Assessing Local Anesthesia in Octopus bimaculoides to Provide Analgesia

Abstract

Despite the substantial growth of cephalopod research over the past 15 years, effective pain management for these animals remains largely unstudied, and no US federal regulations currently address cephalopod analgesia. This 2-phase study evaluated the feasibility and efficacy of lidocaine, bupivacaine, and extended-release bupivacaine injections for producing analgesia in juvenile Octopus bimaculoides in an established model of arm regeneration. Phase I served as a pilot to assess adverse effects, practicality, and histopathologic consequences of drug administration. Forty-nine animals were randomized to the saline control or treatment groups. Phase II assessed lidocaine, bupivacaine, and saline in an additional 34 animals, using behavioral metrics at multiple time points, including a posture and gait analysis during a walking assessment and mechanical allodynia evaluation via the von Frey filament test. Health indicators, such as weight change and respiratory rate, were also monitored. Finally, wound healing and arm regeneration were followed with photo documentation. Although all animals displayed statistically significant changes in arm posture and gait immediately postsurgery, these changes invariably resolved within 72 hours irrespective of treatment group. Moreover, health and regenerative outcomes did not differ significantly among the treatment groups. The failure of local anesthetic efficacy highlights the need for alternative analgesic regimes for octopus research.

Introduction

Cephalopod research has accelerated over the past 15 years across ecological, evolutionary, neurobiological, molecular, and behavioral studies.^1,2 For example, the number of publications using octopuses increased by an average of 10% annually from 2011 to 2021.¹ During this time, a new biologic understanding has emerged regarding cephalopod responses to noxious stimuli. Earlier studies documented nociceptive behaviors in cephalopods but questioned whether these responses indicated a capacity to experience pain^3–7; however, more recent literature has argued that these animals must experience some level of pain, as evidenced by learned avoidance^8,9 and wound-directed attention.^8–10 Reflecting this shift in attitude, many countries have included cephalopods in their regulatory frameworks.^11–17 Most notably, the European Union published “Guidelines for the Care and Welfare of Cephalopods in Research” in 2015 requiring that animals that may suffer pain after general anesthesia should be treated with perioperative analgesics. These guidelines recognized that, at the time of publication, there was no information on the efficacy of any analgesics in cephalopods.¹⁸

Despite the need for enhanced and standardized care of cephalopods, NIH and USDA have not yet established specific regulations regarding cephalopods at the time of this publication. In 2023, NIH issued a Request for Information to gather comments on a regulatory framework for cephalopod welfare, including critical areas such as the use of analgesia.¹⁹ These imperatives form the basis of our study, which aims to address the use of analgesics in cephalopod research and veterinary medicine. Cephalopods have undergone a range of invasive procedures similar to those performed in vertebrates; however, published techniques often lack detailed descriptions, particularly regarding the use of analgesia.¹⁸ The lack of adequate pain management in vertebrates is a failure of proper animal welfare and increases variability in research outcomes. In cephalopods, the absence of analgesia similarly raises welfare concerns, as it results in an unknown level of pain and distress.² Cephalopod research does offer several trusted methods of anesthesia,^9,20–23 but analgesic protocols for adequate desensitization to the immediate and long-lasting effects of assumedly painful procedures are limited.

When selecting an analgesic drug, it is essential to consider the physiology specific to the class of animals. The conservation of sodium channels in invertebrate species suggests that local anesthetics, such as lidocaine, may provide analgesic effects in cephalopods.²⁴ Evidence supporting this conjecture includes a range of experimental approaches from a cellular to an organismal level. Electrophysiological studies have confirmed this possibility when using lidocaine in both excised limbs and anesthetized cephalopods, including Octopus bocki, Ascarosepion bandense, and Octopus vulgaris.^8,22,25 In conscious animals, behavioral efficacy in the context of induced noxious stimuli has been evaluated through wound-directed behaviors, conditioned place preference, and responses to mechanical stimulation, most commonly associated with acetic acid exposure. Using these paradigms, lidocaine has been reported to reduce nocifensive and painful behaviors in Octopus bocki, Sepia pharaonis, and Ascarosepion bandense.^8,10,22 In contrast, a study evaluating nonlocal anesthetic agents employed von Frey testing, behavioral scoring, and electrophysiological recordings, reporting efficacy for high-dose buprenorphine in Euprymna berryi,⁹ while low-dose buprenorphine, ketorolac, ketamine, dexmedetomidine, and acetaminophen demonstrated only modest efficacy.⁹

Pharmacokinetics and pharmacodynamics are also crucial factors to consider. In the dwarf cuttlefish, Ascarosepion bandense, the onset of lidocaine action following injection has been reported to be approximately 3 minutes 20 ± 46 seconds with a duration of at least 11 ± 4 minutes.²² The temporal activity of other local anesthetics has not yet been established in cephalopods. In mammals, bupivacaine has an onset of approximately 20–30 minutes and a duration of approximately 4–6 hours,²⁶ while extended-release bupivacaine (Nocita) can reach a maximum concentration within 30–60 minutes and last up to 72 hours due to the liposomal distribution of the drug over time.²⁷ Redosing lidocaine and bupivacaine to manage potential pain over a longer timeline may be inappropriate, as it poses a risk of acute cardio- and neurotoxicity in mammals.²⁶ In addition, redosing necessitates multiple injections. Therefore, drugs with extended action may provide a more practical option for sustained relief.

Finally, when considering the practicality of drug selection, easily accessible analgesic options warrant further exploration, particularly for research laboratories that may not have access to controlled substances. Local anesthetics are not controlled and are typically easy to obtain. In addition, lidocaine and bupivacaine are relatively cost effective and offer a reasonable shelf life after vial puncture.^28,29 Evaluating the efficacy of an extended-release bupivacaine, which has not yet been explored in cephalopods, represents a logical next step in the progression of this research. Despite the increased expense and limited stability after opening, the extended-release option may offer worthwhile benefits through prolonged analgesic efficacy compared with a standard formulation presenting a shorter duration of action.

The objective of our 2-phase study was to evaluate local anesthetics for juvenile O. bimaculoides. Juvenile cephalopods were selected as a model because their full life histories can be monitored from hatching and because arm loss and regeneration are a natural part of their biology,^30,31 making them well suited for studies of injury and recovery.^32–34 Accordingly, this study aimed to identify analgesic options that mitigate procedural pain and distress while preserving the integrity of the regenerative cephalopod model.³⁴ Phase I involved the preliminary assessment of arm injections with 3 local anesthetic agents: lidocaine, bupivacaine (low, intermediate, and high dosing), and extended-release bupivacaine (low and high dosing), compared with a saline control. Advancement to phase II was contingent upon 2 predefined criteria: the absence of pronounced morbidity and mortality and the feasibility of administration. Phase II evaluated the local anesthetic efficacy associated with amputation of the second right arm (R2) in 6-week-old juvenile octopuses.

To evaluate efficacy, multiple measures were assessed at baseline, following drug administration and amputation of R2, and at subsequent time points up to 72 hours and 21 days, depending on the metric studied. Both novel and established metrics were employed. Posture and gait were evaluated using an adapted walking assessment. While gait analysis has not yet been established for octopus or other cephalopod species, similar approaches have been used in other taxa, including sows, cattle, and fish.^35–37 Our study adapted the posture and gait analysis for use with octopus and so establishes a novel efficacy metric for potential indicators of pain and discomfort in this species.

Mechanical allodynia was evaluated using the von Frey filament test, as this test is not only a standard in the assessment of pain in rodents,^38,39 but it has also been used to demonstrate nociceptive responses in other cephalopods.^3,7,9 Additional health indicators, including weight change and respiratory rate, were monitored to provide a systemic evaluation of physiologic status. Finally, because the effects of analgesic drugs on wound healing and tissue regeneration have not previously been studied in cephalopods, these parameters were assessed to determine whether the local anesthetic agents interfered with the animals’ innate regenerative capacity.

We hypothesized that extended-release bupivacaine would produce the most prolonged analgesic effect, followed by bupivacaine and then lidocaine. This enhanced analgesia was expected to be reflected in health and behavioral metrics. In addition, we anticipated that none of the local anesthetic agents would adversely affect the healing and regeneration process.

Materials and Methods

Ethics review.

All procedures were approved by the University of Chicago IACUC and were in compliance following the European directive related to the use of cephalopods as experimental organisms (Directive 2010/63/EU). The University of Chicago animal care program is accredited by AAALAC International and has a Public Health Service assurance with Office of Laboratory Animal Welfare.

Study design.

The experiments were separated into 2 phases to ensure that the drugs tested did not cause overt adverse effects. Within phase I, we compared 3 treatment groups: 12 mg/kg 2% lidocaine (Covetrus North America; n = 8), 12 mg/kg bupivacaine (2.5 mg/mL; n = 6; Hospira, Lake Forest, IL), and 35 mg/kg extended-release bupivacaine (13.3 mg/mL; n = 6; Nocita; Elanco, Indianapolis, IN) against 0.9% sodium chloride saline (Baxter Healthcare, Deerfield, IL) control group (n = 7) over an observational period of 3 days. The lidocaine dose was based on previously published work evaluating its efficacy in Sepia pharaonis when coadministered with acetic acid.¹⁰ The bupivacaine dose was subsequently extrapolated from this established lidocaine dose to represent a ratio accepted among other taxa. Because extended-release bupivacaine is a novel formulation that has not been evaluated in aquatic species, its dose was derived by extrapolation from a rodent study.³⁹ The volume of saline used for control treatments was selected to fall within the range of the administered drug volumes. Drug volumes were not adjusted for dose and concentration differences and resulted in the administration of approximately 0.1 µL lidocaine, 1 µL bupivacaine, and 0.5 µL extended-release bupivacaine, depending on cohort weight averages. Saline was administered at 0.5 µL. No surgical intervention occurred within phase I. Dosing was achieved by weighing in batches of octopuses (n = 3–7) and calculating an average weight for that group.

Phase I assessed preliminary health parameters, including posture and gait changes, the presence of skin lesions, and appetite level. Posture and gait changes were assessed based on the amount of arm curling or tucking as well as the frequency of use. Skin lesions were evaluated if tissue loss beyond the site of injection was observed. Appetence was assessed as a systemic marker for health and welfare by scoring whether animals took food within one minute on a yes or no basis. Injection parameters included estimating the amount of leakage that occurred at injection, the presence of subcutaneous bubbles, and the number of injection attempts. Due to adverse consequences at the injection site, reduced doses were explored for both 6 mg/kg bupivacaine (n = 4) and 3 mg/kg bupivacaine (n = 8) as well as 8.75 mg/kg extended-release bupivacaine (n = 10), which resulted in the administration of approximately 0.1 µL lidocaine, 0.25 µL bupivacaine, 0.12 µL extended-release bupivacaine, depending on cohort weight averages, as well as 0.5 µL saline.

Phase II compared 2 treatment groups: 2% lidocaine (12 mg/kg) (n = 11) and bupivacaine (3 mg/kg, 2.5 mg/mL) (n = 11) against the saline control group (n = 12) over a period of 1–3 weeks. Drug volumes were not adjusted for dose and concentration differences, as in phase I, and resulted in the administration of approximately 0.1 µL lidocaine and 0.25 µL bupivacaine, depending on individual weights. The volume of saline was retained at 0.5 µL from phase I. Juveniles were weighed individually at baseline for drug dosing, immediately after amputation, on day 7, and on day 21. Respiratory rate, walking assessment parameters, von Frey assessment parameters, and healing parameters were assessed at baseline, immediately postamputation, and days 1, 2, and 3. Regeneration parameters were assessed at days 7, 14, and 21.

A negative control group (amputation without prior injection) was not included, as extensive experience in the collaborating Ragsdale laboratory has demonstrated reproducible wound healing and regeneration following identical amputations performed without prior injections.³⁴ These observations demonstrated clear, robust, reproducible arm regeneration and therefore informed the design of the current study. Accordingly, and consistent with the principles of Replacement, Reduction, and Refinement (3Rs), inclusion of an additional control group was not considered necessary.

Experimental animals.

Wild-caught O. bimaculoides clutches with the brooding females were purchased from Aquatic Research Consultants (field-collection venture operated by Dr. Chuck Winkler). Upon arrival, each clutch was separated from its brooding female and placed near an area of high oxygenation within a 250-gallon 3-tiered artificial saltwater (ASW) tank equipped with a recirculating filtration system. The ASW within the tank system was prepared from pharmaceutically pure sea salt (33 gallon/liter; Tropic Marin “classic”; Tropic Marin, Wartenberg, Germany) dissolved in reverse osmosis water. The water quality and temperature of the tank system were monitored daily. The eggs were monitored until hatching, which was induced via water agitation during the final stages of development. Within both phases, newly hatched animals from 2 clutches were assigned to isolated mesh containers (Imagitarium Aquarium Net Breeder) without systematic allocation (simple randomization). All individuals were of similar size and could not be sexed, as anatomic differences do not appear until sexual maturity at approximately 5 months of age.⁴⁰ Animals exhibiting prior injury to R2 at the time of experimentation were excluded. The animals were fed approximately 5 adult brine shrimp (Artemia salina; Carolina Biologic Supply, Burlington, NC) or 5 adult mysids (Americamysis bahia; US Mysids, Saint Augustine, FL) daily until they were 6 weeks of age. To ensure consistent consumption of food while on study, animals were hand-fed daily with pieces of wild-caught, uncooked, shelled frozen shrimp (Whole Foods, Austin, TX) that were approximately the size of their interocular distance. The animals studied were between 6 and 10 weeks of age.

Anesthesia (phase I and II).

Animals were anesthetized before imaging, injections, and amputations. Juveniles were transferred to 100 mm diameter glass Petri dishes (catalog no. BR455743; Millipore Sigma, Burlington, MA) and anesthetized in a solution of 2% ethanol in filtered artificial sea water (fASW). It was determined that juveniles reached a deep plane of anesthesia once they exhibited a decreased respiratory rate, lack of righting reflexes when disturbed, and generalized pallor.^20,23 After the conclusion of injections or amputations, animals recovered in a Petri dish filled with fresh fASW. Juveniles were determined to have reached full recovery based on increased respiratory rate, return of righting reflexes, and return to normal chromatophore color and texture.^20,23 Juveniles were returned to their home tanks, in phases I and II, or were reassessed for behavioral parameters at time 0 of phase II.

Imaging (phase I and II).

While in the Petri dish filled with anesthetic solution, baseline imaging of the whole body of the juvenile was acquired by using a dissecting scope affixed with a camera. Baseline images were taken to assess and document any anatomic abnormalities before experimental intervention, such as abrasions or missing distal sections of the arm due to injuries within the home tank. Animals with abnormalities located on R2 were excluded from phase II. Immediately postinjection, R2 was imaged to document the injection site.

Within phase I, follow-up images were taken on days 1, 2, and 3 to evaluate healing of injection site and document any potential adverse consequences, such as lesions beyond the injection site. Within phase II, to evaluate healing of the amputation, images were taken immediately postamputation and on days 1, 2, and 3.

Injections (phase I and II).

Due to the small size of the juveniles’ arms, which entailed small injection volumes, customized needles were created for attachment to a pneumonic picospritzer (PV830 Pneumatic PicoPump; World Precision Instruments, Sarasota, FL). Needles were composed of filamented borosilicate 1.0 × 0.75 × 10-cm capillary glass (Sutter Instruments) and pulled using a P-97 Pipette Puller (Sutter Instruments, Novato, CA). Needle ends were snapped using calipers set to 0.07 mm. Drug or saline was dosed based on group or individual weights, as described in the study design, and mixed with 0.5 μL (phase I) or 0.1 μL (phase II) of fast green dye (Sigma-Aldrich, Saint Louis, MO) to localize the deposit.⁹ The fast green dye was sterilized by passing a 0.1% solution of dye and 0.9% saline through a 0.22-μm filter (Milex Durapore PVDF; Millipore Corporation, Burlington, MA). The volume of dye was reduced between phases due to potential adverse consequences arising from excess volume injected into a small tissue space. Dose mixtures were aliquoted onto a sterile surface using a micropipette for vacuum suction into the customized needle.

The juveniles were maintained in a Petri dish filled with anesthetic solution. The second right arm (R2) was gently isolated using forceps (Dumont #3c; Roboz, Gaithersburg, MD). The customized needle filled with dyed drug was injected at a depth of approximately 0.4 mm and approximately 6 mm from the distal tip of the arm in an attempt to penetrate the intramuscular space. Juveniles were maintained for 15 minutes in an anesthetic solution to allow for the average, extrapolated onset of action to take effect.^22,26,27 For phase I, recovered animals were transferred back to their home tank.

Amputation (phase II).

In phase II, R2 was gently pulled outward following injection, and the arm was amputated at the midpoint of the injection site, as close to the puncture marks as possible, with iridectomy scissors (RS-5630; Roboz, Gaithersburg, MD). This location was selected to maximize the potential drug exposure at the amputation plan, as the extent of local anesthetic spread is currently unknown, and to maintain sufficient proximal arm length to allow continued use for postoperative prey capture and consumption. Care was taken to ensure that amputation occurred parallel to the vertical and horizontal planes, creating a clean transection, which facilitates healing and normal regeneration. Closure of any fashion was not performed, as cephalopods constrict blood vessels rapidly, preventing hemorrhage.^41,42

Weight (phase I and II).

At baseline, sample sizes were n = 11 for lidocaine and bupivacaine and n = 12 for saline. Sample sizes at later time points were reduced due to predetermined early removal under the approved IACUC protocol and one incidental mortality of undetermined cause. Sample sizes at day 7 were reduced to n = 10 for each treatment group. Differences in retention periods reflect a staggered enrollment design, in which animals enrolled earlier in the study were euthanized at day 7, whereas animals enrolled later were retained to day 21. This enabled additional longitudinal assessments, resulting in sample sizes for body weight measurements between day 7 and day 21 (n = 4 each for saline and lidocaine; n = 6 for bupivacaine).

Respiratory rate (phase II).

Juveniles were placed in 100-mm-diameter plastic Petri dishes filled with approximately 0.4 cm ASW. The water level was chosen to recapitulate natural walking ambulation and postures while mitigating swimming movements for appropriate evaluation of gait. Juveniles were acclimated to the Petri dish, undisturbed, for at least 15 minutes. After 15 minutes, juveniles within the Petri dish were placed into a light box (Photo Studio Light Box; DUCLUS) to facilitate filming. Juveniles were initially observed for a period of 2 minutes, during which their respiratory rate was determined. Respiratory rate was counted while juveniles were in a stationary position within the Petri dish for 15 seconds, and that count was multiplied by 4 to yield a value in breaths per minute.

Animals exposed to elevated room temperatures during data collection resulted in increased respiratory rates unrelated to treatment, and these data points were excluded from both the benchmark calculation and subsequent analysis. In addition, one animal within the bupivacaine group had a markedly low baseline respiratory rate at 12 breaths per minute, which was over 4 SD below the mean (40.67 ± 6.939); therefore, this animal was similarly excluded. Final group sizes were adjusted accordingly for all time points: saline (n = 10), lidocaine (n = 8), and bupivacaine (n = 7).

Walking assessment (phase II).

Animals were observed in the Petri dish for 2 minutes. Posture and gait were evaluated and scored while the juvenile was ambulating within the dish. For both parameters, juveniles were scored on a range of 0 (normal posture and normal elongation and shortening of arm) to 3 (tucked posture and little-to-no use of arm), as described in the results (Table 1). If the juvenile did not move during that 2-minute period, a plastic pipette was used to gently prod the juvenile into ambulating.

Table 1.

Ethogram Scoring

Score	Behavior description
R2 posture
0	Normal posture in comparison to nonsurgical arms
1	Postsurgical limb is observed to have mild curling
2	Postsurgical limb is observed to have moderate-to-severe curling
3	Postsurgical limb is observed to have tucking
R2 arm gait
0	Normal amount of elongation and shortening with crawling stride, sucker contact with Petri dish is normal, no abnormal positioning of arm under any circumstances
1	Mildly decreased elongation and shortening with crawling stride, sucker contact with Petri dish is mildly decreased, abnormal arm positioning variably observed during movement
2	Moderately decreased elongation and shortening with crawling stride, sucker contact is moderately decreased, abnormal arm positioning is frequently observed during movement
3	Little-to-no elongation and shortening with crawling stride, sucker contact is severely decreased or absent, abnormal arm positioning is consistently observed

Postural changes and arm gait of R2 in Octopus bimaculoides during a walking assessment categorized on a score of 0–3 to record behaviors that progressively increased in severity.

Initial sample sizes were saline (n = 12), lidocaine (n = 11), and bupivacaine (n = 11). Due to removal before experimental endpoints, the lidocaine group was reduced to n = 10 after the 0-hour time point. In the saline group, the sample size was reduced to n = 10 after 0 hour and to n = 9 after 48 hours.

von Frey assessment (phase II).

The juvenile was transferred within the Petri dish to the stage of a Stemi 2000 dissecting scope affixed with a Nikon D3200 camera for photography and filming. The von Frey assessment was adapted from previous literature^3,9 and employed a series of ascending manual filaments (North Coast Medical, Morgan Hill, CA) with forces ranging from 0.008 to 500 g. The mantle was probed until the filament bent 3 times before moving up in force and filament size. The testing was done on the mantle both to assess allodynia at the organismal level and to ensure, given the small size of a juvenile arm, accurate and consistent probing. The animal was only probed when in a stationary position within the Petri dish. To ensure the response was purely nociceptive rather than visual, the experimenter moved slowly and kept the probe out of the juvenile’s direct line of sight. The test ended once the juvenile reached threshold, which was demonstrated by retreating (walking or jetting opposite to the direction of the probe), grooming the probed area, or performing whole-body chromatophore changes.

One animal within the saline group was excluded, as it reached a threshold of 0.6 g at 0 hour, which is over 229 SD above the mean (0.01150 ± 0.002574). Each treatment group initially comprised n = 11 animals. Due to early removal, final sample sizes were reduced to n = 10 in the lidocaine group after time 0 hour and to n = 10 and subsequently n = 9 in the saline group after the 0- and 48-hour time points, respectively.

Regeneration analysis (phase II).

To assess regeneration, images were collected on days 7, 14, and 21. On day 7, blastema formation was scored using a three-point scale: 0 indicated minimal or no growth, 1 indicated blunted growth (<50% of normal), and 2 indicated normal growth (≥50% of normal).³⁴ On day 14, sucker formation was evaluated, with a score of 0 denoting the absence of suckers and 1 denoting the presence of suckers. On day 21, the presence of chromatophores was similarly scored, with 0 indicating the absence of chromatophores and 1 indicating the presence of chromatophores.

Euthanasia (phase I and II).

Euthanasia occurred at day 3 in phase I and day 7, 14, or 21 in phase II. The animal was transferred to a glass Petri dish filled with a solution 4% ethanol in fcASW. Juveniles were kept in the solution until their respiratory rate ceased for 15 minutes. The whole body of the juveniles were then transferred to 4% paraformaldehyde for 24–72 hour for fixation. The whole-body specimen was then transferred to phosphate buffered saline (PBS) and stored at 4 °C.

Histology (phase I).

Samples were selected to span the observed spectrum of tissue responses within each treatment group: saline, n = 2; lidocaine, n = 3; bupivacaine, n = 3; and extended-release bupivacaine, n = 3. Fixed arm tissue was embedded in paraffin on its side, and then serial, parasagittal 5-μm sections were collected. Histologic sections were prepared 10 μm apart, at 5 µm, 15 µm, 25 µm, and 35 µm, for each arm and stained with hematoxylin and eosin (H&E). Histologic sections were evaluated by a board-certified veterinary pathologist with expertise in the interpretation of tissues from cephalopods.

Statistical analysis.

The surgeon was blind to drug injections throughout data collection, with the exception of extended-release bupivacaine in phase I, due to its white opacity and viscosity. Interpretation of histopathologic samples was also performed blind. Metric counts and scoring were recorded during live observation and verified through subsequent review of photographic images and video recordings. Data collected and were organized into spreadsheets for record keeping (Google Sheets, Google; Excel, Microsoft) and analyzed using commercial software (GraphPad Prism version 10.4.2, for Windows; GraphPad Software, San Diego, CA).

Values of P < 0.05 were considered significant. Data for posture and gait, percent weight change, respiratory rate, and von Frey filament assessments are presented as mean ± SE. Posture and gait were each analyzed using a Kruskal-Wallis test with a Dunn multiple comparisons. Percent weight change, respiratory rate, and von Frey filament were each analyzed using 2-way ANOVA with a Tukey post hoc multiple comparisons. Finally, blastema scores are presented as the percentage of juveniles in each score category per treatment group and were analyzed using the Fisher exact test.

In phase II, one animal receiving lidocaine escaped from a covered Petri dish. Due to exposure to the external environment, which causes unrecoverable morbidity progressing toward mortality, the animal was euthanized. Two animals receiving saline were euthanized due to suspected self-injury, as evidenced by wound healing at different stages. Finally, one animal receiving bupivacaine was found dead within the tank; no antemortem clinical signs were observed. This represented the only spontaneous death in the study. Animals that were removed contributed data up to the point of removal, and subsequent time points for these subjects were treated as missing, with the exception of the respiratory rate analysis. The analyses were performed using appropriate models that accommodated unequal sample sizes across time points without excluding entire subjects from earlier measurements.

Results

Phase I.

We first assessed how naive animals respond to these drugs, as neither lidocaine nor any form of bupivacaine has been evaluated in this species or in juvenile octopuses. Due to the exploratory nature of this phase, the following results are presented descriptively with no formal statistical analyses performed.

Reactions to the injection were common across the various conditions, with 59% of animals across groups exhibiting tissue loss beyond the injection site (Figure 1A-C). In some cases, injury was not grossly visible (Figure 1A). More severe cases showed visible damage. In some animals, this involved localized loss of dermal tissue and, occasionally, underlying muscle (Figure 1B). In the most extreme cases, the entire limb distal to the injection site was lost (Figure 1C). This tissue loss was apparently due to autophagy, a self-injurious response previously documented in octopuses.^43,44 Affected animals displayed prolonged tucking of the injected arm toward their mouth, and tissue was missing once animals resumed a more typical posture.

Figure 1.

Tissue Damage in the Second Right Arm (R2) of Octopus bimaculoides Following Injection. (A) Two puncture sites from pulled glass capillary needle represented by pink arrowheads on the aboral, or nonsucker, side of R2. The typical dispersal of dyed injection solution is evident. (B) Flayed dermal and muscular components beyond injection sites represented by pink arrowheads. Some solution remains within deeper tissue, as indicated by green dye. Darkened chromatophores were noted around the lesion. (C) Complete loss of tissue distal to injection site with nonuniform edges of the dermis. Injection site is not visible, although some solution remains proximal to the injection site, as indicated by green dye. (D-F) Tissue stained with hematoxylin and eosin. Images included are representative examples and are not derived from the same gross specimens. (D) Focal, linear tract of hemocyte-like infiltration that extends from the striated muscle (M), with suspected disruption of myofibers, and into the ANC (N). (E) Epidermal (E) ulceration accompanied by an infiltrate of hemocyte-like infiltrates in the subepidermal connective tissue (C); chromatophores are denser than the surrounding tissue. (F) Hemocyte-like infiltration along the edge of the histologic section, with complete loss of tissue at wound edge.

Because the majority of animals with tissue loss beyond the injection site were from the bupivacaine and extended-release bupivacaine treatment groups, we tested half and quarter doses. A high incidence of suspected self-injury persisted using quarter doses, with 71% of bupivacaine animals and 40% of extended-release bupivacaine animals still presenting with tissue damage. In contrast, the incidence of damage was similar between a full dose of lidocaine and saline, occurring in 25% and 29%, respectively (Figure 2A). The most severe clinical outcome included the loss of the entire distal portion of the limb, which was not observed in any animal from the saline control group and in 13% of animals in the lidocaine treatment group (Figure 2B).

Figure 2.

Percentage of Octopus bimaculoides with Tissue Damage. (A) Percent of animals demonstrating lesions beyond the injection site when given original doses or quartered doses: saline control (n = 7), lidocaine (n = 8), bupivacaine (n = 6), reduced-dose bupivacaine (n = 8), extended-release (E-R) bupivacaine (n = 6), and reduced-dose extended-release bupivacaine (n = 10). (B) Percent of animals demonstrating complete loss of tissue distal to injection within saline control (n = 7), lidocaine (n = 8), bupivacaine (n = 18), and extended-release bupivacaine (n = 16). No saline animals exhibited this clinical sign. Formal statistics were not performed on data. Percents are rounded to the nearest whole number.

Tissue samples from 11 treatment animals were examined histologically (Figure 1D-F). Seven animals had various degrees of hemocyte-like infiltrates at the injection site, including the epidermis, dermis, and muscle, and in some cases, the major neural structure of the arm, the axial nerve cord (ANC). Evidence of these cellular responses was present across all drug groups and did not appear to be correlated with the extent of macroscopic tissue damage.

Potential mechanical confounders were also assessed. Subcutaneous bubbles, which were introduced by our pressure injection system, were observed at the injection site in 33% of animals (Figure S1A). In addition, 59% of animals had at least 2 or more injection attempts, creating additional puncture sites (Figure S1B).

Evaluation of injection sites across all treatment groups showed that 53% of animals had leaked more than half of the solution. Extended-release bupivacaine, irrespective of dosage, resulted in the highest leakage rate, with 63% of animals showing at least 50% leakage. The increased leakage, together with the incidence of lesions, led us to exclude extended-release bupivacaine from phase II.

Finally, postural changes were observed in all animals following injection, with most animals returning to baseline after 3 days. No changes to feeding habits were observed, with the majority of animals accepting food within one minute of offering.

Phase II.

We evaluated the efficacy of lidocaine and bupivacaine injections against saline control after surgical amputation of R2 using several assessments.

Posture and arm gait.

Arm posture and gait were assessed during walking assessments as possible indicators of pain and distress. Both posture and gait were scored on a progressive scale from 0 (normal) to 3 (severe) (Table 1; Figure 3A; Video S1). For arm posture, all groups presented significant changes in posture as tested with post hoc Dunn multiple comparisons, with more severe changes in posture observed at 0 and 6 hours compared with baseline (saline: P < 0.0001; lidocaine: P < 0.0001; and bupivacaine: P < 0.0001. Scores returned toward baseline by 24 hours (Figure 3B). For arm gait, a significant difference appeared at the 6-hour time point between drug groups (P = 0.0388); however, post hoc pairwise comparisons did not reach significance after correction for multiple testing (P > 0.05). There was a clear effect of postsurgical time on arm gait across the groups. The post hoc Dunn multiple comparisons revealed multiple significant differences across all groups, with the greatest effects at 0, 6, and 24 hours, relative to baseline (saline: P < 0.0001; lidocaine: P < 0.0001; and bupivacaine: P < 0.0001). These elevated scores were followed by a return toward baseline by 48 hours (Figure 3C).

Figure 3.

Posture and Arm Gait Captured during Walking Assessment of Octopus bimaculoides in Phase II. (A) Illustrations capturing posture during ambulation within the Petri dish. Scores (0–3) correspond with descriptions in Table 1. (B) Posture during walking assessment was evaluated in juveniles receiving saline, lidocaine, or bupivacaine before amputation from baseline (B) to 72 h. Data represented as mean ± SEM. Statistical analysis was performed using Kruskal-Wallis test with repeated measures, followed by Dunns multiple comparisons. Differences in behavior across the postamputation observation times were significant [saline H(2) = 50.29, P < 0.0001; lidocaine H(2) = 46.57, P < 0.0001; and bupivacaine H(2) = 54.47, P < 0.0001], but there were no significant differences between treatment groups at any time point (P > 0.05). (C) For arm gait, the 6-h time point reached significance between drug groups [H(2) = 6.499, P = 0.0388]; however, post hoc pairwise comparisons did not. Comparisons between lidocaine and bupivacaine trended toward significance (P = 0.0724). As with posture, the differences in behaviors seen across the postamputation time points were significant [saline: H(2) = 43.98, P < 0.0001; lidocaine: H(2) = 40.09, P < 0.0001; and bupivacaine: H(2) = 40.71, P < 0.0001]. The timeline on the x-axis in panels B and C is not a proportional progression.

Respiratory rate.

The mean respiratory rate of naïve animals at baseline was calculated as 38 breaths per minute. No significant differences in respiratory rate were observed between treat-ment groups at any time point (P > 0.05; Figure 4A). There was, in addition, a significant effect of time. Post hoc comparisons revealed statistically significant temporal differences within the bupivacaine group at 6 hours (P = 0.0548). No significant interaction between treatment group and time was observed (P > 0.05).

Figure 4.

Respiratory Rate and Percent Weight Change Were Evaluated in Octopus bimaculoides Receiving Saline, Lidocaine, or Bupivacaine Injections before Amputation from B (Baseline) to 21 d. Data represented as mean ± SEM. Statistical analysis was performed using 2-way ANOVA test with repeated measures, followed by Tukey multiple comparisons. (A) No significant differences were observed in respiratory rate between treatment groups at any time point (P > 0.05); however, there was a trend toward significance between the lidocaine and bupivacaine groups at 6 h (P = 0.0548). There was an overall significance of time (P = 0.0208); post hoc comparisons showed temporal differences within the bupivacaine group from 0 to 6 h (P = 0.0057) and 6 to 48 h (P = 0.0488) (not shown). The benchmark respiratory rate in naïve animals is shown as a dotted gray line. (B) No significant differences in percent weight change were observed between treatment groups at any time point (P > 0.05); however, the differences between both baseline and time 0 compared with days 7 and 21 were significant (P < 0.01–0.0001) (not shown). Timeline on the x-axis is not represented proportionally.

Percent weight change.

The average body weight for naïve 6- to 7-week-old hatchlings was 15.5 g. Percent change of each animal was calculated from baseline to assess normalized growth over time. All groups showed dramatic weight gain over the 3 post-amputation weeks, and there were no statistically significant differences across the groups (P > 0.05; Figure 4B).

von Frey filament test.

Mechanical allodynia was assessed using the von Frey filament test as a measure of nociceptive sensitivity to mechanical stimuli (Figure S2; Video S2). Under the tested conditions, no measurable effect on mechanical sensitivity by treatment or time postamputation was observed, and variability across animals was low (P > 0.05).

Tissue healing and regeneration.

The tissue damage observed in phase I provided significant concerns as to how these drugs may affect tissue repair and regeneration. Therefore, blastema growth was evaluated on R2 at day 7 (Figure 5A and B). In both control and treatment groups, many animals had not regenerated a normal-sized blastema. Some animals had no blastema, others had a smaller blastema, and some had a normal blastema; however, no significant differences were observed between treatment groups as determined by 3 separate Fisher exact tests (P > 0.05; Figure 5C).

Figure 5.

Blastema Growth Evaluated in Octopus bimaculoides Juveniles Receiving Saline, Lidocaine, or Bupivacaine before Injection. (A-C) Example images of each score. (A) R2 demonstrating a score of 0 due to minimal to no blastema growth. (B) R2 demonstrating a score of 1 due to regenerative tissue being less than or equal to 50% of the expected blastema growth. (C) R2 demonstrating a score of 2 due to regenerative tissue being greater than or equal to 50% of the expected blastema growth. (D) Statistical analysis was performed using Fischer exact test. There were no significant differences between treatment groups at this time point (P > 0.05).

The effects at day 7 proved to be a regeneration delay prompted by the solution injections. At days 14 and 21, all animals showed the expected morphologic characteristics of a regenerate arm, including small suckers (by day 14) and some chromatophores (by day 21).³⁴ All juveniles had suckers and chromatophores present at each corresponding time point (data not shown).

Discussion

Phase I, which was exploratory, was designed to inform methodology and to identify potential challenges and adverse consequences associated with administering lidocaine, bupivacaine, and extended-release bupivacaine relative to a saline control. Only lidocaine and bupivacaine advanced to phase II. Ultimately, no treatment group demonstrated consistent analgesic efficacy, and all were associated with some degree of complication. Based on these findings, we do not recommend the use of 12 mg/kg lidocaine or 3 mg/kg bupivacaine injections as analgesic agents in juvenile O. bimaculoides.

We observed tissue damage after injections. Additional consequences of local anesthetic treatment, notably by bupivacaine and extended-release bupivacaine, included an increase in suspected self-injurious behavior. Some animals had such severe lesions that the entire arm distal to the injection was missing. This degree of tissue loss is unexpected, as prior studies have successfully administered injected local anesthetics into cephalopods without reports of complication.^8–10,22 In our study, injection volume may have contributed to local tissue responses; however, the absence of severe tissue injury in saline-treated animals administered an intermediate volume suggests that volume alone does not explain the observed lesion incidence. The absence of severe tissue loss in the phase I saline group further supports that the pharmacological agents themselves were the primary drivers of phase I tissue damage.

These findings contrast with previous reports of lidocaine efficacy,^8,10,22 which differed from the present study in several key aspects, including the cephalopod species examined, developmental stage, and experimental context. Notably, prior studies^8,10,22 evaluated lidocaine delivered with a chemical irritant, whereas lidocaine here was administered as a perioperative treatment.

Our data from phase II indicates that the remaining drugs do not significantly improve behavioral outcomes. This is supported by the markedly different arm posture and gait scores observed immediately following amputation. This statistically significant increase was evident not only in saline-injected animals but was also anecdotally observed in previous work with noninjected, amputated animals (N.G. Schulz, unpublished data, 2023), indicating that the amputation itself reliably produces a behavioral response. Regardless of treatment, scores improved and approached baseline within 1–2 days following amputation, aligning with the typical timeline of wound reepithelialization.^34,42,45 Although this recovery was consistent across groups, bupivacaine consistently yielded lower posture and gait scores compared with saline and lidocaine. The lack of statistical significance precludes confirmation of an analgesic effect; however, a larger sample size may have revealed a significant difference.

The similar recovery trends across both local anesthetic and saline groups suggest that the drugs did not achieve a perceivable therapeutic effect. This contrasts with previous studies reporting successful outcomes using lidocaine; however, those studies used different cephalopod species, age groups, and experimental contexts involving coinjection of chemical irritants with lidocaine rather than as a perioperative treatment.^8,10 Despite the lack of drug-specific effects, the significant increase in walking assessment scores followed by gradual recovery reflects a general response to injury, substantiating that cephalopods warrant regulatory consideration to ensure their welfare is managed in a scientifically informed manner.⁴⁶

As part of an assessment to evaluate potential systemic effects of treatment in juvenile octopuses, percent weight change was measured postoperatively. Percent weight change was assessed as a potential indicator of pain, given that reduced appetite and weight loss are commonly observed in both mammalian⁴⁷ and fish species³⁷ as a chronic response to noxious stimuli. Drug- and saline-treated animals monitored from day 0 to day 21 exhibited consistent growth over time. This pattern suggests that the administered drugs did not interfere with the expected developmental weight gain of these juvenile animals after arm amputation.

Respiratory rate was also evaluated as a systemic metric, as hyperventilation is recognized as a respiratory stress response associated with sustained and uncontrolled pain in mammals⁴⁸ and certain fish species.³⁷ Respiratory rate did not differ significantly between treatment groups over time; however, a significant difference was observed within the bupivacaine group at the 6-hour time point, consistent with an overall, unanticipated decreasing trend. This may reflect local anesthetic penetration into the ANC, a major arm neural structure analogous to the vertebrate spinal cord,⁴⁹ or systemic absorption. In vertebrates, local anesthetics delivered intrathecally or intravenously can depress respiratory function, particularly at high doses, due to their neurotoxic effects on respiratory control centers.^50–52 Given the pharmacokinetics of bupivacaine, which has an onset of action of 20–30 minutes and duration of action is 6–8 hours in mammals,²⁶ compared with the shorter onset and duration of lidocaine,²² our measurements may have captured a peak of respiratory depression.

von Frey scores did not significantly differ across treatment groups nor time points in our study. When considered alongside our behavioral assessments, we propose that the von Frey test may not be a sufficiently sensitive measure of allodynia in O. bimaculoides juveniles, particularly when stimuli are applied at a site remote from the surgical site. This contrasts with its previously reported utility in adult Abdopus aculeatus,³ Loligo pealeii,⁷ and Euprymna berryi,⁹ during which the test detected drug-induced changes in nociceptive thresholds.

The final parameter of phase II focused on evaluating healing and regeneration. Using pharmacological agents, especially via routes that may introduce tissue disruption, may interfere with the regenerative process and compromise the integrity of a regenerative model. All groups, including the saline control, exhibited delays in blastema formation. This suggests a baseline level of mechanical irritation from the injection itself, irrespective of the pharmacological agent used. While the severity of lesions differed by group after injection in phase I, the consistent delay across groups in phase II implies that regeneration was primarily impacted by shared mechanical factors rather than drug-specific effects. In addition, self-mutilation in some animals may have hindered early stages of regeneration, contributing to the stunted growth observed at day 7. Despite this early variability, all animals demonstrated regenerative progress over time, and by day 21, each group had achieved comparable regeneration milestones.

Studies of local anesthetic delivery in snail (Lymnaea stagnalis) cell culture have demonstrated suppression, synapse formation reduction, and neuronal apoptosis and necrosis.^53,54 These findings demonstrate that local anesthetic drugs have the ability to affect neuronal repair processed in molluscs. Our findings extend these observations and suggest that research using cephalopod molluscs as regenerative models should exercise caution when considering the use of local anesthetics.

While our findings contribute to the growing foundation of cephalopod analgesia research, they also highlight key areas for further exploration. In particular, the route of administration is a critical consideration when developing standardized protocols. Our study employed a method that introduced mechanical complications that may have influenced efficacy measures. A critical next step is identifying alternative routes of administration, particularly in a small model, with future studies focusing on the feasibility and efficacy of methods such as immersion or parenteral oral delivery. Future directions also include expanding upon the range of analgesic agents. Building on previous work, which explored a range of general analgesics, continued comparative studies are needed to identify compounds with optimal safety and efficacy profiles.⁹ Taken together, these directions will support the development of species-appropriate, standardized analgesic protocols that improve both research outcomes and animal welfare.

Finally, our study’s application of the walking assessment represents a significant contribution to the field of cephalopod research, offering a novel and practical method for evaluating an animal’s ability to perform normal functions in the face of surgical intervention. Given the current lack of established methods to assess analgesic efficacy in cephalopods, we employed this test alongside other behavioral metrics adapted from vertebrate models and newly applied to cephalopods, as well as standard veterinary assessments, to provide a holistic evaluation of health and recovery. By integrating multiple lines of evidence, our approach offers a comprehensive framework for assessing the perioperative use of local anesthetics in cephalopods.

In conclusion, our study demonstrated that injectable local anesthetics are insufficient in mitigating pain in juvenile O. bimaculoides and are associated with the formation of lesions around the injection site. As such, the methods used in this study are not recommended for use in neither research nor clinical medicine involving cephalopods. Our findings underscore the need for continued research in cephalopod analgesia and emphasize the importance of developing safe and effective protocols to support both cephalopod welfare and scientific integrity of investigations.

Supplementary Materials

Figure S1. ^{(10.2MB, tif)} Examples of confounding mechanical injury in Octopus bimaculoides postinjection. (A) Two subcutaneous air bubbles along with normal dispersal of dye. The injection site is not visible. (B) Three puncture sites from pulled glass capillary needle (pink arrowheads) with normal dispersal of dye.

Figure S2. ^{(2.1MB, tif)} Mechanical allodynia via the von Frey filament test was evaluated in Octopus bimaculoides juveniles receiving saline, lidocaine, or bupivacaine before injection from baseline (B) to 72 hours. Data represented as mean ± SEM. Statistical analysis was performed using 2-way ANOVA test with repeated measures, followed by Tukey multiple comparisons. There were no significant differences between treatment groups at any time point or within each treatment group between time points, nor between the interaction of time and treatment (P > 0.05). The timeline on the x-axis is not represented proportionally.

Video S1. ^{(23MB, mp4)} Von Frey assessment.

Video S2. ^{(28.3MB, mp4)} Walking assessment.

Conflict of Interest

The authors have no conflicts of interest to declare.

Funding

This project was generously funded by the ACLAM Foundation.

Author Contributions

J.N.C. conceived and designed the study, acquired data, analyzed and interpreted the data, drafted the manuscript, and reviewed the final version. N.G.S. contributed to study design, data acquisition, data analysis and interpretation, and reviewed the final manuscript. C.W.R contributed to study design, provided experimental equipment and laboratory space, and reviewed the final manuscript. A.J.C., A.M.O., K.R.L., and G.P.L. contributed to study design and reviewed the final manuscript. D.M. supervised the study and reviewed the final manuscript.

Protocol registration

The Animal Care and Use Protocol for this study was created and is registered with the University of Chicago IACUC. There is no externally registered protocol.

Data availability

Data are available upon reasonable request to the corresponding author.