Postoperative pain following the ovariohysterectomy (OHE) of cats has been most commonly treated with opioid-based analgesia and nonsteroidal anti-inflammatory drugs (NSAIDs) [1,2]. However, concerns regarding opioid-induced hyperalgesia have motivated the investigation of non-opioid drugs to decrease postoperative pain in human and veterinary patients [3-5]. A variety of non-opioid analgesics such as lidocaine, ketamine, and dexmedetomidine have been utilized in multimodal analgesic approaches to improve the effects of perioperative analgesia and minimize small animals’ opioid consumption [4,6,7].

Lidocaine is an amide-based local anesthetic that may be used to provide peripheral or systemic analgesia [8,9]. In dogs, systemic lidocaine has been as effective as morphine to control postoperative pain after ophthalmic surgery [6]. However, systemic lidocaine infusion has not been recommended for cats due to a significant risk of dose-dependent cardiovascular depression and adverse neurological events [10].

The administration of drugs at acupuncture points (pharmacopuncture) can amplify the therapeutic effects of different medications, thereby potentially reducing dosages and decreasing adverse events [11,12]. Clinical application of pharmacopuncture has already been used in human and veterinary patients [11-14].

Previous clinical reports on postoperative pain management in dogs and cats undergoing OHE have shown that the injection of analgesic micro-doses at acupuncture points has been as effective as subcutaneous clinical doses [11,15]. In addition, the injection of lidocaine at acupuncture points in humans has decreased postoperative pain scores and opioid requirements compared to those in the control group [16]. However, there is no information available regarding the administration of local anesthetics at acupuncture points on small animals.

The aim of this study is to investigate how lidocaine administered at acupuncture points could affect analgesic requirements and postoperative pain in cats undergoing OHE. A secondary objective is to study the adverse and sedative effects resulting from this acupuncture technique. The hypothesis is that pharmacopuncture with lidocaine could reduce analgesic requirements and pain scores compared to those from the control group.

1. Design and ethical statement

A prospective, randomized, blinded, controlled clinical study was designed to investigate the perioperative analgesic effects of lidocaine administration at specific acupuncture points. The study protocol was approved by a local ethics committee (protocol 5930/2020 CEUA) at Universidade do Oeste Paulista and conducted in accordance with the reporting guidelines for clinical studies using acupuncture (Standards for Reporting Interventions in Clinical Trials of Acupuncture: STRICTA). Informed written consent for the investigation was obtained from all cat owners.

2. Study population

The researchers enrolled 30 crossbred, client-owned cats who were scheduled for elective OHE between February, 2020 and September, 2021. Inclusion criteria were as follows: cats had to be 6 months or older, exhibit complete blood count and serum chemistry values within accepted normal limits, and maintain good temperament while interacting with veterinarians. The exclusion criteria included pregnancy, lactation, a weight less than 2 kg, and/or systemic diseases. All cats were evaluated before the operation via abdominal ultrasound, to confirm that they were not pregnant. Animals were admitted to the hospital at least 36 hours prior to surgery to help them acclimate to the operating room and observers.

3. Anesthetic protocol

All anesthetic procedures were performed by the same anesthetist, who was unaware of group allocation. After fasting for 8 hours, cats were sedated intramuscularly with 10 µg/kg dexmedetomidine (Zoetis, São Paulo, Brazil). Ten minutes later, an aseptic 24-gauge catheter was placed in the cephalic vein. Lactated Ringer’s solution was then administered intravenously at 3 ml/kg/h, until extubation. Anesthesia was induced with IV propofol (Cristália, Itapira, Brazil) and maintained using isoflurane (Cristália, Itapira, Brazil), then vaporized in 100% oxygen and delivered via a non-rebreathing circuit (Takaoka, São Paulo, Brazil).

Heart rate (HR), oxyhemoglobin saturation (SpO₂), respiratory rate (RR), esophageal temperature (T), end-tidal carbon dioxide concentration (P_ETCO₂), and end-tidal isoflurane concentration (FE’ISO) were continuously measured using a multi-parametric monitor (Digicare Animal Health, Florida, USA). Before each experiment, the gas analyzer was calibrated with a standard gas mixture (CO₂: 5 vol %, N₂O: 70 vol %, O₂: 24 vol % and isoflurane: 1 vol %). The cats were allowed to breathe normally. Mechanical ventilation was instituted only in cats with hypercapnia (PETCO2 ≥ 45 mmHg). Systolic arterial pressure (SAP) was monitored indirectly using a Doppler sphygmomanometry (Doppler 841-A, Parks Medical Electronics), with an appropriately sized cuff that spanned 30-40% of the circumference of the thoracic limb. The probe was placed over the radial artery. Body temperature was maintained between 37℃ and 38℃ using a thermal warming blanket (Hefei Longshore, Anhui Province, China).

4. Study groups

The cats were assigned to an experimental or control group that was treated with pharmacopuncture (Lido group, n = 15) or nothing (Control group, n = 15), as determined by randomization software program (Research Randomizer, Computer software, http://www.randomizer.org/). The Lido group received a single, bilateral injection of lidocaine 5 mg/kg; Cristália, Itapira, Brazil), without a vasoconstrictor at acupoints Stomach-36 (ST-36) and Spleen-6 (SP-6) (Fig. 1). The 2% lidocaine was diluted in saline solution (0.9%), achieving a final volume of 1.0 ml.

Figure 1. Anatomical locations of the ST-36 and SP-6 acupuncture points.

Immediately after intubation, an identical volume of the anesthetic solution (0.25 ml) was administered at each acupuncture point using 1 ml syringes attached to 12.7 mm long, 28 gauge hypodermic needles, which were inserted, normal to the skin surface, to a depth of approximately 10 mm at ST-36 and 5 mm at SP-6, until a muscle twitch response was observed. The sequence of lidocaine administration was ST-36 (right and left), then SP-6 (right and left). All injections were performed by a veterinarian who has practiced acupuncture for 16 years. The locations of the acupuncture points were determined in accordance with Xie and Prost [17].

5. Intraoperative monitoring

Isoflurane vaporizer settings (Sigma Delta, Penlon Limited) were adjusted to maintain a constant surgical depth of the anesthesia to prevent eyeball rotation, loss of palpebral reflexes, and loss of jaw tone as well as autonomic responses to surgical stimulation. The FE’ISO, HR, and SAP were recorded at specific time points throughout the process of anesthetization, as follows: T0 (baseline, 10 minutes of 1.2% FE’ISO; before surgical stimulation); T1 (after skin incision); T2 and T3 (after clamping the first and second ovarian pedicles; T4 (after clamping the uterine cervix); and T5 (after the final skin suture was placed).

The ovariohysterectomy was performed using a standard technique that makes use of a median laparotomy access point. All surgical procedures were performed by the same surgeon, using a 3 cm ventral midline approach and 3-clamp technique. If the SAP or HR increased by more than 20% compared to those from the time point recorded immediately beforehand, then the FE’ISO was adjusted in 0.1-0.2% increments. Similarly, if the SAP or HR decreased by 20% or more from the previously recorded value, then the FE’ISO was decreased in increments of 0.1-0.2%. If a FE’ISO above 1.8% was required, then additional analgesia was provided using fentanyl (2 µg/kg, IV) and these rescue doses were recorded.

The total duration of anesthesia (administration of propofol to the discontinuation of isoflurane), surgery (first incision until placement of the last suture), and time to extubation (termination of isoflurane to orotracheal tube removal, after recovery of the palpebral reflex) was recorded for each cat.

6. Postoperative monitoring

At 24 hours prior to surgery (baseline) and 1, 2, 4, 6, 8, 12, 18, and 24 hours post-extubation, pain and sedation scores were assessed by the same graduate student, who had been trained to assess postoperative pain in cats using behavioral indices and was unaware of the treatment group assignments. Pain was assessed using the Glasgow feline composite measure pain scale (CMPS-Feline; 0, no pain; 20, maximum pain) [18] and UNESP-Botucatu multidimensional composite pain scale (MCPS; 0, no pain; 24, maximum pain) [19]. Morphine (0.2 mg/kg, IM; Cristália, Itapita, Brazil) was given as re-establish analgesia if the CMPS-Feline scores were ≥ 5/20 or MCPS scores were ≥ 6/24. Thirty minutes after morphine administration, an additional pain assessment was performed. If the CMPS-Feline or MCPS scores remained ≥ 5/20 or 6/24, respectively, then a single dose of meloxicam (0.2 mg/kg i.m.; Vetnil, São Paulo, Brazil) was administered. The researchers recorded the number of cats requiring rescue analgesia and the number of rescue doses administered.

The degree of sedation was assessed using a multidimensional composite scale (scale range = 0-16 points) that had been previously used for cats [20,21]. The scale included the following five criteria: spontaneous posture, response to noise (handclap) and noxious stimulus (pressure to a hind paw digit), resistance to being placed in lateral recumbency, and mandibular tonus. The degree of sedation was classified as poor (0-4), clinical (5-13), or profound (> 13) [21].

7. Adverse events

Seizures, nausea, vomiting, and cardiovascular effects (bradycardia, hypotension, and hypertension) were recorded. Bradycardia, hypotension, and hypertension were defined as the appearance of a HR < 80 beats/minute, SAP < 80 mmHg, and/or SAP > 160 mmHg, which persisted for more than 5 minutes.

8. Statistical analysis

A sample size of at least 15 cats per group was utilized to achieve 80% statistical power to detect a postoperative treatment failure (70% in the Control group and 20% in the Lido group). The sample calculation was based on pilot data.

A Shapiro-Wilk test was performed to assess the normality of the variables. Body weight, age, propofol dose, FE’ISO, and procedure durations were compared between groups using an unpaired t-test. A repeated measures ANOVA and Tukey’s test were used to compare the postoperative pain and sedation scores between groups, at various time points. The number of cats requiring perioperative rescue analgesia and the incidence of adverse effects were compared between groups, using Fisher’s exact test. All analyses were performed using GraphPad Prism 7.0. Differences were considered significant when p < 0.05.

Thirty-six cats were screened for enrolment in the study. Six did not meet the inclusion criteria. Two cats were pregnant and four exhibited aggressive behavior. Demographic data and procedural times were comparable between groups. Extubation time was longer in the Lido group (p = 0.01) (Table 1). The mean dose of propofol (mg/kg; Lido = 4.5 ± 1.1; Control = 4.6 ± 1.8; p = 0.11) and mean FE’ISO did not differ between groups (Lido = 1.41 ± 0.16; Control = 1.44 ± 0.11; p = 0.24).

Mean ± standard deviation..

^#Significant difference between groups (unpaired t-test)..

Procedural times: Anesthesia = time elapsed from the administration of propofol to discontinuation of isoflurane; Surgery = time elapsed from the first incision until placement of the last suture; Extubation = time elapsed from isoflurane termination until orotracheal tube removal, upon recovery of the palpebral reflex..

&md=tbl&idx=1' data-target="#file-modal"">Table 1

Demographic data and procedural times in the Lido and Control groups.

Variables	Lido	Control	p-value
Body weight (kg)	2.6 ± 0.6	2.9 ± 0.6	0.43
Age (months)	16.1 ± 8.1	16.7 ± 5.2	0.97
Anesthesia time (min)	45.7 ± 10	43.8 ± 10.2	0.88
Surgery time (min)	16.2 ± 3.9	16.7 ± 6.8	0.66
Extubation time (min)	19.2 ± 8.5#	10.5 ± 7.2	0.01

Mean ± standard deviation..

^#Significant difference between groups (unpaired t-test)..

Procedural times: Anesthesia = time elapsed from the administration of propofol to discontinuation of isoflurane; Surgery = time elapsed from the first incision until placement of the last suture; Extubation = time elapsed from isoflurane termination until orotracheal tube removal, upon recovery of the palpebral reflex..

Intraoperatively, analgesic supplementation was given to 13.3% (2/15) and 26.6% (4/15) of the cats in the Lido and Control groups, respectively. This difference was not statistically significant (p = 0.65) (Table 2).

T0 = baseline (10 minutes of 1.2% FE´ISO, before surgical stimulation); T1 = after skin incision; T2 and T3 = after the clamping the first and second ovarian pedicles, respectively; T4 = after clamping the uterine cervix; T5 = after the final skin suture had been placed..

&md=tbl&idx=2' data-target="#file-modal"">Table 2

Intraoperative analgesic supplementation in the Lido and Control groups.

Group	Intraoperative time points						Rescue doses	Rescued cats
	T0	T1	T2	T3	T4	T5
Lido	0	0	0	2	0	0	2	2/15
Control	0	0	1	3	1	0	5	4/15

T0 = baseline (10 minutes of 1.2% FE´ISO, before surgical stimulation); T1 = after skin incision; T2 and T3 = after the clamping the first and second ovarian pedicles, respectively; T4 = after clamping the uterine cervix; T5 = after the final skin suture had been placed..

Postoperative pain scores and analgesic requirements did not significantly differ between groups. Overall, the Lido group’s mean MCPS and CMPS-Feline scores increased until 2 hours after surgery. The G-Control in the MCPS and CMPS-Feline scores increased until 6 hours post-extubation (p < 0.0001) (Fig. 2A and 2B). Rescue analgesia was given to 67% (10/15) of the cats in each group. Nine cats in each group received a single dose of morphine and one cat in each group received two doses of morphine. A single dose of meloxicam was given to seven cats in the Lido group and four in the Control group (Table 3).

^†Only for persistent pain, a single dose of meloxicam was given 30 minutes after morphine administration..

&md=tbl&idx=3' data-target="#file-modal"">Table 3

Postoperative analgesic rescue medication in the Lido and Control groups.

Drug	Group	Time post-extubation (h)							Total
		0.5	1	2	4	6	8		Rescue doses	Rescued cats
Morphine	Lido	0	2	6	1	1	1		11	10/15
	Control	0	4	2	2	3	0		11	10/15
Meloxicam†	Lido	0	2	4	1	0	0		7
	Control	0	3	1	0	0	0		4

^†Only for persistent pain, a single dose of meloxicam was given 30 minutes after morphine administration..

Figure 2. Pain and sedation scores in the Lido and Control groups. Data are presented as mean ± SEM. *Significant difference from baseline values (Tukey test, p < 0.0001). Pain and sedation assessments were performed 24 hours prior to the surgery (BL) and 1, 2, 4, 6, 8, 18, and 24 hours post-extubation. MCPS = Multidimensional Composite Pain Scale; CMPS-Feline = Glasgow Feline Composite Measure Pain Scale.

Sedation scores did not differ between groups, at any time point. Compared with scores at baseline, sedation scores were significantly higher in both, up to 1 hour post-extubation in the Control group and two hours in Lido group (p < 0.0001) (Fig. 2C).

Vomiting was observed in 60% (9/15) and 40% (6/15) of the cats in the Lido and Control groups, respectively (p = 0.46). Intraoperatively, hypotension was recorded in only one cat from each group.

There were no additional analgesic benefits derived from the administration of lidocaine at the chosen acupuncture points in healthy cats undergoing ovariohysterectomy. These findings were not consistent with previously reported data from human studies, in which pharmacopuncture of lidocaine significantly increased postoperative analgesia compared with that of the control group [16]. Similarly, a clinical study on dogs reported that pharmacopuncture with morphine or carprofen was an effective method to control postoperative pain [11]. Given that the species and methods of pain assessment were not the same, the main inconsistencies between this study’s results and those of previous studies may be related to the different analgesic protocols used in each study.

In order to investigate the feasibility of developing an opioid-free analgesic protocol, opioids were not administered preoperatively in this study. This component of the study design may have skewed the results, since opioidergic pathways play an important role in the analgesic effect triggered by acupuncture [22,23]. Clinical studies on dogs and cats have demonstrated a synergistic interaction between acupuncture and opioids with respect to perioperative pain management [24-27]. Thus, it is possible that the inclusion of an opioid in our analgesic protocol could have increased the effectiveness of the pharmacopuncture techniques utilized.

In addition, the analgesic efficacy of pharmacopuncture can be influenced by acupoint selection. In our study, ST-36 and SP-6 acupoints were chosen based on results from previous clinical reports showing significant analgesic benefits derived from their stimulation in human and veterinary patients undergoing laparotomy [26,28]. Other acupoints such as Liver-3 (LIV-3), Gallbladder-34 (GB-34), and Large Intestine-4 (LI-4) have also been reported to positively impact postoperative abdominal pain [11,15,29]. Accordingly, it is likely that stimulating more acupoints could have intensified the analgesic response in cats used in the current study.

Moreover, to somatic sensory impulses must be triggered by acupoint stimulation, in order to be communicate analgesic blockades to the central nervous system (CNS) [17]. To elaborate on the neural mechanisms behind anesthetic signal amplification from acupuncture, previous researchers investigated the effects of skin penetration and stimulation of subcutaneous tissue surrounding specific acupoints using local anesthetics immediately before needling. The results indicated that the therapeutic effect of the anesthetic was blocked [30,31]. Furthermore, neuroimaging data suggest that multiple brain regions connecting to descending antinociceptive pathways are activated by acupoint stimulation, thereby reducing pain perception [32,33].

Conversely, one study reported that the injection of lidocaine at the ST-36 acupoint prior to needling reduced the number of activated brain areas in humans [34]. Therefore, it is possible that in the current study, pharmacopuncture using 5 mg/kg of lidocaine impaired the transmission of impulses from the acupoints to CNS, which could partially justify the lack of a significant treatment effect. Future studies evaluating the administration of different doses of lidocaine at these and other acupoints are required to clarify whether the dose of local anesthetic can influence the antinociceptive and analgesic effects of these anesthetics.

Although no significant antinociceptive benefits were found in the present study, more cats in the Control group (26.6%, 4/15) required rescue analgesia than those in the Lido group (13.3%, 2/15) during the traction and ligation of ovarian pedicles. These results suggest that stimulation of acupoints ST-36 and SP-6 with lidocaine and intramuscular dexmedetomidine may have improved the visceral antinociception, thereby decreasing sympathetic cardiovascular responses during the worst noxious stimulus presented during the OHE. In rats, the antinociceptive effects trigged by stimulation of ST-36 with bee venom helped activate opioidergic and α₂ adrenergic receptors [13]. Additionally, previous studies have shown improved antinociceptive effects following the combination of acupuncture with α₂ adrenergic agonist agents [35,36].

Given the high affinity of dexmedetomidine to the α₂ adrenergic receptors [37,38], it is possible that premedication with this agent could have amplified the antinociceptive effects induced by the pharmacopuncture in the present study. However, due to the short duration of action of dexmedetomidine [39], it is probable that such an analgesic effect did not extend to the early postoperative period. In goats, the combination of electro-acupuncture with dexmedetomidine (5 µg/kg) increased the pain threshold at 30 minutes compared to that of goats who had received each treatment alone [37]. In the present study, the surgical manipulation of the ovarian pedicles was performed at 29.43 ± 8 minutes after lidocaine acupoint injections, which was when the peak of the antinociceptive effect probably occurred. This phenomenon can explain the reduced fentanyl requirements in the Lido group.

During the postoperative period, both pain scores and the analgesic requirements were comparable between groups. In the present study, 73% of the total postoperative rescue analgesic doses were administered within the first two hours after extubation, which is consistent with previous data reported from feline OHEs [2,26,27,40]. It is likely that the high prevalence of rescue analgesia during this period decreased the pain scores, so the Lido group’s MCPS and CMPS-Feline scores returned to baseline values after 2 hours. However, even with analgesic supplementation, the Control group’s MCPS and CMPS-Feline pain scores returned to baseline values after 6 hours, post-extubation. Thus, it seems that the injection of lidocaine at ST-36 and SP-6 provided at least a modest analgesic benefit to the pharmacopuncture-treated cats.

Furthermore, the sedation scores did not differ between groups at any time point. However, it appears that pharmacopuncture with lidocaine potentiated and prolonged the sedative effect mediated by dexmedetomidine. As evidence, 67% of the pharmacopuncture-treated cats exhibited a clinically significant degree of sedation during the first hour, post-extubation; conversely, only 47% of cats in the Control group exhibited such an effect. In addition, the extubation time was significantly longer in the Lido group, which suggests that the administration of lidocaine at acupoints ST-36 and SP-6 may have potentiated the CNS depression provided by the anesthetic protocol.Previous studies have also found postoperative sedative effects following intravenous infusion of lidocaine in dogs [41,42].

It is important to emphasize that prolonged sedation in the pharmacopuncture-treated cats may have interfered in pain recognition. Although the MCPS and CMPS-Feline scores are clinically validated methods for pain assessment in cats [17,18], these scoring systems may be biased by sedation. During the first two hours post-extubation, analgesic supplementation was given to 70% of the rescued cats in the Lido group, of which 30% (3/10) exhibited a sedation degree ≥ 5. In these cats, the pain scores were increased due to some behavioral responses assessed according to both scoring systems such as activity (the cat was quiet and reluctant to move) and overall demeanor (depressive behavior), which could be related to the residual sedation and not to pain.

On the other hand, none of the 60% (6/10) of Control group cats who received rescue analgesia during the same period exhibited signs of sedation. Thus, it is possible that the prolonged sedative effect may have interfered with the pain recognition of cats in the Lido group. Moreover, postoperative analgesic supplementation may have biased the pain scoring systems.

To avoid the interference of rescue analgesia during these pain assessments, scores recorded after the first postoperative analgesic intervention should be removed from statistical analyses. However, this approach was not possible in the current study because 47% (7/15) and 40% (6/15) of the cats in the Lido and Control groups, respectively, required rescue analgesia during the first two hours, post-extubation. Therefore, the exclusion of these cats would have resulted in a sample size that was too small and reduced the statistical power of the study.

The lidocaine dose was based on literature from veterinary studies on cats. High doses of local anesthetics are known to be toxic [10]. The dose used in this study was determined to be safe, since minimal adverse events were recorded during the study period. No pharmacopuncture-treated cats exhibited neurological distress or cardiovascular depression that would indicate toxic effects from lidocaine. Transitory hypotension was recorded intraoperatively in only one cat from each group and was successfully reversed by decreasing the concentration of isoflurane. Vomiting was the most frequent adverse effect and was observed mainly after dexmedetomidine administration. This was expected, considering that dexmedetomidine-induced vomiting is a dose-dependent effect, which has been related to activation of the α2-adrenoceptors receptors within a chemoreceptor trigger zone [43].

A potential limitation of this study was the small sample size, which was estimated according to the prevalence of postoperative analgesic supplementation at 70% and 20% in the Lido and Control groups, respectively. Nevertheless, the need for rescue analgesia was the same in both groups, which limited the statistical power of the study. Additionally, despite the fact that training had been provided to the observer in order to score pain in cats using videos, it is possible that the observer’s lack of experience with pain research may have skewed the results.

The injection of lidocaine at ST-36 and SP-6 acupoints did not provide significant perioperative analgesic benefits in healthy cats undergoing OHE, when compared with results from the Control group. Given the limited information regarding administration of lidocaine at acupuncture points, further studies are need to investigate the optimal dosage and synergistic effects of this technique when combined with opioids or other analgesics in small animals.

The authors thank Dr. Rejane Batista Brinholi for performing abdominal ultrasound on the cats in this study. The authors also thank the owners of the cats and clinicians in the Department of Veterinary Surgery and Anesthesiology, Unoeste, for their support.

This study was financed, in part, by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brazil (CAPES) – Finance Code 001.

Camila Menossi Sueza Lima: recruiting and enrolling animals, behavioral scoring, rescue analgesia, and manuscript drafting; Camila Zanetti Segatto: perioperative care, data acquisition, and data management; Gabriel Montoro Nicácio: surgical procedure; Gustavo Ricci Zanelli: anesthesiologist, postoperative care; Renata Navarro Cassu: study design, acupuncture treatment, data analysis, statistical analysis, manuscript writing. All authors approved the final manuscript.

The authors declare no conflict of interest.