Perioperative drug management in non-critical companion animals: a retrospective study at a Spanish veterinary teaching hospital (2018–2022)

Abstract

There is little information in veterinary literature on the perioperative pharmacological management of small animal patients, despite the existence of common protocols and the importance of properly managing this period to reduce anaesthesia-related detrimental effects. This study aimed to analyse the current use of perioperative drugs in companion animals treated at the Veterinary Teaching Hospital of the University of León (HVULE) in Spain over a period of 5 years (2018–2022), describe the prescription patterns of these medicines, and identify the main variables associated with their prescription to explore possible strategies to promote their appropriateness. A total of 3438 cases were included in this study. The animals that most frequently underwent surgery were dogs (58.2%), females (57.0%), and adults (73.0%). The primary procedures performed were reproductive (56.6%) and traumatological (19.8%) surgeries. Regarding pharmacological treatments, more than half (62.3%) belonged to the ATCvet classification QN group (nervous system), and the most common compounds were isoflurane (13.5%), methadone (13.5%), and propofol (12.7%). Amoxicillin/clavulanic acid (3.7%) and marbofloxacin (2.8%) (categories C and B in the European Medicines Agency categorisation, respectively) were the most prescribed antibiotics. These findings provide detailed data to help veterinary policymakers improve drug use during surgical procedures.

1. Introduction

Anaesthesia is a frequent practice in many situations in veterinary medicine, such as immobilisation, analgesia, diagnostics, or surgery, and contributes to improving animal welfare and promoting patient safety (Ryan and Nielsen 2010). Drugs such as inhalation agents, local anaesthetics, sedative-hypnotics, tranquillizers, alpha-2 adrenergic agonists, neuromuscular blockers, opioids, and dissociatives are used to achieve this goal (Raffe 2020). Despite their wide use, there are currently few medicines specifically developed for veterinary use, which means that human medicines must be used. Additionally, some of these medications are not labelled for all target species. Under these circumstances, veterinarians can use these drugs in an extralabel manner. Many species require sedation with sedative or neuroleptic drugs before anaesthetic induction, whereas others require only light or moderate sedation. Pre-anaesthetic sedation is usually administered intramuscularly because of the increased need for immobilisation when drugs are administered intravenously. After adequate sedation, patients are more cooperative and the intravenous route is less stressful for both animals and veterinary staff (Schroeder and Smith 2011).

For all surgical interventions, the administration of analgesics (opioids, NSAID, and local anaesthetics) should be planned. Opioids are highly effective against perioperative pain and their benefits have been documented (Berry 2015; Gruen et al. 2022). Local anaesthetic drugs used in local and regional blockades can completely block the transmission of painful or nociceptive signals, providing profound analgesia (Grubb and Lobprise 2020a; Grubb and Lobprise 2020b). Alpha-2 agonists also have some perioperative functions, such as adrenergic analgesia, attenuation of vasodilatation, and potentiation of sedation, as well as improving respiratory and haemodynamic function (Di Bella et al. 2020; Gruen et al. 2022). NSAID may not be appropriate for all patients, and their administration before or after surgery should be considered individually (Kukanich et al. 2012; Gruen et al. 2022).

Perioperative drug management also includes surgical antimicrobial prophylaxis (SAP) to minimise the morbidity and/or mortality associated with surgical site infections (SSI) despite the negative effects of their use, especially antimicrobial resistance (AMR) (European Union 2019; Menz et al. 2021). SAP is not necessary for all surgical procedures and should be administered according to well-defined principles. Nevertheless, excessive SAP administration occurs frequently worldwide and contributes to overall antibiotic consumption in human surgical units (Worldwide Antimicrobial Resistance National/International Network Group (WARNING) Collaborators 2023). Approximately 15% of all antibiotics prescribed in human hospital settings are SAP related and are crucial drivers of AMR (Sartelli et al. 2016). Limited research has been published regarding the perioperative use of antibiotics in small animals, particularly in Spain (Knights et al. 2012; Ekakoro and Okafor 2019; Gómez-Beltrán et al. 2021; Otero Balda et al. 2023). The use of antimicrobials should be addressed within the One Health approach, as a way of integrating people, animals, and the environment. In the European Union (EU), the EMA’s Antimicrobial Advice Ad Hoc Expert Group (AMEG) has developed a veterinary antibiotic classification system that considers the risk that their use may cause public health problems, such as AMR development, and the need to use them in veterinary medicine (European Medicines Agency 2019).

Prevalent surgical procedures include spaying, neutering, dental extractions, and excision of benign skin growths. The prevailing urgent care surgeries for pets include addressing skin abscesses or lacerations, repairing fractures, managing internal bleeding, resolving intestinal obstructions related to foreign bodies, treating torn cruciate or cranial cruciate ligament ruptures, removing masses, addressing bladder stones or urethral blockages, and managing malignant skin tumours (East Orlando Animal Hospital 2021).

Expert panels have provided several guidelines for standardising anaesthesia protocols in small animals (Grubb et al. 2020; Federation of European Companion Animal Veterinary Associations 2021; Monteiro et al. 2022). Furthermore, although different cross-sectional studies assessing anaesthetic and analgesic procedures employed in general veterinary practice show that veterinarians implement comparable protocols regardless of geographical region and level of training (Wagner and Hellyer 2000; Sano et al. 2018; Gates et al. 2020a), little is known about the actual perioperative pharmacological management at veterinary clinics and hospitals.

To our knowledge, this is the first study to evaluate the current use of perioperative drugs in elective and emergency surgeries at a Veterinary Teaching Hospital in Spain. Therefore, this study aimed to analyse the current use of perioperative drugs in companion animals treated surgically at the Veterinary Teaching Hospital of the University of León (HVULE) (Spain) over a period of 5 years (2018–2022), describe the prescription patterns of these medicines, identify the main variables associated with this prescription, and explore potential strategies to promote their appropriate use, including the prudent use of antibiotics in surgery.

2. Materials and methods

2.1. Study site

HVULE is a veterinary teaching hospital of the Faculty of Veterinary Medicine at the University of Leon (Spain). This hospital is organised into two separate clinical departments according to the target species, small and large animals, both with their emergency facilities. All teaching staff were qualified veterinarians during the study period. For companion animals, three trained veterinarians performed surgeries at the HVULE. They were accredited by the Spanish Small Animals Veterinary Association (AVEPA) in traumatology/orthopaedics and soft tissue surgery and were also recognised at the European level.

2.2. Study design and data collection

This retrospective study was designed to evaluate the drug prescription pattern in both elective and emergency surgical interventions performed at the HVULE between 1 January 2018 and 31 December 2022.

Data on dogs and cats during the 5-year period were collected using an inductive method. Data were obtained by reviewing individual medical records stored on the online management platform GestorVet (GestorVet, Las Palmas de Gran Canarian, Spain). The following data were collected for each animal: history number, year of clinical record, date of birth, species, breed, sex, weight, age group (paediatric, < 5 months, adult, 5 months–7 years; and geriatric, > 7 years) (Arnason Hart et al. 2022), American Society of Anesthesiologist (ASA) classification (ASA 1, healthy patient; ASA 2, mild systemic disease; ASA 3, moderate systemic disease; ASA 4, severe systemic disease that is a constant threat to life; and ASA 5, not expected to survive 24 h without the operation) (Portier and Ida 2018), surgical procedure, and the drugs employed in the surgery, including the route of administration and dosage regimen. All medications were categorised according to the WHO anatomical-therapeutic-chemical veterinary (ATCvet) classification system (World Health Organization Collaborating Centre for Drug Statistics Methodology 2023). The European Medicines Agency (EMA) categorisation of antibiotics for use in animals was used to classify antibiotics (Category A: Avoid; Category B: Restrict; Category C: Caution; Category D: Prudence) (European Medicines Agency 2019). Data were exported to Microsoft Access and Excel (Microsoft Office 2019). Patients whose medical records lacked essential information, those with an ASA score ranging from 3 to 5, or those who died before or during surgical intervention were excluded.

Assuming a margin of error of 0.05, and a significance level of 0.05, the minimum sample size was estimated to be 385 animal surgeries (Stevenson 2020).

The Strengthening the Reporting of Observational Studies in Epidemiology—Veterinary Extension (STROBE-Vet) Statement was used to report data (O’Connor et al. 2016).

2.3. Calculation of the administered daily amount

The Daily Administered Amount (ADA, expressed in grams) was calculated for each drug, using the following formula:

$$\begin{array}{l}\text{Administered\hspace{0.33em}Daily\hspace{0.33em}Amount\hspace{0.33em}}\left(\text{ADA}\right)\text{\hspace{0.33em}=\hspace{0.33em}Amount\hspace{0.33em}of\hspace{0.33em}drug\hspace{0.33em}}\\ \times \text{\hspace{0.33em}Proportion\hspace{0.33em}of\hspace{0.33em}active\hspace{0.33em}ingredient\hspace{0.33em}in\hspace{0.33em}this\hspace{0.33em}drug}\end{array}$$

2.4. Calculation of the prescription diversity

Prescription Diversity (PD) is defined as ‘the frequency and variety with which a practice prescribes pharmaceutical classes (PC) within a determined pharmaceutical family (PF)’ (Singleton et al. 2018). It is calculated as follows, and adjusted to a scale of 0–1, where 1 represents maximal diversity:

$$\text{Prescription\hspace{0.33em}Diversity\hspace{0.33em}}\left(\text{PD}\right)\text{\hspace{0.33em}=\hspace{0.33em}1}-\hspace{0.17em}\frac{\sum \text{np\hspace{0.17em}(np}-\text{1)}}{\text{NP\hspace{0.17em}(NP}-\text{1)}}$$

Where np is the number of prescriptions for a particular PC within a PF, and NP is the total number of prescriptions within a PF.

2.5. Data management and statistical analysis

Data analyses were performed using IBM SPSS Statistics version 26 (IBM Corporation, Armonk, NY, United States). Descriptive statistics (means, standard deviations, ranges, and frequencies with 95% confidence intervals) were used to evaluate the study population. Chi-square (χ²) and Fisher’s exact tests were used to compare qualitative variables. Odds ratios (OR) were calculated using their respective 95% confidence intervals (95% CI). Multivariate forward-step ordinal logistic regression analysis was conducted to assess the impact of each predictor on the type of surgery and the use of antibiotics. A p-value of ≤ 0.05 was considered significant.

3. Results

Data were available from a total of 20563 medical records. Of those, 17483 (85.0%) were medical records from the 5-year study period, with 3519 surgical procedures (20.1%). After applying the exclusion criteria, 3438 animals were included in the study. All patients had an ASA status scale ≤ 2.

Table 1 summarises the background and clinical information of the assessed animals. The proportion of surgical interventions increased significantly between 2018 and 2019 and between 2021 and 2022 (χ² test = 45.243; p < 0.001), whereas in 2020 they dropped. Dogs were the most frequent animals (58.2%) undergoing surgery, with mongrels (27.5%) and greyhounds (10.9%) being the most common breeds. Among the cats, the most common breed was Common European (94.9%). When considering sex distribution within species, a predominance of females over males in both species (65.9% felines and 52.8% canines) was observed. Male dogs were 1.7 times more likely to undergo surgical intervention than male cats (OR = 1.731; 95% CI = 1.502-1.994; p < 0.001).

Table 1.

Background and clinical characteristics of the companion animal surgeries at the HVULE (Spain) between 2018 and 2022.

Characteristics	Frequency (%)(n = 3438)	95% CI
Year of surgery
2018	587 (17.1%)	0.158-0.183
2019	724 (21.1%)	0.197-0.224
2020	635 (18.4%)	0.172-0.198
2021	676 (19.7%)	0.183-0.210
2022	816 (23.7%)	0.223-0.252
Animal species
Canine	2.000 (58.2%)	0.573-0.590
Feline	1438 (41.8%)	0.410-0.427
Sex
Female	1960 (57.0%)	0.562-0.579
Male	1406 (40.9%)	0.401-0.417
Unknown	72 (2.1%)	0.019-0.023
Age group
Paediatric	76 (2.2%)	0.020-0.025
Adult	2511 (73.0%)	0.723-0.738
Geriatric	714 (20.8%)	0.201-0.215
Unknown	137 (4.0%)	0.037-0.043
Surgical intervention system
Reproductive	1943 (56.6%)	0.557-0.574
Traumatological	682 (19.8%)	0.192-0.205
Nodulectomy	196 (5.7%)	0.053-0.061
Dental	93 (2.7%)	0.024-0.030
Digestive tract	91 (2.6%)	0.024-0.029
Ocular	66 (1.9%)	0.017-0.022
Nephrourological	32 (1.0%)	0.008-0.011
Endoscopy	29 (0.8%)	0.007-0.010
Laparotomy	26 (0.8%)	0.006-0.009
Other procedures	280 (8.1%)	0.077-0.086
Number of drugs given in each procedure
1	24 (0.7%)	0.006-0.008
2	31 (0.9%)	0.008-0.011
3	72 (2.1%)	0.019-0.024
≥ 4	3311 (96.3%)	0.959-0.966

CI: confidence interval.

Almost three-quarters of participants were adults. Paediatric cats were 29 times more likely to undergo surgery than paediatric dogs (OR = 28.878; 95% CI = 16.345-51.030; p < 0.001). Conversely, geriatric dogs were 19 times more likely to undergo surgery than geriatric cats (OR = 18.908; 95% CI = 13.402-26.678; p < 0.001).

Regarding sex, reproductive (65.6% females and 32.0% males) and traumatological (44.1% females and 54.8% males) surgeries were the most frequent procedures. Females were 2.5 times more likely to undergo a reproductive surgery than a traumatological one (OR = 2.557; 95% CI = 0.759-1.118; p < 0.001). In terms of age (paediatric, adult, and geriatric animals), the prevalence of reproductive (43.6%, 66.3%, and 18.7%, respectively) and traumatological interventions (32.0%, 18.5%, and 24.9%, respectively) was clearly higher than that of other surgeries and differed between age groups. Regarding surgical procedures of the reproductive system, almost all (n = 1907; 97.9%) were spaying, mainly ovariohysterectomies (n = 1275; 66.9%) and orchiectomies (n = 632; 33.1%). In all categories of surgical intervention, the species ratio favoured dogs, with the exception of procedures involving the reproductive tract. Notably, over 87.3% of cats underwent reproductive tract surgeries compared to 34.3% of dogs.

3.1. Perioperative drugs

Regarding perioperative drug treatments (n = 26065), the mean number of medicines administered in each surgical procedure was 7.4 ± 2.2 (range 1-14; median 7). Four or more drugs were used in 96.3% of the cases, with 23852 (91.5%) of the drugs being included in the ATCvet classification. Almost half of the patients who underwent reproductive (47.6%) or traumatological surgeries (51.8%) received a high number of medications (8–9 and 7–8 drugs, respectively). Table 2 lists these treatments according to the first level of ATC classification, showing that more than half belonged to group QN (nervous system). When the fifth level was considered, the most common chemical substance was isoflurane.

Table 2.

Anatomical groups (first-level ATCvet and ATC) and subgroups for chemical substances (fifth-level ATCvet and ATC) used in surgical interventions at the HVULE (Spain) between 2018 and 2022.

First and fifth-level ATCvet	Frequency (%)(n = 23852)	95% CI
Group QA Alimentary tract and metabolism	1630 (6.8%)	0.067-0.070
QA04AD90 maropitant	1630 (6.8%)	0.067-0.070
Group QB Blood and blood-forming organs	3109 (13.0%)	0.128-0.132
QB05BB01 electrolytes	1390 (5.8%)	0.056-0.060
QB05BB02 electrolytes with carbohydrates	5 (0.02%)	0.0001-0.0003
QB05CX01 glucose	14 (0.06%)	0.0004-0.0007
QB05XA03 sodium chloride	1700 (7.1%)	0.069-0.073
Group QH Systemic hormonal preparations, excluding sex hormones and insulins	11 (0.05%)	0.0003-0.0006
QH02AB04 methylprednisolone	11 (0.05%)	0.0003-0.0006
Group QJ Antiinfectives for systemic use	1540 (6.5%)	0.063-0.067
QJ01CR02 amoxicillin and beta-lactamase inhibitor	872 (3.7%)	0.035-0.039
QJ01MA93 marbofloxacin	668 (2.8%)	0.027-0.029
Group QM Musculoskeletal system	2668 (11.2%)	0.110-0.114
QM01AC06 meloxicam	982 (4.1%)	0.040-0.043
QM01AH91 robenacoxib	1686 (7.1%)	0.069-0.072
Group QN Nervous system	14894 (62.3%)	0.621-0.625
QN01AB06 isoflurane	3230 (13.5%)	0.133-0.138
QN01AB08 sevoflurane	146 (0.6%)	0.006-0.007
QN01AX05 alfaxalone	916 (3.8%)	0.037-0.040
QN01AX10 propofol	3024 (12.7%)	0.125-0.129
QN01BB02 lidocaine	353 (1.5%)	0.014-0.016
QN02AB03 fentanyl	763 (3.2%)	0.032-0.035
QN02AC90 methadone	3226 (13.5%)	0.133-0.138
QN02AF01 butorphanol	69 (0.3%)	0.002-0.003
QN05CM18 dexmedetomidine	2940 (12.3%)	0.121-0.126
QN05CM91 medetomidine	195 (0.8%)	0.008-0.009
First and fifth-level ATC	Frequency (%)(n = 2213)	95% CI
Group J Antiinfectives for systemic use	561 (25.4%)	0.245-0.263
J01DB04 cefazolin	561 (25.4%)	0.245-0.263
Group N Nervous system	1652 (74.6%)	0.737-0.756
N01BB01 bupivacaine	522 (23.6%)	0.226-0.246
N05CD08 midazolam	1130 (51.0%)	0.500-0.522

CI: confidence interval.

3.1.1. General anaesthetics (QN01A)

Inhalation agents were used in almost all surgical interventions (98.2%). Isoflurane was the primary choice for maintenance (93.9%), whereas sevoflurane was used in only 4.2% of cases. Thus, the PD of halogenated hydrocarbons was 0.09. Isoflurane was used in 91.4% and 97.5% of dog and cat surgeries, respectively, with inhalation doses ranging between 1 and 1.5%. As for sevoflurane, doses ranged from 2 to 2.5% and were used in 6.4% of canine surgeries and 1.3% of feline surgeries.

Other commonly used general anaesthetics were propofol and alfaxalone, both of which are preferentially used to induce anaesthesia. They were only used for maintenance in 1.7% of surgeries. The PD for the other general anaesthetics was 0.36. Propofol doses ranged from 2 to 4 mg/kg by intravenous (IV) route, and for alfaxalone between 0.2 and 2 mg/kg IV and 0.1-5 mg/kg for the intramuscular (IM) route. Propofol and alfaxalone were administered in 85.7% and 16.1% of surgeries in dogs and 91.2% and 41.4% of surgeries in cats, respectively.

3.1.2. Hypnotics and sedatives (QN05 and N05)

This group was used in 93.1% of all surgeries, with a PD of 0.5. Dexmedetomidine was used in 85.5% of the interventions (82.6% in dogs and 89.6% in cats) as a premedication agent, with doses ranging 0.3–30 µg/kg IM or 0.5–10 µg/kg IV. For medetomidine (7.6% cats and 4.3% dogs), doses ranged from 4 to 40 µg/kg IM. There are no veterinary medicinal products available for midazolam, but it is also widely used as a veterinary surgery premedication (30.2% in dogs and 36.5% in cats), with IV doses of 0.03–0.5 mg/kg.

3.1.3. Opioids (QN02A)

Opioids were used for both preoperative and postoperative analgesia (96.6% of the procedures). Methadone (91.8%) was the most commonly administered drug, followed by fentanyl (32.0%) and butorphanol (2.9%). These three drugs were also commonly used in cat surgeries, although the frequency was lower for the latter two substances (methadone, 96.7%; fentanyl, 8.7%; and butorphanol, 0.8%). The PD of the opioids were 0.33. IV dose of methadone was 0.2 mg/kg, whereas for the IM route ranged from 0.04 to 1.3 mg/kg; IV doses of fentanyl ranged from 2.5–3 µg/kg, and IM doses of butorphanol were 0.1–0.4 mg/kg.

3.1.4. Antiinflammatory (QM01)

Drugs in this group were administered postoperatively in 77.5% of all patients (n = 2665). Robenacoxib was used in 1686 procedures (49.0%) and meloxicam in 982 (28.6%). The PD of NSAIDs was 0.47. Robenacoxib was used in 49.9% of canine surgeries, followed by meloxicam (18.0%). In cats, both the drugs showed similar frequencies (47.8% for robenacoxib and 43.3% for meloxicam). Robenacoxib doses were 2 mg/kg IM and 1-2 mg/kg by subcutaneous (SC) and IV routes. As for meloxicam, doses were 0.2 mg/kg per oral (PO), 0.1-2 mg/kg SC, and 0.05-0.2 mg/kg IM and IV.

3.1.5. Antibiotics (QJ01 and J01)

These drugs were used for surgical prophylaxis in 59.9% of cases, and the PD was 0.66. Amoxicillin/clavulanic acid (34.3%), cefazolin (25.3%), and marbofloxacin (1.6%) were the most commonly used antibiotics in dogs. Nevertheless, the frequencies in cats were clearly different, as marbofloxacin (44.3%) was the most prescribed perioperative antibiotic, followed by amoxicillin/clavulanic acid (13.0%) and cefazolin (3.9%). The doses administered ranged from 15 to 20 mg/kg for amoxicillin/clavulanic acid SC, IM, and PO. Cefazolin doses used were 20 mg/kg SC, IM, and 50 mg/kg IM. Marbofloxacin was administered intravenously (2 mg/kg) or SC (5 mg/kg).

Table 3 shows the results of the multivariate analysis performed to identify the variables associated with antibiotic use. Being female and having undergone traumatological or reproductive surgery had a significant impact on the use of antibiotics. A significant annual increase was observed during the study period.

Table 3.

Multivariate ordinal logistic regression analysis of factors relevant for the use of antibiotics in surgery.

Variables	OR (95% CI)	p-value
Sex (female)	1.202 (1.046-1.381)	0.010
Year	1.933 (1.827-2.045)	< 0.001
System surgery (other vs. reproductive)	2.077 (1.760-2.451)	< 0.001
System surgery (other vs. traumatological)	2.548 (2.063-3.147)	< 0.001

CI: confidence interval; OR: odds ratio.

According to the EMA Categorisation of antibiotics for prudent and responsible use in animals (European Medicines Agency 2019), 668 of the active ingredients administered were classified as Restrict (marbofloxacin; 31.8%) and 1433 as Caution (cefazolin and amoxicillin/clavulanic acid; 68.2%). Most antibiotics were administered parenterally (n = 2100), and only one animal was administered antibiotics orally. Figure 1 shows the antibiotic prescriptions according to EMA categorisation and year of study.

Figure 1.

Antibiotic prescription according to their european Medicines agency (EMA) categorisation and study year.

3.1.6. Local anaesthetics (QN01B and N01B)

Local anaesthetics were used in 16.7% of surgical procedures, and the PD was 0.48. These drugs were associated with general anaesthetics, NSAIDs, alpha-2 adrenergic agonist, and opioids in 16.6%, 93.9%, 96.2%, and 96.9% of cases, respectively. Although bupivacaine is not approved as a veterinary medicinal product, it was the most commonly administered drug (59.7%), followed by lidocaine (40.3%). Both molecules were mainly used to perform regional nerve blocks in different regions (92.6%), including the abdominal (56.5%), mesovarium (31.5%), intratesticular (1.7%), dental (0.7%), and thoracic (0.3%) areas. Differences were observed between the main indications for both drugs, as in the intratesticular and mesovarium splashing blocks only lidocaine was employed, and in most of abdominal surgical interventions (including the intraperitoneal technique), the drug employed was bupivacaine (98.6%). Bupivacaine was used more frequently (10.7% and 21.5%) than lidocaine (7.0% and 14.9%,) in both dog and cat surgeries, respectively. The doses of lidocaine were 2-6 mg/kg for nerve blockade, 1-3 mg/kg/h for continuous rate infusion (CRI), and 1 mg/kg topically (TOP). Bupivacaine was administered at doses of 2-4 mg/kg for nerve block and 2 mg/kg IV or TOP.

3.1.7. Antiemetics (QA04AD)

Antiemetics were administered during almost half of the procedures (47.4%). Maropitant was the only drug used in this group (44% in dogs and 37.5% in cats). Doses were 0.3 mg/kg IM, 0.2-3 mg/kg SC and 1-2 mg/kg IV.

3.1.8. Glucocorticoids

Methylprednisolone was administered only in 11 surgical procedures in dogs and as an antiemetic or antiinflammatory treatment (1–2 mg/kg IV).

3.1.9. Blood substitutes and perfusion solutions (QB05)

Fluid therapy was used in 3008 surgeries (87.5%), with intravenous solution additives (QB05X) being most commonly administered (54.7%), followed by intravenous solutions (QB05B) (44.8%). The first group (QB05X) was used in more than half of cats (55.1%) and nearly half of dogs (45.4%), whereas the second group (QB05B) was used in 36.2% of cats and 43.8% of dogs.

Table 4 shows the ADA prescribed for animals undergoing surgical treatment at the HVULE. Altogether, a total of 557.7 grams of antibacterial agents for systemic use were documented from animal surgeries over a five-year period. No significant differences were observed in the use of antibiotics between dogs and cats (OR = 1.007; 95% CI = 0.940-1.079; p = 0.844).

Table 4.

The daily amount prescribed for companion animals undergoing surgical treatment at the HVULE for chemical substance (fifth-level ATCvet and ATC).

Pharmacotherapeutic group	mg	g	%
General anaesthetics
Propofol (QN01AX10)	160630.9	160.6	96.7
Alfaxalone (QN01AX05)	5743.9	5.7	3.3
Total	166074.8	166.3	100
Hypnotics and sedatives
Dexmedetomidine (QN05CM18)	205.6	0.2	5.6
Medetomidine (QN05CM91)	27.9	0.03	0.8
Midazolam (N05CD08)	3462.9	3.5	93.7
Total	3696.4	3.7	100
Opioids
Butorphanol (QN02AF01)	687.8	0.7	5.1
Methadone (QN02AC90)	12717.4	12.7	94.6
Fentanyl (QN02AB03)	38.5	0.04	0.3
Total	13443.7	13.4	100
Antiinflammatory (NSAID)
Meloxicam (QM01AC06)	1552.5	1.5	3.6
Robenacoxib (QM01AH91)	41367.9	41.4	96.4
Total	42920.4	42.9	100
Antibiotics
Amoxicillin/clavulanic acid (QJ01CR02)	321333.9	321.3	57.6
Cefazolin (J01DB04)	221250.6	221.2	39.7
Marbofloxacin (QJ01MA93)	15208.8	15.2	2.7
Total	557793.3	557.7	100
Antiemetics
Maropitant (QA04AD90)	22584.5	22.6	100
Total	22584.5	22.6	100
Glucocorticoids
Methylprednisolone (QH02AB04)	200.6	0.2	100
Total	200.6	0.2	100

NSAID: non-steroidal antiinflammatory drugs.

3.2. Surgical interventions

3.2.1. Reproductive surgery

More than a half of the surgeries were performed on the reproductive system (56.6%). As mentioned above, these were ovariohysterectomies (66.9%) and orchiectomies (33.1%). Females had a 2.2 times higher risk of undergoing reproductive surgery (OR = 2.179; 95% CI = 1.845-2.577; p < 0.001).

Table 5 shows the results of multivariate analysis performed to identify the variables associated with reproductive system surgery in animals. Female sex; being a cat; and the use of NSAID, local anaesthetics, and antiemetics were associated with a significant probability of this type of surgery. Additionally, there was a significant year-on-year increase.

Table 5.

Multivariate ordinal logistic regression analysis of factors relevant to undergoing reproductive surgery.

Variables	OR (95% CI)	p-value
Animal species (cat)	11.012 (8.899-13.626)	< 0.001
Sex (female)	1.734 (1.437-2.092)	< 0.001
Year	0.806 (0.718-0.904)	< 0.001
Antiinflammatory (NSAID)	9.490 (7.168-12.565)	< 0.001
Local anaesthetics	3.941 (2.871-5.410)	< 0.001
Antiemetics	1.700 (1.168-2.476)	0.006
Antibiotics (not use)	1.993 (1.539-2.581)	< 0.001

CI: confidence interval; NSAID: non-steroidal antiinflammatory drugs; OR: odds ratio.

3.2.2. Traumatological surgery

Traumatological was the second most frequent surgical intervention (19.8%). Within this group, fracture repair (33.9%), hemilaminectomy (14.7%), resolution of cruciate ligament rupture (13.4%), and other procedures such as elbow or knee dislocation or amputation (38.0%) were the most frequently performed.

Multivariate analysis was also performed for variables associated with traumatological surgery (Table 6). Being a dog; male; and using antibiotics, hypnotics, sedatives, and antiemetics had a significant impact on this surgery.

Table 6.

Multivariate ordinal logistic regression analysis of factors relevant to undergoing traumatological surgery.

Variables	OR (95% CI)	p-value
Animal species (dog)	5.771 (4.386-7.594)	< 0.001
Sex (male)	1.359 (1.120-1.648)	0.002
Antibiotics	3.989 (3.008-5.053)	< 0.001
Hypnotics and sedatives	2.303 (1.521-3.488)	< 0.001
Antiemetics	1.629 (1.140-2.326)	0.007
Antiinflammatory (NSAID, not use)	3.131 (2.460-3.985)	< 0.001
Local anaesthetics (not use)	6.079 (3.825-9.661)	< 0.001

CI: confidence interval; NSAID: non-steroidal antiinflammatory drugs; OR: odds ratio.

4. Discussion

To our knowledge, this is the first report describing the perioperative drug patterns in pets in a veterinary teaching hospital over a long period. Most studies found in the literature are surveys (Clarke and Hall 1990; Dodman and Lamb 1992; Wagner and Hellyer 2000; Nicholson and Watson 2001; Sano et al. 2018; Kramer et al. 2022; Otero Balda et al. 2023), not retrospective studies, such as ours. In recent decades, diagnostic and surgical procedures have become increasingly sophisticated and long. Consequently, anaesthesia caseloads in veterinary teaching hospitals have become more complex, requiring more time and expertise to manage them effectively (Wagner and Hellyer 2000).

We observed that the number of surgeries increased in the HVULE over the 5 years of this study, with the exception of 2020, which coincided with the onset of the COVID-19 pandemic. A similar reduction was described in another study where spay-neuter surgeries performed between March and April 2020, with a subsequent rebound in May 2020 (Arnason Hart et al. 2022). This decrease may be attributed to the suspension of non-urgent surgical interventions, such as preventive care, treatment of minor or chronic conditions, and routine spay-neuter surgeries (American Veterinary Medical Association 2020; Hoffman 2021; Kogan et al. 2021).

Dogs are the most prevalent surgical cases in the HVULE. In Europe, 90 million households (46% of the total) own at least one pet, with cats being their preferred choice (> 113 million) (FEDIAF European Pet Food 2020). However, in Spain, dogs were the favoured animal (21.9%), followed by cats (8.2%) (Ministerio de Agricultura, Pesca y Alimentación 2015). In 2021, there will be over 30 million pets in Brazil, including over 9 million dogs and over 5 million cats (Asociación Nacional de Fabricantes de Alimentos para Animales de Compañía 2021). A study conducted at Ashanti Regional Veterinary Clinical Hospital (Ghana) also indicated that surgeries were more frequent in canines than in other companion animals (Eyarefe and Dei 2014).

When patient age was considered, a higher frequency of traumatological surgery was observed in geriatric animals (24.8% vs. 18.5% in adult animals). Contrastingly, reproductive surgery was more common in adults than in geriatric animals (66.4% vs. 18.7%). Nodulectomies were more frequent in geriatric patients (18.5% vs. 2.2% in adults). Finally, antibiotic therapy was prescribed more frequently to adults than geriatric animals (60.8% vs. 43.1%).

Regarding surgical procedures, those related to the reproductive system were the most frequently performed, including practically all spaying surgeries. Canine and feline ovariectomies and ovariohysterectomies are the most prevalent elective surgical procedures in small-animal clinical practice (Gates et al. 2020a) to prevent or reduce the risk of future health problems, such as pyometra, mammary gland or reproductive tract neoplasias (Root Kustritz 2012), unplanned breeding, and solve the global issue of pet overpopulation through spaying or neutering procedures (Hiby et al. 2017). According to a Spanish study, 70.2% of cats were sterilised and only 32.3% of dogs were sterilised in 2015 (Ministerio de Agricultura, Pesca y Alimentación 2015), a trend that is likely to increase in the coming years, at least in this country, due to the implementation of Law 7/2023 on the Protection of the Rights and Welfare of Animals, which requires domestic cats to be sterilised (Boletín Oficial del Estado 2023). In the US, spaying is an integral part of shelter policies, with ‘neuter-before-adoption’ strategies gaining acceptance in the 1990s and being supported by veterinary training programs (Shivley et al. 2018; Gates et al. 2020b; Rigdon-Brestle et al. 2022). In Ghana, surgeries involving wounds and trauma are the most prevalent among dogs, followed by neutering at the second place (Eyarefe and Dei 2014). In terms of anaesthetic and analgesic protocols for ovariohysterectomies, propofol (66.6%) and isoflurane (73%) were the predominant anaesthetics used by US veterinarians, whereas carprofen and buprenorphine were the most common postoperative analgesics (Kramer et al. 2022). These protocols aligned closely with the procedures followed in the HVULE.

Although guidelines have been published by the American Animal Hospital Association (AAHA) (Grubb et al. 2020) and The World Small Animal Veterinary Association (WSAVA) (Monteiro et al. 2022), and an initiative by The Federation of European Companion Animal Veterinary Associations (FECAVA) (Federation of European Companion Animal Veterinary Associations 2021) on basic anaesthesia and analgesia practices, there are no universally recognised good practices for veterinary anaesthesia. Regarding perioperative treatment, isoflurane was the most frequently prescribed active ingredient (93.9%). Halogenated inhalation anaesthetics, such as isoflurane and sevoflurane, are extensively utilised to maintain general anaesthesia in animals because no venous infusion pump is required, which makes the procedures less expensive (Steffery et al. 2015). Isoflurane is the most commonly used drug in North America (Brosnan and Steffey 2018), followed by sevoflurane (Steffery et al. 2015). Isoflurane undergoes less hepatic biotransformation than sevoflurane and is cheaper (Eger 1994; Safari et al. 2014), whereas sevoflurane is less soluble than isoflurane, resulting in fast induction (Pottie et al. 2008; Brosnan and Steffey 2018). Nevertheless, the risks posed by these compounds to veterinarians must be considered. Additionally, beyond the exposure by professionals working in operating rooms to anaesthetics (Mastrangelo et al. 2013; Casale et al. 2014; Paes et al. 2014; Souza et al. 2016; Braz et al. 2018), residual quantities of these substances may be eliminated outdoors, resulting in an environmental impact (Vollmer et al. 2015). The environmental implications of anaesthetic gases are also related to the One Health concept (Kaplan et al. 2009; One Health Initiative 2023). Moreover, halogenated anaesthetic drugs are considered greenhouse gases, and they contribute to harm the global environment (Campbell and Pierce 2015). In this sense, the European Parliament and Council published a regulation on fluorinated greenhouse gases, which states that the use of desflurane as an inhalation anaesthetic will be banned from 1 January 2026 except when its use is strictly necessary and no other anaesthetic can be used on medical grounds. In these exceptional cases, desflurane, similar to all the other gases, should be captured (European Union 2024).

Propofol was extensively used in this study. It is administered as an induction agent and, in some cases, for maintenance of anaesthesia (Simon and Steagall 2020). One of its advantages is dosing flexibility, which allows for a wide spectrum of responses ranging from light to profound sedation and muscle relaxation. Moreover, it is rapidly and consistently metabolised (Posner 2018a). Some surveys have shown that propofol is the preferred induction agent in dogs (Clarke and Hall 1990; Dodman and Lamb 1992; Nicholson and Watson 2001). Alfaxalone is a synthetic neurosteroid anaesthetic drug that shares some characteristics with propofol, and both are used as preferred induction agents in cats (Sano et al. 2018). In our study, alfaxalone was used mainly in females (62.2%) and adults (75.8%), in reproductive surgery (68.9%). In the latter type of surgery, 87% of the cats were administered IM. Several studies and protocols have considered alfaxalone a suitable option for the induction and maintenance of anaesthesia in neutering cats (Beths et al. 2014; Warne et al. 2015; Federation of European Companion Animal Veterinary Associations 2021; Monteiro et al. 2023). Moreover, IM alfaxalone provided adequate sedation to allow for better management of this animal species (Tamura et al. 2015; Guo et al. 2024). IM alfaxalone was also common in dogs (92.6%), as it is associated with medetomidine and/or butorphanol and is considered a clinically effective protocol, allowing the administration of lower doses without causing severe cardiorespiratory depression in healthy dogs (Kato et al. 2021). Moreover, this drug is useful in aggressive or nervous patients.

Over the past few decades, multimodal analgesia has gained importance in veterinary medicine for perioperative pain management. This strategy involves combining different analgesic drugs to target distinct pain pathways using lower doses, minimising at the same time those side effects associated with the treatment (Lamont 2008). Alpha-2 adrenergic agonists, opioids, NSAID, and local anaesthetics are used in HVULE. Alpha-2 adrenergic agonists are administered pre- and postoperatively to provide sedation and analgesia, opioids in the perioperative period to manage pain and reduce anaesthetic requirements, and NSAID to diminish opioid requirements in the postoperative period and provide more analgesia with fewer side effects (Lemke and Creighton 2010; Valverde and Skelding 2019).

Dexmedetomidine and medetomidine are selective alpha-2 adrenergic agonists currently used in small animal practice (Lemke and Creighton 2010; Valverde and Skelding 2019). In the HVULE, dexmedetomidine was mainly used in females (57.2%), adults (74.9%), the IM route (98.9%), and reproductive surgeries (59.4%). Alfaxalone was also associated with premedication (29.1%), mostly during reproductive system surgeries in cats (64.2%). Alfaxalone is widely used in clinical practice as a sedative drug, alone or with dexmedetomidine or opioids (Kellihan et al. 2015; Trimble et al. 2018; Arenillas et al. 2020), owing to its analgesic properties and minimal influence on the respiratory drive (Giovannitti et al. 2015; Adami et al. 2016; Di Bella et al. 2020; DeGroot et al. 2020; Pleyers et al. 2020; Guo et al. 2024). Moreover, dexmedetomidine was associated with opioids in almost all surgeries (98.4%). It is an active isomer of medetomidine and approximately twice as potent as medetomidine. Medetomidine has a rapid onset after parenteral injection, and its duration of action is approximately 1 h in dogs. It was predominantly used in reproductive surgeries (64.6%) and cats (80.2%), primarily in very nervous, young, and/or aggressive patients.

Midazolam is a benzodiazepine which produces sedation, hypnosis, anxiolysis, anterograde amnesia, and muscle relaxation and has an anticonvulsive effect (Posner 2018b). It is more potent than diazepam, has similar pharmacodynamics, and is often used interchangeably in clinical settings. Midazolam is frequently used in veterinary medicine as it can be administered via several routes, does not cause pain upon injection, and is less expensive than diazepam (Posner 2018b). This drug was used in 32.7% of the surgeries performed at the HVULE to facilitate recovery in anaesthetised animals prone to excitability, provided that adequate analgesia was available. Midazolam-propofol (86.4%) was administered during the induction phase of anaesthesia. This combination also reduced the required propofol dose (Aguilera et al. 2020). The association midazolam-opioid was also employed with a high frequency (97.5%), which may be related to the fact that it provides cardiovascular stability and can be used to sedate dogs undergoing cardiovascular examinations (Ferreira et al. 2022). It should be noted that midazolam does not produce effective sedation in healthy dogs (BSAVA 2023) and cats (Ilkiw et al. 1996) and may actually induce abnormal behaviour.

Opioids are the most effective class of drugs for managing acute and chronic pain, and should be used as perioperative drugs combined with tranquillizers or sedatives (Grubb et al. 2020). Butorphanol, morphine, hydromorphone, and fentanyl are the drugs most commonly used in small animals (Lemke and Creighton 2010; Simon and Steagall 2017). At the HVULE, the veterinarians used methadone, fentanyl, and butorphanol. Fentanyl can be administered via IV as a preanesthetic agent, intraoperatively (at a constant infusion rate) or transdermally (Simon and Steagall 2017). It is approximately 100 times more potent than morphine and has a shorter duration of action. The drug was rarely used (22.2% of the surgeries), mainly in traumatological procedures (42.1%), such as fractures and reproductive surgical interventions (34.1%), and particularly in neutering (95%). A combination of fentanyl, propofol, methadone/butorphanol, and dexmedetomidine/medetomidine was used in most traumatological (85.1%) and reproductive (87.3%) surgeries. Furthermore, inhalation anaesthesia was used in most reproductive surgeries (98.6%). The combination of propofol and fentanyl infusions provided stable cardiovascular function and satisfactory surgical outcomes in dogs (Andreoni and Lynne Hughes 2009) and cats (Guilherme et al. 2003).

Butorphanol is an opioid partial agonist and antagonist that is most commonly used in small animals. However, this drug was rarely used in the present study. The analgesia is not as deep and its duration of action is short (1-2 h) (Pascoe 2000; Dyson 2008). Consequently, it is recommended when a surgery may cause mild to moderate pain, such as ovariohysterectomy and neutering (Mathews 2000). Moreover, side effects are less severe than those caused by morphine, hydromorphone or fentanyl, and it can be used as a preanesthetic drug (0.2-0.4 mg/kg IM) (Lemke and Creighton 2010), and postoperatively at lower doses (0.05 mg/kg IV) to reverse sedation and respiratory depression (Simon and Steagall 2017). In our study, this drug was administered during very few surgeries (2.0%), mainly in dogs (82.6%) and for diagnostic procedures (47.4%).

Methadone is a synthetic mu opioid agonist that has been licensed in Canada for veterinary use as a preanesthetic drug in cats (0.5 mg/kg IM), and in Australia, New Zealand, the United Kingdom (UK), and the European Union (EU) as an analgesic and preanesthetic agent in dogs (0.5-1 mg/kg IV, IM and SC) and cats (0.3-0.6 mg/kg SC). In these species an extralabel use could be also considered, as a preanesthetic or analgesic drug (0.2-0.5 mg/kg IV, IM or SC). It is characterised by its effects on pain pathways and a low incidence of vomiting (Kerr and Swanton 2023). Therefore, it is considered the most efficacious analgesic in ovariohysterectomies and other moderate-to-severely painful surgeries, such as orthopaedic surgeries (Hunt et al. 2013). In our study IV dose was 0.2 mg/kg, whereas the range for IM route was 0.04-1.3 mg/kg. In this case, a lower dose may minimise the adverse effects and maintain adequate intraoperative analgesia (Ryan et al. 2022). Moreover, the duration of the analgesic effect was approximately 4 h. Consequently, additional drugs may be administered, depending on the individual signs of pain in each animal (Kerr and Swanton 2023).

Regarding NSAIDs, two COX inhibitors were used exclusively (meloxicam and robenacoxib), both approved in the EU for perioperative use in companion animals (Committee for Veterinary Medicinal Products 2011; Committee for Veterinary Medicinal Products 2020). They are widely used in dogs and cats, either alone or with opioids in routine procedures according to multimodal analgesia techniques (Hunt et al. 2015), with better evidence than the unimodal protocol (Zanuzzo et al. 2015). However, patients treated with systemic analgesics, such as opioids or combinations of opioids with NSAID, showed lower mortality (Gil and Redondo 2013). Robenacoxib is a highly selective COX-2 inhibitor with analgesic, antiinflammatory, and antipyretic characteristics and a better safety margin for gastrointestinal and renal adverse effects than other COX-2 inhibitors (King et al. 2009). The lower use of meloxicam in the HVULE group may be related to the fact that it should be avoided in animals with hepatic, cardiac, or renal failure and bleeding disorders, as well as in severely dehydrated or hypovolemic animals, because of its potential risk of renal toxicity (Committee for Veterinary Medicinal Products 2011). Not using NSAIDs was significantly correlated with traumatological procedures (Table 6). NSAID inhibit the COX enzyme by decreasing prostanoid synthesis, which may delay bone healing; however, there is no consensus on the safety of this pharmacological group in orthopaedic surgeries (Marquez-Lara et al. 2016; Ozturan and Akin 2024).

Local anaesthetics prolong the duration of anaesthesia, potentiate its effects, and decrease the required doses of other perioperative drugs (Schwartz et al. 2022). Lidocaine and bupivacaine are the most commonly used agents, with ropivacaine gaining popularity (Martin-Flores 2019). Lidocaine acts faster and for a shorter time than bupivacaine or ropivacaine (Martin-Flores 2019), and is a low-risk and low-cost drug with a fast onset but short duration (DeRossi et al. 2016). In the HVULE group, local anaesthetics were employed in 84.5% of reproductive surgeries, with 62.1% of local blocks in neuter procedures. In 41 cases (4.7%), lidocaine was administered via CRI because of the risk of cardiac arrhythmias or because visceral analgesia was required. These local anaesthetics were mostly combined with alpha-2 adrenergic agonists and opioids (96.6 and 99.1%, respectively), as this combination prolonged the duration of the sensory block, increased the pain-free period, and diminished the postoperative analgesic requirements (Valverde and Skelding 2019; Grubb and Lobprise 2020a). Furthermore, local anaesthetics were mostly associated with NSAID (91.3%), as they produce effective analgesia via a dual mechanism, including inhibition of sodium ion influx through nerve axons and lower local expression of mediators by inhibition of the COX enzyme (De OL Carapeba et al. 2020; Margeti et al. 2024). As in the case of NSAID, the nonuse of local anaesthetics was significantly correlated with traumatological surgeries. In human medicine, efforts are being made to reduce the use of general anaesthetics and replace them with nerve blocks in orthopaedic surgery (Wang et al. 2016; Lee et al. 2022). However, in veterinary medicine, general anaesthesia is necessary because of the complex handling of patients, which implies that local anaesthetics are used to a lesser extent in this type of surgery. Several studies have indicated that locoregional anaesthesia also reduces anaesthesia-related mortality, as it improves cardiovascular and respiratory stability during the procedure and reduces hypnotic doses and perioperative stress (Perez et al. 2013; Romano et al. 2016).

While maropitant was employed at the HVULE to prevent opioid-induced vomiting, it was avoided in patients with liver disease and in animals with or predisposed to heart disease. Maropitant is administered with opioid-based premedication, reducing the frequency of vomiting and nausea and the time to eat postoperatively, thus improving overall comfort in dogs and cats (Hay Kraus 2013; Martin-Flores et al. 2017). This contrasts with the low use of methylprednisolone (in only 11 animals) as an antiemetic in the postoperative period, considering that the main indication for glucocorticoids in anaesthesia is the prevention of nausea and vomiting during the postoperative period (De Oliveira et al. 2013). However, it was administered at our hospital because of an associated neoplastic process (27.3%) and reduced spinal cord swelling (72.7%) (Nishida et al. 2016).

Preoperative antibiotic prophylaxis is defined as the administration of antibiotics before surgery to decrease the risk of postoperative infection (Crader and Varacallo 2023). Surgical site infections (SSI) are a major problem in veterinary medicine that contribute to increased morbidity, mortality, and costs (Espinel-Rupérez et al. 2019). However, surgical antibiotic prophylaxis (SAP) practices in veterinary medicine remain underexplored, with significant variations influenced by clinical environment, surgical caseloads, and geographical factors (Launcelott et al. 2019; Cockburn et al. 2022). SAP is commonly used to prevent infections, although there is an ongoing debate regarding the choice of appropriate antibiotics for each indication and treatment duration (Otero Balda et al. 2023). Although several studies have examined antibiotic use in small animals (Jessen et al. 2017; Tompson et al. 2021; Farrell et al. 2024), research focusing specifically on perioperative antibiotic use in companion animals, particularly in Spain, is lacking (Knights et al. 2012; Ekakoro and Okafor 2019; Gómez-Beltrán et al. 2021).

The Danish Antibiotic Use Guidelines for Companion Animal Practice specify that SAP cannot replace aseptic surgical techniques. Moreover, it should be administered immediately before surgery and should not normally be continued beyond the closure of the surgical incision. Ideally, the initial dose should be administered intravenously 30-60 min before skin incision and repeated at intervals of twice plasma half-life (in cefazolin, every 4 h) (Jessen et al. 2019). Similarly, Swedish guidelines advise against antimicrobial prophylaxis as a substitute for good asepsis and only as an adjunct to it, with clear indications (Bergvall et al. 2009). Similarly, the Finnish guidelines do not recommend prophylaxis for clean or clean-contaminated soft tissue surgeries lasting less than 60 min, but similar to the Danish guidelines, it advises a first-generation cephalosporin 30–60 min before incision for clean orthopaedic surgeries, continuing administration until the end of the procedure (The Finnish Food Safety Authority and the Faculty of Veterinary Medicine at the University of Helsinki 2018).

General guidelines advise against perioperative SAP in sterile settings and clean procedures, and postoperative SAP is rarely indicated (Jessen et al. 2019; Frey et al. 2022). The British Small Animal Veterinary Association (BSAVA) does not recommend prophylactic antimicrobials for routine neutering or uncomplicated dental procedures in dogs and cats, reserving SAP for specific cases, such as longer surgeries, procedures involving implants, immunosuppressed or debilitated patients, significant infections, bowel surgeries, dental procedures with periodontal disease, or contaminated wounds (Monteiro et al. 2022). Despite these guidelines, surveys have revealed that veterinarians frequently use antimicrobials in routine procedures including neutering, ovariohysterectomies, and dermal mass removal (Knights et al. 2012; Hardefeldt et al. 2017), which may contribute to the emergence of AMR (Branch-Elliman et al. 2019). In our study, patients who underwent reproductive surgeries were 2 times more likely to not use antibiotics (OR = 1.993; 95% CI = 1.539-2.581). In contrast, traumatological ones were four times more likely to use them (OR = 3.989; 95% CI = 3.008-5.053).

In this study, we observed that antibiotics from category A (Avoid), which should be administered to companion animals under exceptional circumstances, were not used at the HVULE. Nevertheless, those belonging to category B (Restrict), which are restricted to animals because of their critical importance in human medicine, were used in 668 cases (19.4%), although cultural examinations guiding the judicious use of antimicrobials were not routinely performed. According to the EMA Categorisation of antibiotics for use in animals, the use of category B (Restrict) antibiotics should be based on antimicrobial susceptibility testing (AST), wherever possible (European Medicines Agency 2019). Moreover, the World Organization for Animal Health (WOAH) has indicated that fluoroquinolones should not be used for prevention or as a first-line treatment, and their use should be based on the results of microbiological culture and AST (World Organisation for Animal Health 2024). European countries, such as Finland and Germany, are required to perform AST for the use of fluoroquinolones, such as marbofloxacin, and third- or fourth-generation cephalosporins (Bundes­ministeriums für Ernährung und Landwirtschaft 2017; The Finnish Food Safety Authority and the Faculty of Veterinary Medicine at the University of Helsinki 2018), and Spanish legislation has also made this requirement since January 2025 (Ministerio de la Presidencia, Relaciones con las Cortes y Memoria Democrática 2023). In Europe, the VETCAST group, a subcommittee of the European Committee on Antimicrobial Susceptibility Testing (EUCAST), is developing standards and operating processes for the AST of veterinary pathogens together with the European Network for Optimization of Antimicrobial Therapy (ENOVAT) to optimise veterinary antimicrobial use (Damborg et al. 2024).

Moreover, antibiotics belonging to Category C (Caution), which should be used only when there are no effective Category D (Prudence) antimicrobials, are frequently administered. Although category D was the first choice, it was not used in our study. Recent European legislation aims to reduce antimicrobial use by restricting prophylactic use, except in exceptional cases. Therefore, antibiotic products for prophylaxis should be administered only to individual animals, as in surgical procedures (European Union 2019). Veterinarians should always be aware that beyond the use of antibiotics in animals, they have an added responsibility for AMR in humans. In this respect, although antibiotic consumption increased in the last few years of this study in parallel with the number of animals treated, first-line treatments should initially be those compounds belonging to group D (Prudence) (European Medicines Agency 2019). These advances in clinical guidelines have important implications in promoting the prudent use of antibiotics, combating global AMR, and protecting both animal and human health. Therefore, veterinarians should be familiar with these guidelines to ensure optimal practices and outcomes.

A scoping review for the ENOVAT revealed differences between countries in terms of antimicrobial use for SAP. Cefazolin was predominantly used in studies from Europe and North America, and cefuroxime was predominantly used in the UK (Sørensen et al. 2024). This study showed that beta-lactam antimicrobials (primarily cephalosporins) are the most commonly used drugs during perioperative surgery, in line with the guidelines described above (The Finnish Food Safety Authority and the Faculty of Veterinary Medicine at the University of Helsinki 2018; Jessen et al. 2019). In our study, these treatment patterns were followed in dogs, in which amoxicillin/clavulanic acid was employed, but not in cat procedures, in which marbofloxacin was the most widely used drug, possibly to prevent surgical infections (Staphylococcus intermedius, Escherichia coli, and Pseudomonas aeruginosa), as indicated in the SPC (AEMPS 2009). Despite being the most widely used antimicrobials in practice, amoxicillin/clavulanic acid is not recommended as a first-line treatment, as clavulanic acid has increased its usage in human beings with multidrug-resistant bacteria (Davies and Davies 2010). Therefore, the use of fluoroquinolones should be restricted for similar reasons (Davies and Davies 2010). Similar to our study, in a Spanish survey on perioperative antimicrobial use, beta-lactamase-resistant penicillins, such as amoxicillin/clavulanic acid, were the most frequently indicated, followed by first-generation cephalosporins, such as cefazolin, and fluoroquinolones, such as marbofloxacin (Otero Balda et al. 2023). Moreover, cefazolin is considered the antibiotic of choice for orthopaedic surgeries because of its rapid distribution in plasma and surgical wound fluid, favourable antimicrobial spectrum, prolonged tissue levels, low toxicity, and cost-effectiveness (Välkki et al. 2020). This is in line with our results, in which cefazolin was used in 76.6% of the orthopaedic surgeries. In addition, according to Danish guidelines, it is the antibiotic of choice against Staphylococci and Pasteurella spp. (Jessen et al. 2019).

Finally, fluid therapy was employed in 3008 surgeries. One of its most common uses is to support the patient during the pre-anaesthetic period, with several benefits such as supporting cardiovascular function, maintaining plasma volume during long anaesthetic periods, and counteracting hypotension (Davis et al. 2013).

Prescription diversity (PD) is a metric that reflects both the number of distinct drug classes prescribed and the uniformity of their use (Singleton et al. 2018). This metric is particularly valuable for analysing antibiotic prescription strategies aimed at curbing the development of resistance (Redding et al. 2019). PD provides insight into whether a hospital adopts a more homogeneous (low PD) or heterogeneous (high PD) prescribing approach and serves as a tool for evaluating shifts in prescribing patterns (Redding et al. 2019). Our antibiotic PD (0.66) was lower than that determined by other authors for dogs (0.92 in 2017, 0.93 in 2018) and cats (0.89 in 2017, 0.88 in 2018) (Schnepf et al. 2021). This implies that our hospital does not use a wide range of antibacterial agents during perioperative surgeries. Many factors may influence prescription diversity, such as market dominance, standard operating procedures, and drugs perceived as leaders in terms of effectiveness, habits, and preferred providers.

We also calculated the administered daily amount (ADA) for all drugs prescribed in the HVULE, which is based on the human defined daily dose (DDD) (WHO Collaborating Centre for Drug Statistics Methodology 2023). This parameter helps to identify treatment patterns, optimise drug use, and reduce the risk of AMR to antimicrobials (Bosman et al. 2019; Lardé et al. 2020). Few studies have been conducted on antimicrobial consumption metrics in companion animals, and different parameters have been proposed and evaluated (Hopman et al. 2019; Schnepf et al. 2021). In the literature, we found only one veterinary teaching hospital in Germany that assessed the same parameters for antimicrobials and for all types of treatments and procedures (not only perioperative ones) (Schnepf et al. 2021). In our study, cefazolin was mostly given to dogs (0.22 kg), but in lower amounts than that determined in Germany for dogs in only one year (0.29 and 0.26 kg in 2017 and 2018, respectively). Marbofloxacin was the most prescribed compound in cats in the HVULE during the five years studied (0.015 kg), whereas in the German hospital similar values were achieved for the same species in only one year (0.01 kg in 2017 and 2018). Notably, the calculated amounts strongly depend on the recommended doses, which are highly variable from one antibiotic to another. Knowledge of current drug use, specifically antimicrobials, is increasingly important in both people and animals, but indicators need to be clarified in companion animals, and methodologies need to be validated to enable comparisons across animal populations within species, between species, and internationally.

A potential limitation of this study is the representativeness of the sample as the data were collected from only one hospital. Therefore, the results may not be generalisable to other clinics or veterinary hospitals. Additionally, the experiences of practitioners and individual patient characteristics should be considered. However, although significant efforts were made to identify demographic characteristics, clinical signs, and prescriptions using manual and semi-automated methodologies, unclear or missing descriptions may have been overlooked by the authors. Additionally, the COVID-19 pandemic occurred during the study period and may have affected hospital visits. However, despite these limitations, our findings increase the knowledge of perioperative drug management in current veterinary practice and may help raise further discussions on the prudent use of several pharmacological groups.

5. Conclusions

This is the first study to describe perioperative drug patterns in companion animals and the variables associated with prescription in a veterinary teaching hospital over a long period. Overall, the data presented herein provide a comprehensive overview of the drugs used perioperatively, including those approved for human use. Dogs were the most frequent animals undergoing surgery, and reproductive and traumatological procedures were the most common surgical interventions. Isoflurane was the most commonly administered drug. Amoxicillin/clavulanic acid (Category C-Caution) and marbofloxacin (Category B-Restrict) were the most common antibiotics in dogs and cats, respectively. Antibiotics were more likely to be used in females and during traumatological surgeries.

Our study contributes to a better understanding of perioperative drug management in surgical veterinary patients, which may be useful for all veterinarians to compare with their clinical practice. The findings of this study provide detailed data for veterinary policymakers and educators to improve rational drug use in surgical procedures. Additionally, it may serve as a basis for standardising and improving drug protocols for different types of surgery. Furthermore, the results obtained regarding the use of antibiotics emphasise the need to promote rational use in this pharmacological group according to the current AMR guidelines, which include the prioritisation of antibiotics with lower restriction categories wherever possible.

Funding Statement

This study received no external funding.

Author contributions

Conceptualisation, B. R. and R. D.; methodology, B. R. and R. D.; formal analysis, J. S.; investigation, J. R. A.; resources, R. P.; data curation, A. M. S. and M. V.; writing-original draft preparation, B. R.; writing-review and editing, R. D., A. M. S., J. M. R., and C. L.; visualisation, J. M. R.; supervision, A. M. S. and R. D.; project administration, C. L. All authors have read and agreed to the published version of the manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).