Factors associated with postoperative complications in healthy horses after general anesthesia for ophthalmic versus non-ophthalmic procedures: 556 cases (2012-2014)

Antimicrobial prophylaxis is not indicated for horses undergoing general anaesthesia for elective orthopaedic MRI

BACKGROUND

Post-anaesthetic fever is a known complication of general anaesthesia, however, its incidence in horses undergoing elective magnetic resonance imaging (MRI) is unknown.

OBJECTIVE

To determine the incidence of post-anaesthetic fever in horses undergoing elective orthopaedic MRI and determine whether prophylactic antimicrobial therapy would be associated with a reduction in the incidence of post-anaesthetic fever. We hypothesised that prophylactic antimicrobials would be associated with a reduction in the incidence of post-anaesthetic fever.

STUDY DESIGN

Retrospective cross-sectional study.

METHODS

This retrospective study included 791 elective orthopaedic MRIs in systemically healthy horses between June 2006 and March 2020 that recovered from general anaesthesia and did not undergo surgery or intensive medical therapy soon after recovery. Potential factors associated with post-anaesthetic fever were evaluated using multivariable logistic regression. Case signalment, travel time, preanaesthetic haematology and fibrinogen abnormalities, use of prophylactic antimicrobials, peri-anaesthetic nonsteroidal anti-inflammatories, anaesthesia time and recovery time were all evaluated for association with post-anaesthetic fever.

RESULTS

Of 791 MRI cases, 44 (5.6%) developed a post-anaesthetic fever. Horses that received prophylactic antimicrobials were [odds ratio (OR) 3.8, 95% confidence interval (CI) 1.98-7.46; p ≤ 0.001] more likely to develop a post-anaesthetic fever than those that did not receive antimicrobials. Young horses (1-4 years of age) were (OR 2.8, 95% CI 1.26-6.17; p = 0.01) more likely to develop fever compared with adult horses (≥5 years of age).

MAIN LIMITATIONS

Limitations of this study pertain to retrospective analysis including nonrandomised case selection and incomplete data records.

CONCLUSIONS

While fever may indicate infection, the majority of early post-anaesthetic fevers resolved before discharge from the hospital with no identified cause. The use of prophylactic antimicrobials to reduce the risk of post-anaesthetic fever for elective MRI is not supported by this study.

Pain scoring systems in hospitalized horses with ocular disease

BACKGROUND

Pain recognition in hospitalized horses is challenging, and the utility of pain scoring systems in horses with ocular disease has not been well-described.

HYPOTHESIS/OBJECTIVES

Evaluate the horse grimace scale (HGS) and behavior pain score (BPS) in hospitalized horses with ocular disease. We hypothesized that HGS and BPS would be associated with different clinical progressions.

ANIMALS

Privately owned horses hospitalized for ocular disease between September 2018 and September 2020.

METHODS

Retrospective observational study. The HGS and BPS were recorded daily throughout hospitalization. Clinical progressions were categorized as: (a) discharge from hospital after medical treatment, (b) ophthalmic surgery (eg, keratectomy, conjunctival flap, amniotic membrane transplantation, corneal transplant), or (c) enucleation. Temporal trends in HGS and BPS were assessed using linear regression. Correlations among slope, intercept, and progression were determined using the Kruskal-Wallis test.

RESULTS

Of 65 horses that met inclusion criteria, 29 (45%) were discharged after exclusively medical management, 28 (43%) underwent ophthalmic surgery, and 8 (12%) underwent enucleation. Two horses (3%) had 2 ophthalmic surgeries performed. The BPS scores at admission were higher in horses that were managed medically than in those that underwent enucleation (P = .01). Horses requiring enucleation had higher increases in HGS (P = .02) and BPS (P = .01) during hospitalization than horses that were medically managed and a higher increase in BPS (P = .04) than horses that required ophthalmic surgery.

CONCLUSIONS AND CLINICAL IMPORTANCE

Pain scoring may represent a useful tool for monitoring progression and response to treatment in hospitalized horses with ocular disease.

Incidence and risk factors for complications associated with equine general anaesthesia for elective magnetic resonance imaging

BACKGROUND

Complications are a practical consideration for elective magnetic resonance imaging (MRI) studies performed under general anaesthesia but relatively little is known about their distribution and risk factors.

OBJECTIVES

To describe the incidence of complications associated with MRI performed under general anaesthesia at a large referral facility and evaluate potential risk factors for these complications.

STUDY DESIGN

Retrospective case-control study.

METHODS

Patient information and details of the MRI procedure were collected retrospectively from medical records of all horses that had undergone an MRI under general anaesthesia at the University of Pennsylvania, New Bolton Center, between September 2005 and April 2012. Complications and categorical variables were examined by chi-squared or Fisher's exact tests as appropriate. A mixed-effects logistic regression approach was used to evaluate associations between explanatory variables and the outcome variable (complications or pyrexia). A univariable screen was used to select variables (likelihood ratio test p < 0.2) for inclusion in the multivariable analysis. Statistical significance was inferred when p ≤ 0.05.

RESULTS

Complications were noted after MRI in 51 (17.4%) of 293 events eligible for inclusion. Complications included pyrexia (n = 35), pneumonia (n = 14), colic (n = 10), facial/nerve paralysis (n = 6), diarrhoea (n = 4), and other (n = 3). The odds of developing a post-anaesthetic complication were significantly decreased in horses that received peri-anaesthetic antimicrobials (OR 0.29, 95% CI 0.14-0.63, p = 0.002). Increased age (OR 0.87, 95% CI, 0.76-0.99, p = 0.03) and peri-anaesthetic antimicrobial administration (OR 0.23, 95% CI 0.08-0.65, p = 0.005) were associated with a decreased odds of developing pyrexia.

MAIN LIMITATIONS

Single centre retrospective design.

CONCLUSIONS

Potential complications including pyrexia, pneumonia and colic should be recognised when pursuing MRI under general anaesthesia. The administration of peri-anaesthetic antimicrobials decreased the odds of a complication and warrants consideration, particularly in horses that might be classified as high risk.

Effect of Volumes of Solutions on Intraocular Pressure During Intravitreal Injection of Low Dose Gentamicin in Horses With Recurrent Uveitis: A Randomized Controlled Study

Horses with recurrent uveitis can be treated by intravitreal injection of low dose gentamicin under sedation and after local anesthetic techniques including the retrobulbar nerve block. Since it is reported that retinal degeneration can be secondary to an acute increase of intraocular pressure (IOP), the current randomized controlled study was carried out in order to investigate the changes in IOP following retrobulbar anesthesia, with two different volumes of local anesthetic (lidocaine) solution (10 and 5 mL), and intravitreal injection of 6 mg gentamicin in two different volumes of solution (0.3 and 0.15 mL). Multivariate regression showed that IOP was significantly lower in the pathologic eye at baseline (estimated effect size -6.1 mmHg; P = .0001) and after sedation (estimated effect size -9.4 mmHg; P < .0001). The IOP was significantly higher after a 10 mL block (P .0002) but this effect was very small (+0.5 mmHg). There was no significant difference in IOP after the injection of 0.15 or 0.3 mL solution of gentamicin. There was no significant combined effect of the volume of local anesthetic used for the block and the volume of the gentamicin solution. Since the expected anesthetic effects (desensitization and akinesia) were met in all cases, the smaller volume of 5 mL of lidocaine solution would be preferable for retrobulbar block for intravitreal injections, while intravitreal injections volumes of 0.15 and 0.30 mL can be used indifferently.

Comparison of image quality of corneal and retinal optical coherence tomography using sedation and general anesthesia protocols with or without retrobulbar anesthesia in horses

OBJECTIVE

To compare image quality and acquisition time of corneal and retinal spectral domain optical coherence tomography (SD-OCT) under 3 different sedation-anesthesia conditions in horses.

ANIMALS

6 middle-aged geldings free of ocular disease.

PROCEDURES

1 randomly selected eye of each horse was evaluated via SD-OCT under the following 3 conditions: standing sedation without retrobulbar anesthetic block (RB), standing sedation with RB, and general anesthesia with RB. Five regions of interest were evaluated in the cornea (axial and 12, 3, 6, and 9 o'clock positions) and fundus (optic nerve head). Three diagnostic scans of predetermined quality were obtained per anatomical region. Image acquisition times and total scans per site were recorded. Corneal and retinal SD-OCT image quality was graded on a subjective scale from 0 (nondiagnostic) to 4 (excellent).

RESULTS

Mean values for the standing sedation without RB, standing sedation with RB, and general anesthesia conditions were 24, 23, and 17, respectively, for total cornea scan attempts; 23, 19, and 19 for total retina-scan attempts; 14.6, 13.2, and 9.2 minutes for total cornea scan time; 19.1, 9.2, and 13.0 for total retina scan time; 2.0, 2.3, and 2.5 for cornea grade; and 2.7, 2.9, and 2.5 for retina grade.

CONCLUSIONS AND CLINICAL RELEVANCE

The RB facilitated globe akinesia and improved the percentage of scans in frame and region of interest accuracy for retinal imaging via OCT in horses. Retrobulbar blocks improved clinical image acquisition while minimizing motion artifact.

Demographic, preoperative and anaesthesia-related risk factors for unsatisfactory recovery quality in horses undergoing emergency abdominal surgery

OBJECTIVE

To determine demographic, preoperative and anaesthesia-related variables that may be associated with unsatisfactory recovery quality in horses undergoing emergency abdominal surgery (colic) in an equine teaching hospital.

STUDY DESIGN

Retrospective case series.

ANIMALS

A total of 313 horses.

METHODS

The anaesthetic records of horses admitted for surgical treatment of colic between 2005 and 2018 were examined. Overall quality of recovery was assessed as dangerous, poor, fair, good or excellent. The following categories were constructed as a dichotomic variable: unsatisfactory recovery (poor and dangerous recoveries) and satisfactory recovery (excellent, good and fair recoveries). Univariable and multivariable analyses were performed to evaluate the association between all studied variables and recovery.

RESULTS

All recoveries were unassisted. Unsatisfactory recovery quality totalled 17.2% (3.5% and 13.7% were dangerous and poor recoveries, respectively), whereas satisfactory recoveries totalled 82.8% (26.2%, 40.9% and 15.7% were fair, good and excellent recoveries, respectively). Univariable analysis showed that unsatisfactory recoveries were associated with high preoperative packed cell volume, pain behaviour, poor premedication and induction quality, high intraoperative mean heart rate, low mean arterial blood pressure, dobutamine dose ≥1.5 μg kg^-1 minute^-1, non-administration of romifidine, long anaesthesia time and prolonged time to stand. The multivariable model showed that factors strongly associated with unsatisfactory recovery quality were dobutamine dose ≥1.5 μg kg^-1 minute^-1 [adjusted odds ratio (AOR) = 6.60; 95% confidence interval (CI), 2.91-14.96], poor premedication quality (AOR=4.60; 95% CI, 1.73-12.23) and a time to stand > 70 minutes (AOR=2.59; 95% CI, 1.13-5.91).

CONCLUSIONS AND CLINICAL RELEVANCE

Our study shows that high dobutamine requirements, poor premedication quality and a prolonged time to stand are risk factors for unsatisfactory recovery quality in horses undergoing anaesthesia for colic surgery. Addressing these factors may enable clinicians to improve the quality of recovery phase.

Recovery of horses from general anaesthesia: A systematic review (2000-2020) of risk factors and influence of interventions during the recovery period

BACKGROUND

In equine anaesthesia, the recovery period is a time of considerable risk and has been the focus of prolific research. Risk factors, including age, type and duration of procedure or temperament may influence recovery quality. Unfortunately, the anaesthetist is unable to control for these factors, therefore various pharmacological interventions and recovery methods have been developed with the objective of improving recovery quality. However, no consensus among anaesthetists has been reached for many of these interventions and their implications for recovery-related mortality and morbidity.

OBJECTIVES

To conduct a systematic review of the published evidence relating to risk factors and interventions in the recovery period which may influence recovery quality from equine general anaesthesia (GA).

STUDY DESIGN

A systematic evaluation of the equine veterinary literature using the GRADE framework.

METHODS

A literature search was performed and studies were assessed for eligibility by both authors utilising PRISMA guidelines. Studies meeting inclusion criteria were evaluated by both authors, categorically summarised and the quality of evidence for each sub-topic was assessed using the GRADE framework.

RESULTS

Thirty-nine studies were identified which directly assessed the impact of risk factors and recovery interventions on recovery quality after equine GA. There was evidence to support that peri-anaesthetic risk factors such as anaesthesia duration, American Society of Anesthesiologists (ASA) physical status and surgical procedure influenced recovery quality. We also identified sufficient evidence that administration of α-2 adrenoreceptor agonists immediately prior to recovery, improves recovery quality.

MAIN LIMITATIONS

The validity of the results of some studies may have been compromised by missing data and small sample sizes.

CONCLUSIONS

Recovery quality is influenced by factors including: anaesthesia duration, ASA physical status and surgical procedure. Recovery quality can be improved by the administration of an α-2 adrenoreceptor agonist immediately prior to recovery.

Retrobulbar lidocaine injection via the supraorbital fossa is safe in adult horses but produces regionally variable periocular anaesthesia

BACKGROUND

Injection techniques for retrobulbar anaesthesia are published in horses, but neither safety nor anaesthetic efficacy and duration have been evaluated objectively in vivo.

OBJECTIVE

To characterise the safety and efficacy of one published technique for retrobulbar anaesthesia.

STUDY DESIGN

Randomised, controlled descriptive experiment.

METHODS

Unilateral retrobulbar injection with 10 mL lidocaine (2%) was performed in eight sedated adult mares. Contralateral eyes served as untreated controls. Neurophthalmic parameters, intraocular pressure (IOP), and corneal and periocular sensation were measured awake, post-sedation and at periodic time points for 24 hours following injection. Adverse effects were documented.

RESULTS

Injection of 10 mL lidocaine significantly increased IOP for up to 2 hours (P < .05) maximally at 30 min (mean [95% CI]: 6.0 [2.7, 9.2] mm Hg, P < .001). Six of the eight treated eyes developed mild to moderate reversible chemosis for 2 to 24 hours. One eye developed severe chemosis and superficial corneal ulceration at 24 and 48 hours following injection respectively. Corneal sensitivity significantly decreased for 6 hours (P < .05), maximally at 10 min (-44.4 [-34.6, -54.1] mm, P < .001). Periocular sensitivity (measured as increase in applied force) significantly decreased dorsally and medially for up to 2 hours (maximal at 2 hours (367.1 [238.5, 495.7] g, P < .001, and at 30 min: 345.8 [202.6, 488.9] g, P < .001) respectively). Ventral and lateral sensitivity were not effectively decreased beyond 30 min. Optic nerve function was not consistently reduced following injection.

MAIN LIMITATIONS

Investigators were not masked to the treated eye.

CONCLUSIONS

Retrobulbar injection using 10 mL lidocaine is safe in normal eyes of adult horses, but carries risk in structurally compromised or glaucomatous eyes due to transient IOP increase. Reversible chemosis commonly develops 2-4h following injection, and may be severe in some horses with risk for corneal ulceration. Corneal anaesthesia is rapid and prolonged, but all periocular regions are not consistently anaesthetised. Retrobulbar injection should be combined with other local anaesthetic injections for eyelid surgeries or enucleations.

Recovery after General Anaesthesia in Adult Horses: A Structured Summary of the Literature

Simple Summary

Recovery is the most dangerous phase of general anaesthesia in horses. Numerous publications have reported about this phase, but structured reviews that try to reduce the risk of bias of narrative reviews/expert opinions, focussing on the topic are missing. Therefore, the aim of the present article was to publish the first structured review as a summary of the literature focussing on the recovery phase after general anaesthesia in horses. The objective was to summarise the available literature, taking into account the scientific evidence of the individual studies. A structured approach was followed with two experts in the field independently deciding on article inclusion and its level of scientific evidence. A total number of 444 articles, sorted by topics and classified based on their levels of evidence, were finally included into the present summary. The most important findings were summarised and discussed. The present structured review can be used as a compilation of the publications that, to date, focus on the recovery phase after general anaesthesia in adult horses. This type of review tries to minimise the risk of bias inherent to narrative reviews/expert opinions.

Abstract

Recovery remains the most dangerous phase of general anaesthesia in horses. The objective of this publication was to perform a structured literature review including levels of evidence (LoE) of each study with the keywords “recovery anaesthesia horse”, entered at once, in the search browsers PubMed and Web of Science. The two authors independently evaluated each candidate article. A final list with 444 articles was obtained on 5 April 2021, classified as: 41 “narrative reviews/expert opinions”, 16 “retrospective outcome studies”, 5 “surveys”, 59 “premedication/sedation and induction drugs”, 27 “maintenance with inhalant agents”, 55 “maintenance with total intravenous anaesthesia (TIVA)”, 3 “TIVA versus inhalants”, 56 “maintenance with partial intravenous anaesthesia (PIVA)”, 27 “other drugs used during maintenance”, 18 “drugs before/during recovery”, 18 “recovery systems”, 21 “respiratory system in recovery”, 41 “other factors”, 51 “case series/reports” and 6 “systems to score recoveries”. Of them, 167 were LoE 1, 36 LoE 2, 33 LoE 3, 110 LoE 4, 90 LoE 5 and 8 could not be classified based on the available abstract. This review can be used as an up-to-date compilation of the literature about recovery after general anaesthesia in adult horses that tried to minimise the bias inherent to narrative reviews.

A retrospective study of fecal output and postprocedure colic in 246 horses undergoing standing sedation with detomidine, or general anesthesia with or without detomidine

OBJECTIVE

To determine time to first passage of feces, total fecal piles and incidence of colic in the first 24 hours postprocedure in horses undergoing standing sedation with detomidine, or general anesthesia with or without detomidine.

STUDY DESIGN

Retrospective cohort study.

ANIMALS

A total of 246 horses.

METHODS

Records of all horses that underwent standing sedation or general anesthesia between December 2012 and March 2016 were reviewed. Horses aged <6 months, admitted for colic or cesarean section, with inadequate data, and those not administered xylazine and/or detomidine were excluded. Records included patient signalment, fasting duration, procedure performed, drugs administered, time to first feces, number of fecal piles during 24 hours postprocedure and mention of colic. Chi-square, Fisher's exact and Tukey's post hoc comparison tests were used. Parametric data were reported as mean ± standard deviation with significance defined as p <0.05.

RESULTS

In total, 116 and 57 horses underwent general anesthesia without detomidine (group GA) and with detomidine (group GA-D), respectively, and remaining 73 horses underwent standing sedation with detomidine (group S-D). Detomidine dose was significantly higher in group S-D than in group GA-D. Time to first feces was longer (7.1 ± 4.2 hours), and group S-D horses passed one fewer fecal pile (6.3 ± 2.4) than group GA horses. There was no interaction between detomidine treatment and preprocedure food withholding and the time to first feces or the number of fecal piles in the first 24 hours postprocedure. Overall, seven horses (2.8%) showed signs of colic (five, one and one in GA, GA-D and S-D, respectively).

CONCLUSIONS AND CLINICAL RELEVANCE

Detomidine administration, as part of an anesthetic protocol or for standing sedation procedures, should not be expected to contribute to postprocedural colic.