Behavioral Assessment of Tramadol and Meloxicam Effects on Postoperative Pain in a Rat Craniotomy Model

Abstract

Stereotaxic surgery is a common procedure in neuroscience, yet effective analgesic protocols require further study and refinement to optimize the analgesia used in invasive procedures and to improve animal welfare. This study evaluated the effects of tramadol and meloxicam, alone or combined, on pain management following craniotomy in rats. Forty Wistar-Han rats were divided into 5 groups: saline + anesthesia (SAL+ANE), saline + surgery (SAL+SUR), tramadol + surgery (TRA+SUR), meloxicam + surgery (MEL+SUR), and tramadol/meloxicam + surgery (TRA/MEL+SUR). Treatments (saline, 0.2 mL; tramadol, 17.8 mg/kg; meloxicam, 1.5 mg/kg) were administered subcutaneously every 12 h for 72 h. The animals underwent anesthesia or surgery 30 min after the first injection. Postoperative assessments included open field testing, a grooming transfer test, nesting behavior, body weight, and food/water intake. Surgery induced behavioral changes occured within 48 h. SAL+SUR and MEL+SUR groups showed increased locomotion and rearing, while SAL+SUR, TRA+SUR, and TRA/MEL+SUR groups had reduced grooming. TRA/MEL+SUR and SAL+SUR groups had the lowest grooming transfer test scores, and TRA/MEL+SUR rats displayed reduced nesting behavior. Craniotomy caused mild pain lasting at least 48 h. Although no optimal analgesic was identified, providing analgesia and refining surgical techniques are essential to ensure animal welfare.

Introduction

Stereotaxic surgery is one of the most commonly performed invasive experimental procedures in neuroscience and involves removing a section of the skull to access the brain.^1,2 It is used to implant cannulas or electrodes into specific brain regions to assess the effects of localized neurotransmitter manipulation, to inject drugs to evaluate their impact on the brain function, or to investigate signaling pathways in awake animals.¹

Stereotaxic surgery in human patients is known to cause minimal pain,³ and although the principle of analgesic use is universal, challenges and implementation of protocols differ between species. Furthermore, there is a lack of standardized pharmacological protocols or evidence-based recommendations for managing craniotomy pain in rodents.² Carbone and Austin⁴ found that 8% of studies (2 out of 25) involving craniotomy explicitly reported use of postoperative analgesia possibly as a result of underreporting. Recent data showed an increase in reported analgesic use, rising from 12.2% in 2009 to 25.2% in 2019, indicating greater awareness of pain management. Nonetheless, most studies from both reports did not indicate the administration of analgesics or local anesthetics, suggesting that postsurgical pain remains largely uncontrolled in most mice and rats undergoing craniotomy.⁵ Because inadequate pain management can lead to significant physiologic and behavioral changes, potentially affecting animal welfare and compromising the validity of scientific data, use of standardized analgesic protocols could offer benefity in terms of improving both animal welfare and experimental reproducibility.⁶

The goal of pain management is to eliminate pain using both analgesic drugs (alone or in combination) and nonpharmacological methods, such as refinement of invasive techniques and nursing care.⁷ To achieve this, analgesics should be administered preemptively (before surgery) to minimize sensitization and maintained throughout the postoperative period, particularly during the first 48 h when most signs of pain occur.⁸ Effective pain management can increase survival rates and improve the general condition of animals following surgery.¹

Meloxicam and tramadol were selected for this study as analgesic treatments due to their widespread use in veterinary medicine, as well as their complementary mechanisms of action in pain management. Meloxicam, a nonsteroidal anti-inflammatory drug (NSAID), preferentially inhibits cyclooxygenase-2 (COX-2), reducing prostaglandin synthesis and providing effective antiinflammatory and analgesic effects with fewer ulcerogenic properties compared with nonselective NSAIDs.⁹ Tramadol, a centrally acting μ-opioid receptor agonist, attenuates pain perception by modulating the opioid system and enhancing analgesia through serotonin and norepinephrine reuptake inhibition. Its efficacy is often enhanced when combined with NSAIDs, making it a suitable choice for multimodal pain management.

Pain evaluation presents challenges, as many studies attempt to indirectly assess pain using non-evoked behaviors, such as responses to mechanical, heat, or cold stimuli. However, these assays are impractical for assessing pain related to surgical procedures involving the head.² Various preclinical headache models instead rely on spontaneous behaviors (such as locomotor activity, rearing, nesting, and food or water intake) and nociceptive behaviors (such as grooming) to better gauge pain levels and evaluate analgesic efficacy.^10,11

For this reason, the present study aimed to evaluate the efficacy of tramadol and meloxicam, alone or in combination, to alleviate pain following craniotomy in rats through various behavioral tests.

Ethical review.

The project of the study was reviewed and approved by the Ethics Committee of the School of Veterinary Medicine and Animal Science, University of São Paulo, São Paulo, Brazil and followed their guidelines for humane care and usage of animals in science.

Materials and Methods

Animals.

Forty male Wistar-Han rats (Rattus norvegicus), aged 8 to 12 wk, were obtained from the Institute of Biomedical Science, University of São Paulo. The animals were group-housed (4 per cage) in polypropylene cages (41 × 34 × 16 cm) with pine bedding (Granja R.G., São Paulo, Brazil) and maintained on a 12-h light/12-h dark cycle (lights on at 06:00), at a temperature of 22 °C (±2 °C) and a relative humidity of 55% (±10%). They were provided with an irradiated diet (Nuvilab CR1, Quimtia, Colombo, PR, Brazil) and water ad libitum. The animals were confirmed to be free of endoparasites and ectoparasites, and complete health reports, including microbiologic status based on FELASA recommendations,¹² were provided by the vendor. Paper towels were added to the cages as environmental enrichment. Behavioral assessments were conducted between 0600 and 1200 after 1 wk of acclimatization to the new facility. All procedures adhered to the guidelines of the Ethics Committee of the School of Veterinary Medicine and Animal Science, University of São Paulo, São Paulo, Brazil (no. 3611111119).

Experimental groups.

Animals were allocated using a block randomization technique (randomizing participants within blocks to ensure that an equal number were assigned to each treatment) into one of the following groups (n = 8): saline + anesthesia (SAL+ANE), saline + surgery (SAL+SUR), tramadol + surgery (TRA+SUR), meloxicam + surgery (MEL+SUR), and tramadol/meloxicam + surgery (TRA/MEL+SUR). Drugs were prepared daily and dissolved in saline (1:5; 0.9%) administered at the following dosages: tramadol (50 mg/mL; Intradol, Alcames, San José, Costa Rica) at 17.8 mg/kg and meloxicam (5 mg/mL; Meloxic, Provet, Bogotá, Colombia) at 1.5 mg/kg. The TRA/MEL+SUR group received the corresponding dosages for each drug. Animals in the SAL+ANE and SAL+SUR groups received 0.5 mL of physiologic sterile saline (0.9%; Baxter, San José, Costa Rica). Dosages were determined based on previous studies.^11,13 In addition, the SAL+ANE, TRA+SUR, MEL+SUR, and TRA/MEL+SUR groups received a 0.2-mL injection of lidocaine at the incision site (20 mg/mL; Xylestesin, Cristália, São Paulo, Brazil).

Preoperative preparation and anesthesia.

At 30 min before anesthesia induction, the rats received their assigned treatments via subcutaneous injection. The person administering the injections was blinded to the treatments. Each rat was then placed in an induction chamber and anesthetized with 5% isoflurane (1 mg/mL; Isoflurine, Cristalia, São Paulo, Brazil) carried with 100% oxygen at a flow rate of 1 L/min. Once the righting reflex was lost, the rats were positioned in ventral recumbency on a preheated thermal blanket to maintain body temperature between 35 and 37 °C. Anesthesia was maintained with 1.5% to 2% isoflurane via a nose mask. Sterile ocular lubricant (2 mg/kg; Vidisic Gel, Bausch + Lomb, São Paulo, Brazil) was applied to both eyes.

Each rat’s head was positioned in the stereotaxic frame and hair shaving, aseptic cleaning, and surgical draping were performed over the surgical area. A 0.2-mL SC injection of lidocaine was administered to the incision area in SAL+ANE, TRA+SUR, MEL+SUR, and TRA/MEL+SUR groups. The SAL+SUR group did not receive any analgesic treatment. The SAL+ANE group was anesthetized for 30 min (the average surgery duration), while the remaining groups underwent craniotomy. Physiologic parameters and reflexes were monitored every 5 min until the righting reflex returned.

Surgical procedure.

Five minutes after lidocaine application, a midline incision ∼2.5 cm long was made along the scalp, extending from between the eyes to the back of the ears. The skin was retracted with 4 bulldog clamps to maintain exposure, and the subcutaneous tissue was removed with a sterile cotton swab to expose the skull. A 0.9-mm outer diameter hole was drilled with a sterile hand drill (5 mm caudal to the bregma and 3 mm lateral to the sagittal suture on each parietal bone). A 1- × 2-mm screw was inserted and secured with dental acrylic resin (25 g; Dencrilay, Dencril, São Paulo, Brazil). The incision was then closed with a simple suture using 0.3 nylon. Finally, clostebol and neomycin spray (30 g; Neowell, Wellcopharma, Guatemala City, Guatemala) was applied to the wound.

Postoperative analgesia and care.

Once the rats regained their righting reflex, they were placed alone in a cage with bedding material. An infrared lamp was positioned 1 m from the cage to provide supplemental warmth and prevent hypothermia during recovery. Treatments were administered subcutaneously every 12 h for 3 d. The MEL+SUR group received saline for the second injection each day, as meloxicam has a 24-h duration of action.¹⁴ The rats’ health status was monitored by a veterinarian for 72 h after surgery, as shown in the Figure 2 timeline.

Figure 2.

Experimental timeline. The graphical representation of the timeline of shows procedures made during the experiment. The first half of the experiment consists of the preoperative treatment application; 30 min later the surgery occurs, and 90 min after the surgery the OF test is made. The second half of the experiment consists of a 72-h cycle where the GTT, body weight, food and water intake, NB, and treatment administration are repeated. The graphical design was made using the BioRender.

Behavioral analysis.

At 1.5 h after surgery, once the animals had fully recovered from anesthesia, a series of behavioral tests was initiated. Each rat was placed at the center of an open field (OF) arena, and its behavior was observed for 10 min. The arena consisted of a circular, wooden chamber (90 cm diameter), covered with a black laminate sheet and illuminated with white light (∼106 lm over the OF arena). At 24 h after surgery, the OF test was repeated. Behaviors were video recorded for later analysis. The arena was thoroughly cleaned with 5% ethanol between tests to eliminate any scent cues left by previous rats that could bias results.

Locomotion was measured using EthoVision XT video tracking software (version 15.0.1416, Noldus Information Technology, Wageningen, the Netherlands) and recorded in meters. Rearing frequency and grooming duration were manually scored by trained observers, blinded to treatment conditions, using Solomon Coder software (version 17.03.22; https://solomoncoder.com/download.php). Grooming behavior included self-directed movements such as hand rubbing, face washing, unilateral and bilateral strokes over the head and ears, body licking, head and body scratching, and tail licking. The grooming sequences were categorized as previously described,^15,16 with 2 main variables: anatomic distribution and complexity.

Grooming was classified anatomically as cephalic (head- and forehand-directed sequences), caudal (directed to the body), and sequential (chained sequences of head and body grooming). Sequences that included hindpaw use were considered more complex and categorized as variations of the standard form, resulting in 6 different subtypes. Microgrooming episodes (<1 s) were also counted, while isolated scratching events were excluded.

Grooming transfer test (GTT).

Following the completion of the OF test, each rat was placed individually into a clean cage (41 × 34 × 16 cm) containing bedding material and relocated to an adjacent room. A single drop (∼50 µL) of UV fluorescent gel (Glo Germ Oil, Glo Germ, Moab, UT) was applied to the forehead of each animal. Fluorescence was visualized by turning off the lights in the room and exposing the rat to a UV lamp held at a distance of 15 cm.

To assess the presence of the fluorescent gel, a 5-point scale was employed, as described elsewhere.¹⁷ On this scale, a strong fluorescent signal was scored as 1, indicating minimal grooming activity, while a complete absence of fluorescence was scored as 5, signifying complete removal of the gel through self-grooming behavior.

As the gel is removed through self-grooming, the presence of fluorescence is inversely related to grooming frequency. To evaluate the gel’s long-term effects on grooming, fluorescence was checked at 4, 6, 8, 10, and 24 h after administration. Trained observers, blinded to treatment conditions, assigned scores. To assess drug effects during the postoperative period, the gel was applied each morning following treatment administration and fluorescence was examined daily for a total of 72 h.

Assessment of body weight, food/water intake, and nesting behavior (NB).

Each rat was weighed daily during the postoperative period. Food and water intake were monitored for up to 72 h after surgery. NB was evaluated on a 5-point scale, with a score of 5 indicating that the animal actively used the provided material (for nest construction or complete shredding) and a score of 0 indicating no use of the material (Figure 1). To assess NB, 2 sheets of towel paper (36.5 × 28 cm) were placed each day in the back left side of the cage. Trained observers, blinded to treatment conditions, assigned scores.

Figure 1.

Nesting behavior scoring. Score 1: Nesting material remains intact and in its original position. Score 2: Nesting material shows minimal handling, with slight evidence of biting (>90% remains intact). Score 3: Approximately 25% to 50% of the nesting material is shredded. Score 4: Between 50% and 90% of the nesting material is shredded but still largely remains in its original position. Score 5: Nesting material is completely shredded and dispersed throughout the cage.

Endpoints and euthanasia.

Animals exhibiting severe dyspnea, severe hypothermia (<35 °C) during anesthesia or signs of severe pain in the postoperative period were to be euthanized. However, no animals displayed these clinical signs. At the end of the experiment, all animals were euthanized in a CO₂ gas euthanasia induction chamber (Red Industria e Comercio de Equipamentos Hospitalares e Laboratoriais, Caieiras, Brazil). Death was confirmed by the absence of cardiac and respiratory function.

Statistical analysis.

Each group was compared with the SAL+ANE and SAL+SUR groups by analyzing cumulative scores through planned contrasts using one-way ANOVA. The OF test results at 24 and 48 h were compared using 2-way ANOVA with time (90 min and 24 h) and treatment groups (SAL+ANE, SAL+SUR, TRA+SUR, MEL+SUR, and TRA/MEL+SUR) as factors. Grooming subtypes were analyzed by 2-way ANOVA with subtype (cephalic, cephalic with variations, caudal, caudal with variations, sequential, and sequential grooming with variations) and treatment Groups as factors. When appropriate, independent analyses per group were conducted using one-way ANOVA. GTT and NB were analyzed by 2-way ANOVA, with hour (4, 6, 8, 10, and 24 h) and treatment groups as factors. Body weight, food/water intake, and NB were analyzed by 2-way ANOVA, with hour (24, 48, and 72 h) and treatment groups as within-group factors. A Dunnett multiple comparison test was applied for pairwise comparisons when appropriate. The study was powered based on previous behavioral studies with similar effect sizes, using an α of 0.05 and a power of 80%. All analyses were conducted using GraphPad Prism 8.2.1 software (GraphPad Software, La Jolla, CA). For the experiment timeline, see Figure 2.

Results

OF test.

At 24 h after surgery, the highest average distance traveled was observed in the SAL+SUR rats (F(4, 35) = 2.533, P = 0.0398) and MEL+SUR rats (F(4, 35) = 2.533, P = 0.0274) compared with the SAL+ANE group (Figure 3A). In addition, SAL+SUR rats (F(4, 35) = 2.533, P = 0.0398) and MEL+SUR rats (F(4, 35) = 2.533, P = 0.0275) exhibited higher velocity compared with SAL+ANE at 24 h (Figure 3B). At 90 min after surgery, MEL+SUR rats (F(4, 35) = 2.533, P = 0.0488) showed less movement than did SAL+ANE rats, while at 24 h, SAL+SUR (F(4, 35) = 2.428, P = 0.038) and MEL+SUR rats (F(4, 35) = 2.428, P = 0.0392) exhibited increased movement compared with SAL+ANE rats (Figure 3C). Furthermore, the MEL+SUR group demonstrated significant increases in distance (F(4, 70) = 2.291, P = 0.0212), velocity (F(4, 70) = 2.292, P = 0.0212), and movement (F(4, 70) = 3.963, P = 0.0005) between 90 min and 24 h postsurgery (Figure 3A–C).

Figure 3.

Locomotor and rearing behaviors of rats measured 90 min and 24 h after anesthesia or surgery. (A) Distance traveled (cm), (B) average velocity (cm/s), and (C) movement duration (s) across groups in the OF are shown, as are (D) frequency and (E) duration of rearing behavior (s). *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Among rats that underwent surgery, no differences were observed among the groups in rearing frequency at 90 min postsurgery. However, all groups that underwent surgery showed a lower frequency compared with the SAL+ANE group (Figure 3D). At 90 min, the SAL+SUR (F(4, 35) = 3.446, P = 0.0465), TRA+SUR (F(4, 35) = 3.446, P = 0.0085), and TRA/MEL+SUR (F(4, 35) = 3.446, P = 0.0148) groups exhibited significantly shorter rearing duration compared with the SAL+ANE group (Figure 3E). Between 90 min and 24 h, the SAL+SUR group (frequency: F(4, 70) = 2.778, P = 0.0047; duration: F(4, 70) = 2.010, P = 0.0018) and the MEL+SUR group (frequency: F(4, 70) = 2.778, P = 0.0005; duration: F(4, 70) = 2.010, P = 0.0048) showed increased frequency and duration of rearing (Figure 3D, E).

There were no differences in the frequency of grooming between 90 min and 24 h (Figure 4A). The MEL+SUR group showed the longest grooming duration at 90 min, significantly different from both the SAL+ANE (F(4, 35) = 7.032, P = 0.0027) and SAL+SUR groups (F(4, 35) = 7.032, P = 0.0011). Conversely, the SAL+ANE group (F(4, 35) = 3.337, P = 0.0399) had a longer grooming duration at 24 h compared with the SAL+SUR (F(4, 70) = 5.138, P = 0.0399), TRA+SUR (F(4, 70) = 5.138, P = 0.0193), and TRA/MEL+SUR groups (F(4, 70) = 5.138, P = 0.0099) (Figure 4B). In addition, the MEL+SUR group (F(4, 70) = 5.138, P = 0.0028) showed a significant reduction in grooming duration between 90 min and 24 h postsurgery (Figure 4B).

Figure 4.

Grooming behavior and subtypes in rats at 90 min and 24 h after anesthesia or surgery. (A) Frequency and (B) duration (s) of total grooming behaviors are shown. Subtypes of grooming are shown in (C) frequency and (D) duration at 90 min, as well as (E) frequency and (F) duration at 24 h. Grooming subtypes include cephalic (Ceph.), caudal (Caud.), sequential (Seq.), and variations (var.) in grooming or scratching (Scrat.). *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

At 90 min postsurgery, the MEL+SUR group developed the sequential grooming subtype with the highest frequency and the longest duration compared with the other groups. This was significantly different in frequency (F(24, 245) = 1.838, P = 0.0001) compared with the SAL+SUR group and in duration (F(24, 245) = 4.153, P < 0.0001) compared with both SAL+ANE and SAL+SUR groups (Figure 4C, D). In contrast, the TRA+SUR group showed the lowest frequency and the shortest duration of sequential grooming among all groups, with significant differences in both frequency (F(24, 245) = 1.838, P = 0.0398) and duration (F(24, 245) = 1.838, P = 0.0398) compared with the SAL+ANE group at 90 min (Figure 4C, D). The TRA/MEL+SUR group exhibited a significantly shorter duration of sequential grooming (F(24, 245) = 1.838, P = 0.0459) compared with the SAL+ANE group (Figure 4D), and this was the only group that demonstrated scratching grooming, showing a significant difference in frequency (F(24, 245) = 1.838, P = 0.0398) compared with the SAL+SUR group at 90 min (Figure 4C).

At 24 h, the SAL+ANE group exhibited the highest frequency (F(24, 245) = 1.590, P = 0.0042) and the longest duration (F(24, 245) = 4.074, P < 0.0001) of sequential grooming compared with the SAL+SUR group (Figure 4E, F). All groups that underwent surgery showed significantly lower frequency (TRA+SUR: F(24, 245) = 1.590, P = 0.004; MEL+SUR: F(24, 245) = 1.590, P < 0.0001; TRA/MEL+SUR: F(24, 245) = 1.590, P = 0.0459) and shorter duration (SAL+SUR: F(24, 245) = 4.074, P < 0.0001; TRA+SUR: F(24, 245) = 4.074, P < 0.0001; MEL+SUR: F(24, 245) = 4.074, P < 0.0001; TRA/MEL+SUR: F(24, 245) = 4.074, P < 0.0001) of sequential grooming compared with the SAL+ANE group (Figure 4E, F).

GTT.

During the first 24 h postsurgery, the SAL+ANE group exhibited the highest grooming behavior scores, with a significant difference (F(16, 175) = 0.7647, P = 0.0316) compared with the SAL+SUR group at 10 h. The TRA/MEL+SUR group showed the lowest grooming scores at 6 h (F(16, 175) = 0.7647, P = 0.0088), 8 h (F(16, 175) = 0.7647, P = 0.0004), and 10 h (F(16, 175) = 0.7647, P < 0.0001) compared with the SAL+ANE group. The MEL+SUR group also presented significantly lower grooming scores at 8 h (F(16, 175) = 0.7647, P = 0.0088) and 10 h (F(16, 175) = 0.7647, P = 0.0316) compared with the SAL+ANE group (Figure 5A).

Figure 5.

Grooming transfer test scores during the first 72 h after surgery. (A) 24 h, (B) 48 h, and (C) 72 h after surgery. Each graph shows pain scores (y-axis) measured at various time points (x-axis) in hours. Treatment groups include SAL+ANE, SAL+SUR, TRA+SUR, MEL+SUR, and TRA/MEL+SUR. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001. Error bars represent the SEM.

At 48 h postsurgery, SAL+ANE continued to show the highest grooming behavior scores. The MEL+SUR group had the lowest grooming score at the first evaluation at 4 h (F(16, 175) = 0.3656, P = 0.0254). At 6 h, SAL+SUR (F(16, 175) = 0.3656, P = 0.0254), TRA+SUR (F(16, 175) = 0.3656, P = 0.0254), and TRA/MEL+SUR (F(16, 175) = 0.3656, P = 0.0254) showed lower grooming scores compared with SAL+ANE. These groups maintained the lowest grooming scores at the later hours of the second day (Figure 5B).

On the third day of evaluation (72 h), there were no significant differences in grooming behavior among the groups (Figure 5C).

Body weight, food/water intake, and NB.

No differences in body weight were observed among the groups during the postoperative period. However, the MEL+SUR group showed an increase in body weight (F(2, 105) = 3.160, P = 0.0465) between 24 and 48 h (Figure 6A). There were no significant differences in water intake among the groups during the postoperative period (Figure 6B).

Figure 6.

Body weight gain and water and food/intake evaluation during the first 72 h after surgery. (A) Weight gain (y-axis in grams) measured at 24, 48, and 72 h (x-axis). (B) Water intake (y-axis in mL) measured at the same intervals. (C) Food intake (y-axis in grams) measured at 24, 48, and 72 h. Treatment groups include SAL+ANE, SAL+SUR, TRA+SUR, MEL+SUR, and TRA/MEL+SUR. *, P < 0.05; **, P < 0.01. Error bars represent the SEM.

Regarding food intake, statistical analysis (ANOVA) indicated that the TRA/MEL+SUR group had significantly lower consumption at both 24 and 48 h postsurgery compared with the SAL+ANE group (24 h: F(8, 105) = 0.3288, P = 0.0391; 48 h: F(8, 105) = 0.3288, P = 0.0035). In addition, the TRA/MEL+SUR group consumed less food than did the SAL+SUR group throughout the 3 d following surgery (24 h: F(8, 105) = 0.3288, P = 0.0335; 48 h: F(8, 105) = 0.3288, P = 0.0107; 72 h: F(8, 105) = 0.3288, P = 0.0148). Similarly, the TRA+SUR group exhibited a significant reduction in food consumption at 48 h compared with the SAL+ANE group (F(8, 105) = 0.3288, P = 0.05) and at 72 h compared with the SAL+SUR group (F(8, 105) = 0.3288, P = 0.0148) (Figure 6C).

Regarding NB, statistical analysis (ANOVA) indicated that the SAL+ANE group had the highest NB scores, showing a significant difference compared with the SAL+SUR group at 8 h on the first day of evaluation (F(16, 175) = 0.4233, P = 0.0194). On that same day, all groups treated with analgesics exhibited lower NB scores, with significant differences observed at multiple time points compared with the SAL+ANE group: at 4 h: TRA+SUR (P = 0.0194), MEL+SUR (P = 0.0021), TRA/MEL+SUR (P = 0.0067); at 6 h: TRA+SUR (P = 0.0067), MEL+SUR (P = 0.0006), TRA/MEL+SUR (P = 0.0021); at 8 h: TRA+SUR (P = 0.0002), MEL+SUR (P = 0.0002), TRA/MEL+SUR (P < 0.0001); at 10 h: TRA+SUR (P = 0.0021), MEL+SUR (P = 0.0006), TRA/MEL+SUR (P < 0.0001); and at 24 h: TRA/MEL+SUR (P = 0.0006). Among the analgesic-treated groups, the TRA/MEL+SUR group consistently exhibited the lowest NB scores throughout the first postoperative day, maintaining a significant difference compared with the SAL+ANE group at the final evaluation of the day (Figure 7A).

Figure 7.

Nesting behavior scores during the first 72 h after surgery. Shown are the nesting behavior scores (y-axis) for rats across different treatment groups over a 72-h postoperative period. Each graph represents a distinct time point: (A) 24 h, (B) 48 h, and (C) 72 h after surgery. The x-axis indicates time intervals after surgery (4, 6, 8, 10, and 24 h). Treatment groups are SAL+ANE, SAL+SUR, TRA+SUR, MEL+SUR, and TRA/MEL+SUR. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****P < 0.0001. Error bars represent the SEM.

After the second day (48 h) of evaluation, all groups presented similar NB scores (Figure 7B, C).

Discussion

Pain is a significant factor with respect to the welfare of research animals undergoing surgery. Inadequate pain management can distort experimental results, especially in studies that analyze animal behavior. Detecting and preventing pain related to craniotomy is particularly critical in neuroscience research.²

The SAL+ANE group was only anesthetized to serve as a control group without any pain-inducing interventions, allowing comparisons with other groups. This group exhibited similar exploratory behavior at both 90 min and 24 h after surgery in the OF test, but grooming behavior significantly increased at 24 h. Locomotion enables rats to create a spatial and sensorimotor representation of the environment,¹⁸ while rearing helps them assess the surroundings for potential threats.^18,19 Typically, when rats are repeatedly exposed to the same environment in the OF, they habituate because the environment no longer presents a novel experience and they recognize it as nonthreatening.²⁰ As a result, grooming behavior increased both in frequency and duration over time. This finding supports the conclusion that anesthesia, such as isoflurane used in these experiments, did not significantly disrupt the animals’ behavior during the first 24 h following surgery.

The SAL+SUR and MEL+SUR groups exhibited the most significant changes in exploratory behavior in the OF test at 24 h, with notable increases in locomotion and rearing. We hypothesize that the increased exploration observed in the SAL+SUR group could be a response to pain. Several studies involving abdominal surgeries have reported a reduction in locomotion and rearing as indicators of postsurgical pain.^21–23 However, in our study, the surgery was performed on the rats’ skulls, and the procedure may not have had the same impact on mobility.

Measuring pain in animals presents challenges, often relying on reflexive responses in external limbs, but it becomes particularly difficult when assessing conditions such as headache resulting from craniotomy. Depending on the specific pain measures used, these assessments can capture either pain-stimulated or pain-depressed behaviors.²⁴ The OF test has been studied for only a limited number of painful conditions in rodents, and variations in the type of pain or the surgical procedure can lead to different behavioral reactions. This variability may limit the utility of the OF test for clinical pain assessment.²⁵ Previous studies have shown that tramadol and meloxicam can independently influence locomotor activity, rearing, and grooming behavior in naive mice, even in the absence of surgical intervention.^26,27 Alemán-Laporte and colleagues¹⁸ specifically reported that tramadol reduced locomotion and grooming behavior, while meloxicam increased locomotion in rats. These findings suggest that the variability observed in our OF test data may be influenced not only by postsurgical pain but also by the direct effects of analgesic treatments. To improve the utility of the OF test in future studies, we propose incorporating additional control groups that receive analgesia without undergoing surgery. This would help differentiate the behavioral effects of the drugs from those associated with postsurgical pain.

On the other hand, the increase in locomotion observed in the MEL+SUR group is complex to interpret. When comparing this reaction with that of the SAL+SUR group, one might initially conclude that the analgesic treatment was not as effective in alleviating pain. However, in a previous study, we demonstrated that MEL at 1.5 mg/kg injected intraperitoneally can increase locomotion in naive rats in the OF.¹⁸

In the OF, a reduction in grooming was observed across all groups that underwent surgery. However, the MEL+SUR group showed an increase in both the frequency and duration of grooming at 90 min postsurgery. While some studies have linked active movements and reduced grooming to analgesic benefits,^22,28 the elevated grooming frequency and duration levels observed in the MEL+SUR group at 90 min make it difficult to attribute MEL’s effect to pain reduction.

Rats in the SAL+SUR, TRA+SUR, and TRA/MEL+SUR groups also displayed reduced grooming duration, particularly in the sequential grooming subtype. The GTT revealed that this reduction in grooming behavior persisted for up to 48 h in some rats after surgery. The most affected groups were TRA/MEL+SUR and SAL+SUR, which exhibited the lowest GTT scores during this period. Grooming is a complex, patterned behavior that typically follows a cephalocaudal sequence. The sequential grooming subtype begins with paw licking, followed by cleaning of the nose, face, and head, and eventually progresses to the cleaning of the body.²⁹

Considering that craniotomy induces pain, rats may reduce contact with the injured area, which could lead to a decrease in grooming behavior. Consequently, pain might have affected the SAL+SUR group due to the absence of analgesics during the postoperative period. For the TRA+SUR and TRA/MEL+SUR groups, the observed reduction in grooming could be attributed to the effects of TRA. In previously published experiments, we demonstrated that rats treated with TRA exhibited reduced grooming in the OF.¹⁸ This effect may be linked to various factors related to TRA’s mechanism of action, including increased serotonin levels, which can induce restlessness and distract the rat from grooming its entire body.³⁰ In addition, the blockade of serotonin 5-HT2A receptors, which are involved in TRA’s analgesic effect, may also contribute to a reduction in grooming.³¹

A study by Samal and colleagues³² evaluated the impact of tramadol administration through different routes (intraperitoneal compared with oral) in a rat craniotomy model, assessing pain levels through nest building and the grimace scale. Their findings indicate that both administration methods were equally effective in pain management. However, our study differs in several key aspects: we used a combination of behavioral tests (including the OF test and GTT), a different rat strain (Wistar-Han compared with Copenhagen), and a multimodal analgesic approach with meloxicam. Notably, Samal and colleagues observed a reduction in NB in the initial postoperative period, a finding that aligns with our results showing decreased nesting scores in groups receiving analgesia. This suggests that tramadol may influence NB beyond its analgesic properties. In addition, while Samal and colleagues found that tramadol effectively reduced grimace scores over time, we observed that tramadol-treated rats exhibited reduced grooming duration, suggesting a sedative effect. Other authors have observed that tramadol administered at doses >45 mg/kg induced dose-dependent sedation in rats.³³ These differences highlight the importance of considering strain-specific responses and multimodal approaches when evaluating postoperative pain management in craniotomy models.

An unexpected side effect observed in the TRA and TRA/MEL groups was an injection site reaction. Most animals injected with these treatments developed ulcerative skin lesions, even though the drugs were diluted in a 1:5 ratio with saline. Cannon and colleagues³⁴ also reported the appearance of ulcerative lesions in rats injected with TRA, although these lesions occurred in animals that received higher doses (25 to 50 mg/kg). An analysis of the tramadol’s pH used was conducted, in which it was identified that, despite diluting the drug, its pH was 6.6 (pH meter: Orion 3 STAR, Thermo Scientific, Waltham, MA). Because rats require a pH of ∼7.3 to 7.4,³⁵ these alterations may be linked to the difference in pH. Although this formulation is designed for veterinary use and can be administered subcutaneously, it was intended for other species, which may explain the unexpected reactions observed in our rat model.

Only the TRA/MEL+SUR and TRA+SUR groups exhibited a reduction in food intake during the postoperative period. However, this reduction did not lead to body weight loss, as no significant differences were observed in this parameter. This finding is likely due to appetite suppression resulting from the increased opioid activity, which could have been caused by heightened sedation or a nauseating effect.³⁶

The reduction in the NB scores was more pronounced in animals treated with analgesics, particularly in the TRA/MEL group. Changes in general behavior can be indicative of compromised well-being.³⁷ Because the reduction of NB scores could be related to pain, the data might suggest that the analgesics used in this study did not provide a satisfactory level of analgesia. However, the SAL+SUR group showed higher NB scores than the other groups that received analgesic treatment during the first 24 h, making it difficult to attribute this reaction to pain. Previous studies report that tramadol can induce sedation in rats,^33,34,38 which could explain the low NB scores in the groups treated with this drug. However, there are no reports regarding the effects of meloxicam on NB. Therefore, further investigation is needed to clarify these results.

The data for this study were obtained using only male rats, in line with prior research practices. While this approach aligns with earlier studies, we acknowledge that excluding females may restrict the broader applicability of our findings. A recent review highlighted this trend, reporting that 71% of craniotomy studies in 2009 and 68% in 2019 exclusively used male animals.⁵ However, the NIH has emphasized the importance of including both sexes in preclinical research to improve translational relevance. To address this limitation, future studies should include both male and female subjects to enhance the representativeness and generalizability of the results.

All surgeries were performed with the goal of minimizing animal suffering by refining the surgical techniques as much as possible. To achieve this, analgesics were administered 30 min before surgery, local lidocaine was applied, dental resin was used that did not exceed 40 °C, and sterile equipment was maintained throughout the procedure. These measures likely reduced postoperative pain (as evidenced by the lack of differences between the SAL+SUR and SAL+ANE groups in many of the results) and contributed to a swift recovery for the animals.

Conclusions

Taken together, the data suggest that craniotomy induced mild pain in the animals, as the results from most of the tests did not significantly differ from the group without analgesia. Therefore, analgesic treatment is essential, at least during the first 48 h postsurgery, a period during which the most significant behavioral changes were observed in the animals that underwent surgery. Importantly, note that analgesics can produce varying effects on animal behavior, so these factors should be carefully considered in the experimental design and planning. Alternatively, behavioral tests could be conducted 72 h after surgery, once the analgesic treatment has concluded, to avoid confounding effects. These tests may vary depending on the surgical procedure and the intensity of pain. However, it is crucial to conduct a pilot study beforehand to determine which test would be the most suitable choice.

Conflict of Interest

The authors have no conflicts of interest to declare.

Funding

This article is based on a study supported by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES; finance code 001).

Data Availability

The datasets are available from the corresponding author on reasonable request.