Abstract
Objectives
This study aimed to describe perianaesthetic complications in cats undergoing MRI of the brain and investigate if previously reported factors for anaesthetic mortality in cats are associated with an increased risk of anaesthetic complications in this population.
Methods
Records were reviewed of all cats that underwent MRI of the brain in a private referral veterinary hospital between June 2020 and January 2023. Cats that underwent additional procedures other than cerebral spinal fluid (CSF) sampling during the same general anaesthetic (GA) were excluded from the data set. Data were then collected regarding the signalment, presenting clinical signs, physical and neurological examination, GA protocol, complications experienced under GA and known survival time. Selected data were then compared between cats with an intracranial lesion and those without.
Results
A total of 43 cats were included in the study. In total, 41 cats experienced a perianaesthetic complication, which were subclassified as mild (n = 39) and severe (n = 15). Cats with intracranial lesions were more likely to experience severe perianaesthetic complications (odds ratio [OR] 12.4; 95% confidence interval [CI] 2.7–56.7; P <0.001), to receive mannitol under GA (OR 1.8; 95% CI 1.2–2.7; P <0.001) and to experience a delayed extubation (OR 1.2; 95% CI 1.0–1.6; P <0.014) than those without intracranial lesions. Cats with intracranial lesions were more likely to have an abnormal neurological examination than those without intracranial lesions (OR 42.1; 95% CI 8.8–228.6; P <0.001).
Conclusions and relevance
Cats that experienced perianaesthetic complications were more likely to have an intracranial lesion. Abnormal mentation on neurological examination was the best indicator of the likelihood of the cat experiencing perianaesthetic complications. Previously reported risk factors for anaesthetic mortality were not found to be associated with increased risk of anaesthetic complications in this population.
Introduction
The diagnosis of intracranial lesions is most frequently achieved using advanced imaging such as MRI and other diagnostic tests, including cerebrospinal fluid (CSF) sampling. MRI in small animal patients requires general anaesthesia (GA) to maintain immobility.
GA is associated with risk of mortality and morbidity, and this has been well established in the literature, with many studies investigating the risk associated with GA in different species and populations.1–9 Cats are reported to experience a greater anaesthetic risk than dogs. This may be due to their small size, higher body surface area, increased difficulty of intravenous (IV) cannulation and endotracheal intubation. 1 Based on the 2007 Confidential Enquiry into Perioperative Small Animal Fatalities (CEPSAF) report, a number of factors have been identified as being associated with greater risk of mortality and morbidity in cats, which include urgency of procedure, increasing American Society of Anaesthesiologists (ASA) grade, endotracheal intubation and administration of intravenous fluid. 1
Cats presenting for MRI of the brain frequently present with abnormal mentation. The risks associated with GA in dogs for brain MRI have been investigated and it has been reported that dogs with intracranial lesions were more likely to experience complications than those without. 7 It was also reported that dogs with abnormal mentation before GA were more likely to experience anaesthetic complications during this procedure. 7
Mortality and perianaesthetic complications experienced by cats undergoing MRI of the brain have not been previously reported. The objectives of this retrospective study were to describe perianaesthetic complications experienced by cats undergoing MRI of the brain and to investigate if previously reported factors for anaesthetic mortality in cats, such as ASA grade, age, weight and neurological examination findings, were associated with increased risk of perianaesthetic complications in this population. We hypothesised that cats with an intracranial lesion would be more likely to experience perianaesthetic complications than those without.
Materials and methods
The study was approved by the RCVS Ethics Review Panel (2023-045). Written consent was granted by the owners for the use of their cats’ patient data through the consent form signed at admission.
Cats undergoing brain MRI between June 2020 and January 2023 in a private referral hospital were identified by reviewing case records. A 30-month period was selected to provide a representative sample of cases seen. Cats that underwent any additional procedure during the same GA, other than CSF sampling, were excluded. Additional exclusion criteria included MRI of additional anatomic regions under the same GA, more than six data points absent from the anaesthetic record, or if the GA record was missing or illegible. Six missing data points was selected to ensure that useful cases were not lost because of unavoidable missing information that occurred during patient transport from the induction suite to MRI scanner. In cases where more than one MRI of the brain was performed during the inclusion period, only the first GA was included for consideration. The anaesthesia and analgesia protocols employed in the anaesthetic management of these cases was at the discretion of the individual supervising anaesthetist.
Data were recorded using Excel (Microsoft). Information collected included the cats’ age, sex, body mass, body condition score (BCS), presenting clinical signs, duration of clinical signs, concurrent illness, history of seizures, any medications, general physical examination, abnormal blood results, neurological examination findings, date and time of MRI, MRI findings, contrast administration, location of any lesions identified, diagnosis, CSF sampling and results, details of anaesthetic protocol, complications experienced under GA and known survival time.
A neurological examination and MRI review were completed by a board-certified neurologist. The radiological diagnosis was determined by a board-certified neurologist.
For the purposes of this study, the following definitions were employed: hypotension: mean arterial pressure (MAP) <60 mmHg; hypercapnia: end tidal carbon dioxide tension (ETCO2) >45 mmHg; hypocapnia: ETCO2 <35 mmHg; hypothermia: temperature <37°C; severe hypothermia: temperature <34°C.8–12 Delayed extubation was defined as extubation taking place more than 20 mins after cessation of the maintenance anaesthetic agent. 7 Mild GA complications were defined as hypotension, hypothermia and hypercapnia. Severe GA complications were defined as one or more of the following: delayed extubation; mannitol administration under GA; or death (including euthanasia) under GA.
Data were analysed using commercially available statistical software (SPSS Statistics version 29; IBM Corp). Differences were investigated between cats with the variable of interest (eg, intracranial lesions or anaesthetic complications) and those without. The Mann–Whitney non-parametric test was used to evaluate continuous data, such as weight, age and duration of GA, and are presented as median (range). Fisher’s exact test was used to compare categorical data (sex, breed; whether a CSF sample was taken; presence or absence of intracranial lesions; presence of hypotension; hypocapnia or hypercapnia; hypothermia or severe hypothermia; occurrence of severe perianaesthetic complications (mannitol under GA, delayed extubation and death under GA); whether a ventilator was used during anaesthesia; and whether a specific drug was part of the anaesthetic protocol) between groups. Non-normally distributed data are presented as median and 95% confidence intervals (CIs). Values of P <0.05 were considered significant. Where appropriate, data are presented with odds ratios (ORs) and 95% CIs for the association of the variable of interest with cats with an intracranial lesion.
Results
A total of 53 cats underwent brain MRI within the study period. Of these, 10 cats were excluded based on the previously established criteria. In total, 43 cats (14 female spayed, one female entire, 27 male castrated and one male entire) were included in data analysis. The median age was 82 months (range 10–204). Breeds included domestic shorthair (DSH; n = 28), Bengal or Bengal crossbreed (n = 6), Persian and Persian crossbreed (n = 2), domestic longhair (n = 2), British Shorthair (n = 2) and one each of Siamese, Norwegian Forest Cat and Russian Blue. The cats had a median body weight of 4.1 kg (range 2.7–6.7) and a median BCS of 4.5 (range 3–7).
The cats presented with a median duration of clinical signs of 21 days (range 2–1460). There were eight cats with a history of seizures. Of the cats imaged, 10 were considered neurologically normal on examination and 33 cats had an abnormal neurological examination. Of those that were neurologically abnormal, findings included cranial nerve deficits (n = 24), abnormal gait/posture (n = 22), altered mentation (n = 16), abnormal proprioception (n = 11), spinal pain (n = 2) and spinal reflex abnormalities (n = 1).
Premedication consisted of combinations of opioids, alpha (α)2-adrenoceptor agonists and/or benzodiazepines at the discretion of the supervising anaesthetist. Opioids administered were either butorphanol (n = 39) or methadone (n = 4). If an α2-adrenoceptor agonist was administered as part of premedication, it was dexmedetomidine (n = 28). A small number of patients received midazolam as part of their premedication (n = 5). The induction protocols used included the use of a single induction agent: propofol (PropoFlo Plus; Abbot, n = 19) and alfaxalone (Alfaxan; Jurox, n = 3) or the use of coinduction agents: propofol (PropoFlo Plus; Abbot) and midazolam (Dormazolam; Dechra, n = 17); alfaxalone (Alfaxan; Jurox) and midazolam (Dormazolam; Dechra, n = 3); or alfaxalone (Alfaxan; Jurox), midazolam (Dormazolam; Dechra) and fentanyl (Fentadon; Dechra, n = 1). Maintenance agents utilised were either isoflurane (n = 5) or sevoflurane (n = 38). The median duration of anaesthesia was 100 mins (range 75–185). Maintenance agents were delivered in oxygen using either a Mapleson D (paediatric T piece) or a circle breathing system, using paediatric tubing. In most instances, the circle breathing system was only used if delivering intermittent positive-pressure ventilation.
Of the 43 cats, 41 experienced complications. Of these, 15 experienced one or more of the following severe complications: requiring mannitol (n = 9); death including euthanasia (n = 6); and delayed extubation (n = 6). In total, 39 cats experienced one or more mild complications: hypothermia (n = 34); hypotension (n = 23); and hypercapnia (n = 11) (Table 1).
Comparison of major and minor general anaesthetic (GA) complications according to the presence or absence of an intracranial lesion
Data are n (%) for the 41 cats that experienced complications
Five cats were euthanased under general anaesthetic, one cat did not survive 48 hours
Diagnoses included neoplastic mass (n = 12), idiopathic epilepsy and other seizure disorders (n = 8), otitis media/interna (n = 5), empyema (n = 4), open investigation (n = 3) and one each of vestibular syndrome, paroxysmal dyskinesia, cortical dysplasia, thiamine deficiency, feline infectious peritonitis, neurodegenerative disease, polycranial neuritis, temporomandibular osteoarthritis, encephalitis, hypertensive encephalopathy and high-grade second-degree atrioventricular node block. Of the 43 cats, 23 were identified as having a structural intracranial lesion.
Cats with severe perianaesthetic complications were more likely to present with a history of seizures (OR 10.2; 95% CI 6.1–56.6; P = 0.01), have abnormal mentation on neurological examination (OR 1.6; 95% CI 0.2–11.6; P = 0.006) or have an intracranial lesion identified on MRI (OR 5.4; 95% CI 0.7–44.7; P <0.001). Of the 20 cats with intracranial lesions, 13 (65%) had suspected neoplasms, five (25%) had infectious causes (eg, empyema) and a further two (10%) had cortical dysplasia. None of the cats that presented with seizures were found to have intracranial structural lesions. Hypotension, hypothermia, hypocapnia and hypercapnia were not associated with concurrent severe perianaesthetic complications. Specific premedication, induction or maintenance agents were not associated with severe GA complications (opioids: OR 0.9; 95% CI 0.2–3.6; P = 0.93; α2-adrenoceptor agonists: OR 1.3; 95% CI 0.2–7.7; P = 0.54; induction agent: OR 0.8; 95% CI 0.1–5.6; P = 0.31; and maintenance agent: OR 1.6; 95% CI 0.3–4.7; P = 0.26).
Cats with intracranial lesions were more likely (P = 0.03) to have a shorter duration of clinical signs (14 days, 95% CI 14–28) than those without an intracranial lesion (28.5 days, 95% CI 10–60). Cats with intracranial lesions were significantly older (111 months, 95% CI 82–132; P = 0.02) than those without intracranial lesions (72 months, 95% CI 60–115). No significant difference between sex (P = 0.58) or neuter status (P = 0.71) was identified for cats with and without intracranial lesions. Cats with intracranial lesions were also more likely to have an abnormal neurological examination than those without intracranial lesions (OR 42.1; 95% CI 8.8–228.6; P <0.001). Cats with intracranial lesions were more likely to experience severe perianaesthetic complications (OR 12.4; 95% CI 2.7–56.7; P <0.001): to receive mannitol under GA (OR 1.8; 95% CI 1.2–2.7; P <0.001); to experience a delayed extubation (OR 1.2; 95% CI 1.0–1.6; P <0.014); or to experience death (including euthanasia) under GA (OR 1.4; 95% CI 1.1–1.9; P <0.01) than those without intracranial lesions. Nine cats were administered mannitol in the current study, all of which had intracranial lesions, and four of these cats survived to discharge.
Discussion
The objective of this study was to describe perianaesthetic complications in a population of cats undergoing brain MRI at a private referral hospital and to investigate if previously reported factors for anaesthetic mortality in cats were associated with increased risk of complications in this population. We hypothesised that cats with an intracranial lesion would be more likely to experience perianaesthetic complications than those without.
The factors identified in the Confidential Enquiry into Perioperative Small Animal Fatalities (CEPSAF) associated with greater anaesthetic risk (higher ASA grade and extremes of body weight) were not found to be associated with a higher risk of mortality or severe perianaesthetic complications in the current study. 1 However, there are key differences in the respective study populations that could account for this variation. Compared with the CEPSAF report, cats in the current study were anaesthetised during normal working hours and the anaesthetic records reviewed did not contain information on the urgency of the procedure. The cats included had a mix of predominantly ASA 2, ASA 3 and ASA 4 grades, and only underwent MRI rather than a range of surgical and diagnostic procedures. Because of the nature of the MRI procedure, and the patients included experiencing a greater anaesthetic risk, the protocols at the study hospital have been developed and refined to allow preanaesthetic stabilisation, prevent significant hypothermia where possible and provide cardiorespiratory support during the procedure. In addition, there are also adequate emergency resources available in this referral hospital setting should complications occur. In addition, CEPSAF grouped ASA grade as ASA I–II, ASA III and ASA IV–V, which this study did not. CEPSAF also reported that the use of endotracheal tubes (ETTs) and perioperative intravenous fluid therapy (IVFT) was associated with increased risk of mortality. However, these factors were not considered in the current study as all cats included received IVFT and tracheal intubation.1,13 Given all this information, it is challenging to directly compare our results to those reported in CEPSAF. 1
A large number (n = 34) of the cats experienced some degree of hypothermia, which has been reported to contribute to prolonged recovery in dogs and is thought to also occur in cats.14–16 This is likely due to the relatively high body surface area-to-volume ratio of cats and the requirement for the MRI scanning room to be maintained at a lower temperature due to the requirements of the MRI machine.13,17,18 In addition, MRI of the brain in cats is a long procedure, with most scans lasting 40–50 mins, and the use of warming devices is not permitted within the scanning room, further contributing to hypothermia in some cases. Hypothermia as a complication of anaesthesia should not be underestimated as it can have many impacts on the patient’s physiology. It can be associated with altered coagulation, altered drug metabolism, increased incidence of postoperative wound infections, arrythmias and may even result in circulatory arrest and death.19–21 Studies on rat hypothermia have also shown an association with hypotension and cerebral hypoperfusion. 22 Hypothermia in the recovery period may also result in shivering and increase the metabolic oxygen demand. If this increased oxygen demand is not met, inadvertent hypoxaemia can occur. 21
In the current study, cats with severe perianaesthetic complications were more likely to present with a history of seizures, have abnormal mentation on neurological examination or have an intracranial lesion identified on MRI. Intracranial lesions or metabolic causes are the most frequent documented causes of seizures in cats. 23 It is of note that none of the cats with an intracranial lesion were documented as having seizures in this study. This is an unusual finding as a previous study has shown that even in cats with seizures and a normal interictal neurological examination, depending on their age, 5.4–23.1% had a structural intracranial lesion identified on MRI. 24 The findings in the current study are potentially due to the small sample size or artificial selection of patient populations during the referral and diagnostic process. For example, it is possible that older cats with suspected structural lesions as a cause of their seizures are euthanased before referral, although further study would be required to evaluate this further. 24
Cats with intracranial lesions were significantly older than those without intracranial lesions, had a shorter duration of clinical signs, an abnormal neurological examination and as part of their neurological examination, an abnormal mentation. The majority of cats with intracranial lesions (n = 13, 65%) had suspected neoplasms. Based on the neuropathology of feline neoplasms, an older presentation would be expected, as would an abnormal neurological examination based on mass effect in the brain secondary to tumour expansion and vasogenic oedema.23,25 Mannitol is often used as a treatment when elevated intracranial pressure (ICP) is suspected secondarily to an expanding mass and associated vasogenic oedema. 7 In the current study, nine cats were administered mannitol, all of which had intracranial structural lesions, with four of these surviving to discharge. Because of the retrospective nature of this study, the precise decision point for the administration of mannitol was not always recorded. Typically, in this institution, mannitol would be administered when the review of the images indicated concern for increased ICP with evidence of concerning brain herniation as deemed by the board-certified neurologist reviewing the images.
Delayed extubation has been previously defined as removal of the endotracheal tube greater than 20 mins after cessation of the maintenance agent. 7 In the current study, cats with an intracranial structural lesion were at increased risk of delayed extubation. Further information on why extubation was delayed was unfortunately not collected. It could be hypothesised that mass effect within the brain may have reduced certain key parameters, including return of spontaneous breathing or return of observable gag reflex. Further studies would be required in this area. In addition, in cats there may be a risk of laryngospasm associated with delayed extubation.11–13 Laryngospasm may result in respiratory obstruction in recovery.
Coinductions are commonly used in the anaesthetic management of patients with suspected increased ICP. The objective of the use of fentanyl or lidocaine as coinduction agents is the avoidance of coughing during intubation of the patient’s trachea, to avoid further associated increases in ICP.26–29 The use of a coinduction may also result in a reduction in dose requirement of the induction agent, which in turn may minimise cardiovascular alterations experienced by the patient immediately after the induction of GA. 30 Although neither the incidence of peri-induction coughing, nor specific cardiovascular parameters in the induction period were specifically recorded, none of the specified induction protocols were found to be associated with perianaesthetic complications (P = 0.31).
There were several limitations to this study largely owing to its retrospective nature. The anaesthetic protocols and management were undertaken by different individuals, although the anaesthetic protocols used did not vary hugely. All anaesthetic protocols were prescribed in accordance with best clinical practice after a comprehensive physical examination by the attending anaesthetist. In all protocols, an opioid, either methadone or butorphanol, formed part of the premedication. When the attending anaesthetist deemed it suitable, an α2-adrenoceptor agonist was also administered. Intervention points based on complications encountered were decided based on the preferences of the supervising anaesthetist. A neurological examination and MRI review were also undertaken by different individuals. An MRI review resulted in a presumptive diagnosis rather than a definitive diagnosis in most cases. A definitive diagnosis would require very invasive procedures or, alternatively, a post-mortem examination. Duration of clinical signs was based on owner reporting and relied on their accurate recounting of clinical signs. As this was a retrospective study, it was not possible to perform a prospective sample size calculation. The small number of cases included may also have implications for statistical analysis and may have introduced some bias. For example, those cats with severe neurological signs or poorly controlled seizures may have been euthanased before referral. Similarly, after consultation with a specialist neurologist, some owners may have elected to euthanase their cat without performing further investigations (GA and MRI).
Similar to the study conducted in dogs, we report that cats that experienced severe complications were more likely to have an intracranial lesion. We also found that complications were more commonly experienced by cats that presented with abnormal mentation. 7
Conclusions
Although previously described in dogs, to our knowledge, this is the first report on the frequency of perianaesthetic complications experienced by cats undergoing brain MRI. Our findings were in keeping with those reported in dogs. Cats that experienced perianaesthetic complications were more likely to have an intracranial lesion. Abnormal mentation on neurological examination was the best indicator of the likelihood of the cat experiencing perianaesthetic complications. We hope our findings will be considered when performing GA in cats with an abnormal neurological examination or altered mentation.
Footnotes
Conflict of interest
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
Linnaeus Veterinary Limited supported the costs of the open access publication charges.
Ethical approval
The work described in this manuscript involved the use of non-experimental (owned or unowned) animals. Established internationally recognised high standards (‘best practice’) of veterinary clinical care for the individual patient were always followed and/or this work involved the use of cadavers. Ethical approval from a committee was therefore not specifically required for publication in JFMS. Although not required, where ethical approval was still obtained, it is stated in the manuscript.
Informed consent
Informed consent (verbal or written) was obtained from the owner or legal custodian of all animal(s) described in this work (experimental or non-experimental animals, including cadavers, tissues and samples) for all procedure(s) undertaken (prospective or retrospective studies). No animals or people are identifiable within this publication, and therefore additional informed consent for publication was not required.
