular rate for 2 koi could not be visualized because of their body position and the cloudiness of the water. Although opercular rates were not significantly (P = 0.11) different between the groups, 1 koi required resuscitation via manual forward swimming when it was anesthetized at a propofol concentration of 10 mg/L because its opercular movement stopped at anesthetic induction. At propofol concentrations of 2.5, 5, and 10 mg/L, there was a response to a needle insertion in 1 of 10, 2 of 10, and 0 of 10 fish, respectively. Some koi showed nystagmus-like vertical eye movements when they were anesthetized and also when they were heavily sedated (1 mg/L).
Box-and-whiskers plots of anesthetic induction time (A) and recovery time (B) for 10 adult koi (Cyprinus carpio; mean ± SD weight, 325 ± 81 g) exposed to each of 3 concentrations of propofol (2.5, 5, and 10 mg/L) with a 1-week washout period between trials. Values with different letters are significantly (P < 0.05) different. The box represents the interquartile (25th to 75th percentile) range, the horizontal line represents the median, and the whiskers represent the range.
Citation: Journal of the American Veterinary Medical Association 245, 11; 10.2460/javma.245.11.1286
Box-and-whiskers plots of heart rate (A) and opercular rate before (Pre) and after (Post) anesthetic induction (B) for the fish in Figure 1. Values with different letters are significantly (P < 0.05) different.
Citation: Journal of the American Veterinary Medical Association 245, 11; 10.2460/javma.245.11.1286
Anesthesia maintenance with propofol—Anesthetic induction time and recovery time from FADS (mean ± SD) were 5.0 ± 2.0 minutes and 11.6 ± 8.0 minutes, respectively. Physiologic data (Figure 3) and biochemical data (Table 1) were summarized. Blood collected from all fish at the time of anesthetic induction was successfully analyzed, but because of technical difficulties with the blood gas analyzer, there were results for only 4 fish after 20 minutes on the FADS. Therefore, no mathematical comparison was performed between the anesthetic induction and 20-minute blood samples. A positive response to the noxious stimulus was significantly (P = 0.04) more frequent (6/9) at 20 minutes, compared with the proportion (1/9) that responded at 10 minutes. While on the FADS, 8 of 9 koi had weak fin movements.
Box-and-whiskers plots of heart rate (A) and opercular rate (B) for koi (n = 9) anesthetized with propofol (5 mg/L) in which anesthesia was maintained with propofol (3 mg/L) delivered with a FADS. Values with different letters are significantly (P < 0.05) different.
Citation: Journal of the American Veterinary Medical Association 245, 11; 10.2460/javma.245.11.1286
Results of blood gas analysis for caudal hemal arch blood samples obtained from adult koi (Cyprinus carpio; n = 9) following anesthetic induction with propofol (5 mg/L) and after maintenance of anesthesia for 20 minutes with propofol (3 mg/L) delivered with a FADS.
| Biochemical variable | Following anesthetic induction | 20 minutes after anesthetic induction | Awake koi reference range14 |
|---|---|---|---|
| pH | 7.27 ± 0.20 (n = 8) | 7.35 ± 0.15 (n = 4) | 7.39 ± 0.16 |
| Lactate (mmol/L) | 3.98 ± 2.49 (n = 8) | 2.58 ± 1.06 (n = 4) | 1.43 ± 1.37 |
| Po2 (mm Hg) | 71.13 ± 43.00 (n = 8) | 34.25 ± 17.59 (n = 4) | 18.0 ± 9.8 |
| Pco2 (mm Hg) | 10.81 ± 2.22 (n = 8) | 12.6 ± 2.20 (n = 4) | 16.93 ± 4.09 |
| Base excess (mmol/L) | –20.14 ± 3.48 (n = 7) | –18.5 ± 4.20 (n = 4) | –14.8 ± 4.0 |
| HCO3 (mmol/L) | 5.83 ± 1.38 (n = 8) | 7.18 ± 1.80 (n = 4) | Unknown |
Results reported as mean ± SD.
Two fish had a prolonged recovery (> 4 hours after being moved into a recovery bucket). One never fully recovered and was euthanized the next day (approx 15 hours after the start of the recovery phase). For the fish that recovered, heart rates at induction, 10 minutes, and 20 minutes were 72, 52, and 28 beats/min, respectively. For the fish that was euthanized, heart rates at induction, 10 minutes, and 20 minutes were 72, 44, and 40 beats/min, respectively. For the fish that recovered, opercular rates at induction, 10 minutes, and 20 minutes were 8, 64, and 60 beats/min, respectively. For the fish that was euthanized, opercular rates at induction, 10 minutes, and 20 minutes were 24, 44, and 20 beats/min, respectively.
Mortality rate for the koi anesthetized for 20 minutes with propofol was 11% (1/9 koi; 95% CI, 0% to 35%). A necropsy report for this koi revealed no gross or microscopic lesions in the gills, intestines, liver, and other organs.
The pH of the tank water was 7.68. The pH of 1% propofol (10 g/L) was 7.60. The pH of the 1, 2.5, 3, 5, and 10 mg/L solutions was 7.78, 7.65, 7.66, and 7.67, respectively.
Discussion
In the present study, immersion of koi in propofol at concentrations of 2.5, 5.0, and 10.0 mg/L resulted in anesthesia. These findings are consistent with results of another recent study,11 in which silver catfish (Rhamdia quelen) were successfully anesthetized with immersion in propofol at concentrations of 2.5, 5.0, 10.0, and 12.0 mg/L. Clinically, immersion with propofol appears to have a dose-response effect in koi. Anesthetic induction time decreased and recovery time increased as the concentration of propofol increased. It has been recommended that with immersion anesthesia, the anesthetic induction and recovery times should be 3 to 5 minutes or less.13 On the basis of concentrations tested in this study, propofol at 5 mg/L provided anesthetic induction and recovery times within this range for koi. However, this concentration may not be applicable for other fish species. Gressler et al11 reported that a concentration of propofol at 10 mg/L was satisfactory to anesthetize large catfish (mean ± SD weight, 155.74 ± 2.45 g). This may reflect different pharmacodynamic effects in this species of fish or criteria that are different between studies. In the study by Gressler et al,11 no physiologic parameters were measured to aid in assessment of an acceptable concentration. Koi anesthetized in the present study with 10 mg/L were rapidly anesthetized and lived through recovery; however, on the basis of the cessation of opercular movement, this dose was considered too high.
Propofol decreased heart rate, and higher concentrations had a more pronounced effect. At the 10 mg/L concentration, 3 fish had a heart rate < 10 beats/min. Although cardiac output was not measured, this severe bradycardia likely resulted in a considerably decreased cardiac output. Because propofol is known to cause respiratory depression in mammalian species, a decrease in opercular rate was expected. Interestingly, following only anesthetic induction doses, opercular rates were not significantly different between various concentrations or within the same concentration at different time points. The inability to show differences between the groups was likely due to type II error caused by the amount of variability in opercular rates and the small number of koi studied. One koi that was anesthetized in propofol at 10 mg/L had complete cessation of opercular movement. This fish was supported by manual ventilation (gently pushing the carp through the water) until spontaneous opercular movement resumed.
Following anesthetic induction only, 3 of 20 fish responded to a noxious stimulus (needle insertion) with the 2.5 and 5 mg/L concentrations and no fish responded at the 10 mg/L concentration. Fish that did respond to needle insertion may have required a higher dose or longer exposure to prevent response to the noxious stimuli.
Maintenance of anesthesia with propofol at 3 mg/L for 20 minutes resulted in a greater than expected mortality rate. Although all fish were anesthetized with the same concentrations of propofol for 20 minutes on an FADS device, 2 fish had prolonged recoveries and 1 of those fish was euthanized. The cause of the delayed recoveries or death was unknown. Gross and histologic examination did not reveal any abnormalities, and inspection of the heart rate and opercular rate in those fish did not reveal any particular problem. Potential explanations include exaggerated CNS or cardiovascular depression. It is also possible that chemicals in the vehicle may have caused unwanted effects. Interestingly, in the FADS trial, opercular rate following anesthetic induction was significantly lower than baseline in koi anesthetized at a propofol concentration of 5 mg/L, which was not seen in the simple anesthetic induction trial. The mean anesthetic induction time in the anesthetic induction trial was 3.83 ± 1.1 minutes, whereas the mean anesthetic induction time for fish to be placed on the FADS was 5.0 ± 2.0 minutes. Therefore, the most likely cause for the difference between the 2 groups (both immersed in propofol at a concentration of 5 mg/L) was that the koi in the FADS trial were more deeply anesthetized after being exposed for a longer duration of time.
While on the FADS device, 8 of 9 koi had weak fin movements, and 1 and 6 fish responded to the noxious stimulus at 10 and 20 minutes after anesthetic induction, respectively. This indicated that the koi were at a lighter plane of anesthesia at 20 minutes. The mostly likely reason for this change in anesthetic depth was because the maintenance concentration of propofol chosen for the present study (3 mg/L) was inadequate. Fish anesthetized with MS-222 are typically anesthetized with a higher anesthetic concentration than what they are maintained with. On the basis of this practice, the maintenance concentration of propofol chosen was a fraction (60%) of the 5 mg/L administered to induce general anesthesia, which might have been inadequate. Alternatively, uptake and degradation of propofol could have occurred over the 20 minutes. Plasma propofol concentrations were not evaluated, so this hypothesis cannot be further pursued. Considering a mortality rate of 11% (1/9), further studies are needed to evaluate the safety of propofol as a maintenance agent as well as determine the optimal concentration.
Interpretation of biochemical parameters from blood gas results was limited by the small and inconsistent sample sizes at 20 minutes after anesthetic induction. In addition, blood samples were collected from the hemal arch, which includes both an artery and vein. Because it is impossible when taking the sample to tell whether the sample came from the artery or vein, interpretation of parameters such as oxygen tension is limited. Despite this unknown factor, the biochemical parameters from the 2 time frames were clinically similar except for mean Po2 which cannot be further interpreted. Mean Pco2 following anesthetic induction and 20 minutes after anesthesia was 10.8 ± 2.2 mm Hg and 12.6 ± 2.2 mm Hg, respectively. Reference range for Pco2 in awake koi has been reported as 14.1 to 19.8 mm Hg.14 Therefore, the koi in this study were mildly hyperventilating, most likely because of the artificial ventilation of water being run over the gills. Given that opercular rate decreased at anesthetic induction, this was most likely due to the rate of water flow over the gills, which would serve as artificial ventilation.
Blood pH and lactate concentration indicated acidemia with hyperlactatemia. Blood pH at anesthetic induction and 20 minutes after anesthetic induction was 7.27 ± 0.20 and 7.35 ± 0.15, respectively, and lactate concentration was 3.98 ± 2.49 mmol/L and 2.58 ± 1.06 mmol/L, respectively. Interpretation of these values needs to take into account the reference range for awake koi as well as values reported with other anesthetics used in koi. Results of a previous study14 indicate that awake koi have a blood pH reference range as low as 7.28 and lactate concentration as high as 2.38 mmol/L. Koi anesthetized with MS-222 had a mean pH of 7.27 ± 0.09 and mean lactate concentration of 5.33 ± 1.1 mmol/L.14 Koi anesthetized with alfaxalone had a median pH of 7.18 (range, 7.09 to 7.32) and median lactate concentration of 4.1 (range, 2.9 to 5.9) mmol/L.15 Although these values would be considered unacceptable in most mammals following routine anesthesia, they are generally considered clinically acceptable in anesthetized fish.14,15 The finding that lactate concentration in the present study was within clinically reported values and actually decreased over the 20 minutes of anesthesia implied that the koi had sufficient oxygenation and perfusion. Further study with a larger sample size and adequate blood samples is necessary to compare the change in pH and lactate concentration under anesthesia induced by propofol immersion.
The main limitation of the present study was the small sample size that could have resulted in a type II error. Additionally, assessing the depth of anesthesia in fish is subjective, which might have caused variability in measurement of certain parameters, such as anesthetic induction time, between anesthetic induction and the FADS trials. Furthermore, the efficacy of an anesthetic agent in fish might be associated with temperature, pH, and size of fish, and variability in these factors would need to be considered as limitations. Nevertheless, we suggest that these variables would have a minimal impact on the present study because temperature of the room and water was controlled, pH of the water was tested daily, and fish did not differ in weight over the course of the experiment. Although propofol by immersion induced and maintained general anesthesia in koi, it is unknown how other fish species might react. Future studies in different fish species and with a larger sample size are suggested to improve the understanding of the effects of propofol in fish.
Propofol might be an alternative for short-term anesthesia in fish and may provide some advantages. Unlike MS-222, propofol does not require buffering to control pH when it is used as an immersion agent. Similar to MS-222, determination of weight of fish is not necessary, although concentration of propofol may require adjusting depending on general fish size. Furthermore, propofol might provide effective general anesthesia in other species, such as Gulf of Mexico sturgeon, in which MS-222 is not effective.8 Lastly, subanesthetic concentrations of propofol might be useful as a sedative for brief handling and transportation.
Conversely, the use of propofol in fish has its limitations. In addition to the unexpected morbidity rate (22% [2/9]) and mortality rate (11% [2/9]) when used for maintenance of anesthesia, propofol is not licensed for use in food fish, which would limit its use outside of the research or aquarium setting. A further consideration of the use of propofol in fish is the effect of the other components in the emulsion of propofol. In addition to the active ingredient, diisoproylphenol, propofol emulsion contains soybean oil, glycerol, egg lecithin, EDTA, and sodium benzoate. Fish necessarily are immersed in the anesthetic solution, so it needs to be considered whether any of these agents in the doses administered are toxic to fish. According to the material safety data sheet for the propofol used in this study, no ecological toxicity information is available.16 However, it has been reported that propofol is toxic to aquatic animals, and the median lethal concentration for bluegill sunfish was 0.62 mg/L for 96 hours.17 In attempting to interpret these data, it should also be noted that the vehicle for propofol delivery can vary with the manufacturer. Some products have benzyl alcohol and some have EDTA, and the concentrations of those chemicals can vary. The propofol used in this study contained EDTA and no benzyl alcohol. Interestingly, EDTA has been shown to be toxic to fish too, although the median lethal concentration for fish is unknown.18 Although the koi in this study had short-term exposure to these chemicals, it is possible that the prolonged recovery time in fish anesthetized for 20 minutes with propofol could be related to the vehicle component as well as the active ingredient. More research needs to be performed to determine the effects of these drugs on aquatic animals.
In conclusion, propofol concentrations of 2.5 mg/L or higher induced anesthesia in koi. Induction of anesthesia with 5 mg of propofol/L was clinically satisfactory. Maintenance of anesthesia with propofol was associated with prolonged recovery in 2 of 9 koi and death in 1 of 9 (11%; 95% CI, 0% to 35%).
ABBREVIATIONS
| FADS | Fish anesthesia delivery system |
| GABA | γ-Aminobutyric acid |
| MS-222 | Tricaine methanesulfonate |
Blackwater Farms Maximum Growth, Blackwater Creek Koi Farms Inc, Jacksonville, Fla.
Propofol, APP Pharmaceuticals Inc, Schaumburg, Ill.
CS 5000 portable scale, Ohaus Corp, Parsippany, NJ.
Ultrasonic Doppler Flow Detector Model 811-B, Parks Medical Electronics Inc, Aloha, Ore.
iStat Portable Chemical analyzer, Heska Corp, Waukesha, Wis.
CG-4 Abbott Point of Care Inc, Abbott Park, Ill.
Accumet Basic AB15 + pH meter, Fisher Scientific Inc, Pittsburgh, Pa.
SAS, version 9.3, SAS Institute Inc, Cary, NC.
References
1. US FDA. Fish and fishery products hazards and control guidance. 4th ed. Rockville, Md: US FDA, 2011.
2. Palmer LM, Mensinger AF. Effect of the anesthetic tricaine (MS-222) on nerve activity in the anterior lateral line of the oyster toadfish, Opsanus tau. J Neurophysiol 2004; 92: 1034–1041.
3. Hunn JB. Dynamics of MS-222 in the blood and brain of freshwater fishes during anesthesia. US Fish and Wildlife Service investigations in fish control. No. 42. Washington, DC: US Fish and Wildlife Service, 1970.
4. Stoskopf M, Posner LP. Anesthesia and restraint of laboratory fish. In: Fish RE, Brown MJ, Danneman PJ, et al, eds. Anesthesia and analgesia in laboratory animals. 2nd ed. Amsterdam: Elsevier, 2008; 519–534.
5. Stoelting R, Hillier S. Nonbarbiturate intravenous anesthetic drugs. Pharmacol Physiol Anesth Pract 2006; 4: 167–174.
6. Delgado L, Schmachtenberg O. Immunohistochemical localization of GABA, GAD65, and the receptor subunits GABAAα1 and GABAB1 in the zebrafish cerebellum. Cerebellum 2008; 7: 444–450.
7. Uematsu K, Shirasaki M, Storm-Mathisen J. GABA-and glycine-immunoreactive neurons in the spinal cord of the carp, Cyprinus carpio. J Comp Neurol 1993; 332: 59–68.
8. Fleming GJ, Heard DJ, Floyd RF, et al. Evaluation of propofol and medetomidine-ketamine for short-term immobilization of Gulf of Mexico sturgeon (Acipenser oxyrinchus de soti). J Zoo Wildl Med 2003; 34: 153–158.
9. Hunn JB, Allen JL. Movement of drugs across the gills of fishes. Annu Rev Pharmacol 1974; 14: 47–54.
10. Hunn JB, Schoettger RA, Willford WA. Turnover and urinary excretion of free and acetylated MS 222 by rainbow trout, Salmo gairdneri. J Fish Board Can 1968; 25: 25–31.
11. Gressler LT, Parodi TV, Riffel AP, et al. Immersion anaesthesia with tricaine methanesulphonate or propofol on different sizes and strains of silver catfish Rhamdia quelen. J Fish Biol 2012; 81: 1436–1445.
12. Harms C. Anesthesia in fish: zoo and wild animal medicine: current therapy. Amsterdam: Elsevier Health Sciences, 2008.
13. Marking LL, Meyer FP. Are better anesthetics needed in fisheries? Fisheries (Bethesda, Md) 1985; 10: 2–5.
14. Hanley CS, Clyde VL, Wallace RS, et al. Effects of anesthesia and surgery on serial blood gas values and lactate concentrations in yellow perch (Perca flavescens), walleye pike (Sander vitreus), and koi (Cyprinus carpio). J Am Vet Med Assoc 2010; 236: 1104–1108.
15. Minter LJ, Bailey KM, Harms CA, et al. The efficacy of alfaxa-lone for immersion anesthesia in koi carp (Cyprinus carpio). Vet Anaesth Analg 2014; 41: 398–405.
16. MSDS data sheet. Propofol. Available at: editor.apppharma.com/MSDSSheets/MSDS_Propofol_Inj_Emulsion_713.pdf. Accessed Jan 1, 2014.
17. Mankes RF. Propofol wastage in anesthesia. Anesth Analg 2012; 114: 1091–1092.
18. Batchelder TL, Alexander HC, McCarty WM. Acute fish toxicity of the versene family of chelating agents. Bull Environ Contam Toxicol 1980; 24: 543–549.
