In patients with neurological disease a multidisciplinary approach is key to preoperative planning and perioperative care, and this should always include the treating neurology team.
A standardised test of mental function in patients with possible or confirmed cognitive impairment or dementia can help to establish a preoperative baseline and to assess the risk for post-operative delirium and cognitive dysfunction.
Following a stroke, non-urgent surgery should be postponed for at least 6 months, as in the first 6 months the risk of a perioperative stroke is significantly increased. For urgent surgery this risk needs to be taken into account.
Patients with a carotid stenosis >70 per cent and recent associated symptoms should be considered for a carotid endarterectomy (CEA) or stenting. Apart from this, no general recommendations for carotid screening and revascularisation prior to major surgery can be made. Multidisciplinary decisions about individual cases are more appropriate.
Heparin bridging does not reduce stroke rate and increases bleeding risk when warfarin is temporarily withheld perioperatively in patients with atrial fibrillation (AF) or simple cardiac valve disease.
Best possible maintenance or parenteral conversion of regular drug treatment is crucial to prevent symptoms and severe complications in Parkinson’s disease, epilepsy and myasthenia gravis, and in patients on baclofen for spasticity.
In multiple sclerosis small rises of body temperature, even of only 0.5 degrees Celsius may lead to deterioration in nerve conduction, causing paresis hypoventilation and poor cough, hypaesthesia or visual symptoms.
In patients with chronic high spinal injuries (above T6) precipitating stimuli that trigger life-threatening sympathetic hyperreflexia need to be ascertained and avoided. Although these patients may not feel pain below the sensory spinal level, they should be given regional and/or general anaesthesia for surgery in this area to prevent a sympathetic reflex response to surgery.
After major surgery, patients with significant neurological disease should be cared for in a high-dependency or intensive care unit.
preoperative Assessment and Preparation of Patients with Neurological Disease
Neurological diseases are relatively common and frequently impact patients’ physical, cognitive, emotional and psychosocial functioning. The emphasis of perioperative planning and management is to support vital functions impaired by the neurological disease and to minimise the risk of perioperative deterioration of the neurological condition. This risk must be evaluated and discussed with the patient and family or carers.
Patients may express views regarding acceptable quality of life and limitations of treatment and resuscitation. Elderly patients with pre-existing dementia or cerebrovascular and/or other cerebral and general diseases are at risk of post-operative delirium. This warrants discussion with patients and carers as well as post-operative resource planning. Patients with cognitive impairment should have their mental capacity assessed for informed consent prior to surgery. They may need a legal representative with power of attorney to act on their behalf, or the involvement of an independent mental capacity advocate (IMCA).
It is crucial in the preoperative setting to assess the impact of neurological disease on the patient’s mobility, functional status and dependency, and the potential requirement for temporary or permanent care.
With these patients, a multidisciplinary approach is very helpful, and needs to include surgery, anaesthesia, neurology, possibly medicine for the elderly, psychiatry, physiotherapy and other medical or allied health professionals’ support as indicated in each case.
Cerebrovascular disease (CVD) includes ischaemic and haemorrhagic stroke (the term ‘cerebrovascular accident’ is no longer used), as well as atherosclerotic narrowing of major and minor cerebral arteries. CVD is common worldwide, particularly in patients with suboptimally treated cardiovascular conditions, such as hypertension (Hudson and Greene, 2015; Probasco et al., 2013).
Perioperative stroke is defined as stroke during or up to 30 days after surgery (Sacco et al., 2013). Recent American data suggest that it occurs in about 0.1 per cent of patients undergoing non-cardiac and non-neurological surgery and is associated with an eightfold increase in 30-day-mortality (Mashour, Shanks and Kheterpal, 2011). It may also lead to potentially devastating neurological disability. Cardiac surgery is associated with a higher incidence of perioperative stroke, between 1 per cent and 5 per cent overall, and this figure is higher for complex valve surgery (Bucerius et al., 2003; Li et al., 2009). Major vascular, brain and emergency surgery also carry a higher risk of perioperative stroke. Factors that put patients at higher risk for perioperative stroke include advanced age, female sex, previous stroke (particularly if recent), cardiac disease such as atrial fibrillation (AF) or poor ventricular function and vascular disease (cerebral and peripheral). Other risk factors include diabetes, chronic lung and kidney disease, infection and raised white blood cell count as well as perioperative discontinuation of anticoagulation (Bucerius et al., 2003; Jorgensen et al., 2014; Mashour et al., 2011).
A recent large observational study showed that the incidence of post-operative stroke is closely related to the time passed since a previous stroke (Jorgensen et al., 2014). The adjusted odds ratio for post-operative stroke was 16 for any prior stroke, and during the first 3 months after a stroke this ratio was as high as 68. In this study of 862 patients who suffered stroke in the first 3 months prior to non-cardiac surgery, 103 developed another stroke post-operatively. For patients who suffered stroke 3 to 6 months before their surgery, risk was increased by a factor of 24, and more than 12 months prior was associated with a 10- to eightfold increase in risk. This suggests that any non-urgent surgery should be postponed for at least 6 months after a stroke (Hudson and Greene, 2015; Sanders, Jorgensen and Mashour, 2015).
However, patients who have suffered a recent stroke or transient ischaemic attack (TIA) attributable to an ipsilateral carotid stenosis >70 per cent (but not total or near occlusion) are an exception. They are likely to benefit from urgent carotid revascularisation, and currently carotid endarterectomy (CEA) is considered the safest option (Bonati et al., 2012). Ideally, patients should have CEA within 2 weeks of experiencing symptoms to prevent stroke. By 1 month, the benefit may be marginal and by 3 to 6 months, patients may be considered ‘asymptomatic’. Patients with a >50 per cent but <70 per cent stenosis may benefit from a CEA within 2 weeks too. A simple rule of thumb is that a patient with a symptomatic carotid stenosis >70 per cent within the past month should be considered for a revascularisation procedure to prevent stroke. Other scenarios can be discussed within a neurovascular multidisciplinary team (MDT) (Brott et al., 2011).
The risk of perioperative stroke from untreated carotid stenosis in patients undergoing unrelated surgery varies depending on the type of surgery, the degree of uni- or bilateral carotid disease and whether this has led to stroke or TIA, as well as other acute or chronic co-morbidities. These factors increase the general stroke risk, not necessarily in the territory of the affected carotid (Li et al., 2009).
The potential risk reduction from CEA or carotid artery stenting (CAS) has to be weighed against the risk of stroke associated with carotid intervention. In the latest meta-analysis this was 4.5 per cent for CEA and 8 per cent for CAS (Bonati et al., 2012). Thus, general stroke risk factors need to be assessed and optimally managed and patients need to be placed on medical treatment. Only severe symptomatic carotid stenosis should be considered for CEA or CAS prior to other major surgery. Evidence guiding other clinical scenarios is less compelling and they can be treated on their merits in collaboration with a vascular MDT.
Currently, there is no evidence for benefit of screening asymptomatic patients for carotid artery disease and possible intervention prior to major non-cardiac surgery (Kristensen et al., 2014).
In elective cardiac surgery, it is recommended that only patients with symptomatic unilateral or asymptomatic bilateral high-grade carotid artery stenosis should be considered for CEA or CAS. The 2014 European guidelines for coronary revascularisation (Authors/Task Force et al., 2014) emphasise the complexity of decision-making in patients with significant carotid and coronary disease. They suggest that only patients with a symptomatic (previous TIA or stroke) and significant carotid stenosis (>50 per cent in men and >70 per cent in women) may experience a reduction of stroke and mortality after coronary artery bypass grafts (CABG) if a carotid revascularisation is performed prior to CABG. Involvement in asymptomatic carotid artery stenosis should only be considered prior to CABG in men with bilateral severe carotid stenosis or contralateral occlusion if the stroke risk from the carotid intervention can confidently be predicted as less than 3 per cent and the patient’s life expectancy is more than 5 years. Patients should be assessed on an individual basis by a multidisciplinary team which includes a neurologist.
In patients with significant carotid stenosis, the individual risk of perioperative stroke, particularly in cardiac and major vascular surgery, is also related to the adequacy of collateral blood flow. This can be assessed by transcranial Doppler (TCD) ultrasound examination of both middle cerebral arteries. This test may help to assess significant stroke risk in the context of the planned surgery and the individual benefit of carotid revascularisation. The result may guide the choice of surgical and anaesthetic techniques to help minimise the risk of stroke (personal communication, Prof R. W. M. Keunen, Haga Hospital, The Netherlands, at EuroNeuro congress, Barcelona, April 2016).
Perioperative ischaemic stroke may occur after episodes of hypotension and hypoperfusion. Hence, stroke prevention measures include maintenance of blood pressure within the patient’s normal range and avoidance of perioperative hypotension. Factors that can alter cerebral autoregulation, such as chronic hypertension and brain injury, make this management even more crucial.
However, the most common cause of perioperative ischaemic stroke is thromboembolism, most commonly due to atrial fibrillation (AF). Cardiac thrombus may also form secondary to almost any cardiac abnormality, but especially with myocardial infarction, ventricular aneurysm and heart failure. Heart valve replacements, particularly mechanical mitral and aortic valves, also increase this risk. Paradoxical venous embolus may complicate atrial septal defect and patent foramen ovale, present in up to a quarter of the population.
In patients with AF, the risk of thromboembolic stroke is higher with coexisting cardiac failure, hypertension, age above 75, diabetes and previous stroke.
|Co-morbid conditions||CHA2DS2-VASc score|
|Congestive heart failure||1|
|Age ≥75 years||2|
|Sex category (i.e. female sex)||1|
Legend: CHA2DS2-VASc, C=congestive heart failure, H=hypertension, A =age ≥75 years, D=diabetes mellitus, S=stroke/TIA history, V=vascular disease, S=sex category (i.e. female sex), TIA=transient ischaemic attack
The CHA2DS2-VASc score might also help to identify patients at higher risk of perioperative stroke who might benefit from minimisation of the time without anticoagulation in the perioperative period. This decision has to be carefully weighed against the bleeding risk in relation to patient and surgical factors. The following long-term risk factors for increased bleeding in patients anticoagulated for AF have been identified: hypertension, abnormal kidney and liver function, stroke, bleeding history or predisposition, labile international normalised ratio (INR), the elderly and alcohol and drugs such as acetyl salicylic acid (‘HAS-BLED’), as well as diabetes and heart failure (Lip et al., 2011).
However, starting new beta-blockers shortly before non-cardiac surgery cannot be recommended. The POISE trial, which included more than 8000 patients with cardiac risk factors scheduled for non-cardiac surgery but not on beta-blockers, showed that starting of perioperative beta-blockade with metoprolol was associated with a doubling of post-operative stroke from 0.5 per cent to 1 per cent. There was also an overall increase in 30-day mortality from 2.3 per cent to 3.1 per cent despite a significant reduction in post-operative myocardial infarction (4.2% vs 5.7%) (Devereaux et al., 2008).
Current evidence suggests that perioperative continuation of statins is associated with a reduction in cardiovascular morbidity, including AF, and they should be continued perioperatively. Additionally, there may be benefit in starting statins prior to cardiac and vascular surgery, ideally at least 2 weeks pre-operatively (de Waal, Buise and Van Zundert, 2015; Kristensen et al., 2014).
Antiplatelet Agents and Anticoagulation in Patients at Risk of Thromboembolic Stroke
Patients at very high risk of stroke, such as in CEA surgery in patients with carotid stenosis, may benefit from aspirin and possibly another antiplatelet agent like clopidogrel being continued until the day before surgery. On the day of surgery no antiplatelet medication should be given, so that in the event of severe bleeding, transfused platelets would not be inhibited by the plasma activity of these drugs. The antiplatelet effect of these drugs on the patient’s own platelets will persist.
The POISE 2 trial showed that continuing perioperative and new administration of aspirin in patients with cardiovascular risk factors undergoing non-cardiac and non-carotid surgery was not associated with an overall reduction in cardiovascular morbidity or mortality, but caused a higher risk of bleeding. The patients enrolled in this trial were subdivided into strata that had been on long-standing aspirin before and those who had not. They were then randomised to receive aspirin 200 mg/day just before surgery and 100 mg/day or placebo for 30 days post-operatively. In the subgroup not on aspirin before, there was a reduction in perioperative stroke with aspirin (Devereaux et al., 2014). This would suggest that for patients at a high risk of ischaemic stroke, aspirin therapy, usually at the lower daily dose of 75 mg/day, should be continued until the day before surgery and recommenced as early as possible post-operatively.
Full anticoagulation with warfarin may not need to be interrupted for minor dental surgery, cataract extraction and endoscopy. There is some evidence that warfarin at a dose adjusted to an INR of 1.5–2 in patients at very high risk of embolisation can be continued even for major surgery, such as joint replacement, with a 2–10 per cent risk of significant bleeding (Larson, Zumberg and Kitchens, 2005). Generally, however, to reduce the risk of bleeding from major elective surgery, warfarin is stopped 5 days preoperatively and the INR checked before surgery to ensure normal clotting (Baron, Kamath and McBane, 2013).
When assessing whether ‘bridging’ with heparin should be provided during the time off warfarin, the risk of perioperative stroke compared to the risk of bleeding has to be considered on a case-to-case basis. Current evidence suggests that in patients with AF, bridging may not change a relatively low incidence of perioperative stroke, but increase the risk of bleeding. The randomised, double-blind BRIDGE trial was conducted in 108 centres in the United States and Canada between 2009 and 2014. It included 1884 patients with chronic or paroxysmal AF or atrial flutter with and without heart valve disease and other significant stroke risk factors. It showed no difference in occurrence of stroke, TIA and systemic arterial embolism after elective surgery with low molecular weight heparin (LMWH) bridging (0.3%) versus placebo (0.4%). Major post-operative bleeding occurred in 3.2 per cent of the patients who received LMWH bridging and 1.3 per cent of the placebo group. It is important to note that patients with mechanical heart valves and/or stroke or TIA in the 12 weeks prior to surgery were excluded from this trial (Douketis et al., 2015). A systematic review of observational trials which included 12,278 patients with AF or mechanical heart valves also failed to show any difference in stroke or systemic arterial embolism, but a higher rate of bleeding with heparin bridging (Siegal et al., 2012). Based on this evidence, heparin bridging should probably only be considered for patients and/or operations with a significantly increased risk of thromboembolic stroke or systemic embolism, such as patients with AF and a high CHA2DS2-VASc score and/or heart valve disease or replacement(s). The higher risk of perioperative bleeding in these patients should be taken into account, including when considering neuraxial blockade, and discussed with surgeon and patient.
subcutaneous LMWH at therapeutic dose, such as enoxaparin 1 mg/kg or dalteparin 100 IU/kg twice daily, from 3 to 4 days until 12–24 hours before surgery;
intravenous unfractionated heparin titrated to activated prothrombin time ratio (APTTR) of 2–3 until 6 hours before surgery. This option is safer for patients with a creatinine clearance below 30 mL/min or where there is a higher risk of bleeding due to other patient or surgical factors.
Once haemostasis has been secured, LMWH or unfractionated heparin may be restarted 12–24 hours and warfarin 12–72 hours post-operatively depending on the bleeding risk associated with operative and patient factors. Once the normal target INR is reached, heparin should be stopped.
The new direct oral anticoagulants (DOAC), such as the factor Xa inhibitors rivaroxaban and apixaban and thrombin inhibitor dabigatran, are now recommended for stroke prevention in non-valvular AF. Trials and meta-analyses have shown them to be associated with a lower risk of intracranial haemorrhage than vitamin K antagonists (Chatterjee et al., 2013). An added advantage of these agents is a more rapid onset and offset (12–24 hours) than warfarin.
DOACs should be stopped 24 hours before elective surgery with minor bleeding risk and 48 hours before major operations with higher bleeding risk (Heidbuchel et al., 2013; Lai et al., 2014) A longer discontinuation time for these drugs, particularly dabigatran, may be required in patients with renal dysfunction. Perioperative heparin bridging is not indicated in patients on DOACs unless there is a high stroke risk and unpredictable delay in surgery or gastrointestinal pathology, making their bioavailability unreliable. Dabigatran has an inherently low bioavailability, which is decreased further by gastrointestinal pathology and/or surgery and proton pump inhibitors. For this reason, use of parenteral heparin must be considered in any patient at high risk of arterial thromboembolism who cannot take or absorb oral dabigatran. Treatment with DOAC is contraindicated with a creatinine clearance below 30 mL/min and dialysis.
Prior to emergency surgery, the time of last DOAC ingestion and the patient’s renal function must be ascertained. Four factor concentrate and activated prothrombin complex concentrate (50 units/kg) can help reverse their anticoagulant effect. However, the activated concentrate carries a significant pro-thrombotic risk. Haematology advice is often very helpful in these circumstances. Dabigatran can be removed by haemodialysis and haemofiltration.
Antibody fragments which bind and reverse DOACs are becoming available, such as idarucizumab for dabigatran, and may change the management of bleeding and emergency surgery in patients on these drugs.
Dementias are a group of progressive, degenerative, cognitive disorders which are becoming increasingly common as the population ages. Cognitive impairment can be considered as a spectrum from minor changes in cognitive performance, widespread in the older population, to frank dementia. Cognitive impairment and dementia are frequently under-diagnosed in the older population undergoing surgery, and the risks and implications of anaesthesia and surgery in these patients are not always fully appreciated in the preoperative setting.
Cognitive impairment and dementia are closely associated with frailty and the presence of multi-organ co-morbidity. Dementia is strongly predictive of perioperative complications (Hudson and Greene, 2015; Probasco et al., 2013).
In recent years several studies and reviews have implied that anaesthesia and/or surgery might be triggers or risk factors for dementia. However, no consensus has been reached with regard to the risk of post-operative cognitive dysfunction and the type of anaesthesia or anaesthetic technique (Bilotta, Qeva and Matot, 2016; Docherty and Shenkin, 2016).
Definition and Diagnosis of Dementia
Dementia is defined as evidence of significant cognitive decline from a previous level of performance in one or more cognitive domains: learning and memory, language, complex attention, perceptual motor function or social cognition. Cognitive deficits interfere with independence. These patients require assistance with daily activities such as managing their medication and paying bills. The deficits should not occur exclusively in the context of delirium and should not be part of another mental disorder such as depression or schizophrenia.
The major dementia syndromes are:
Alzheimer’s disease (AD): is the most common type of dementia, constituting up to 70 per cent of all cases. One of the hallmark pathologic abnormalities in this disease are deposits of the protein fragment beta amyloid (plaques) and intracellular tangles of the protein. Early symptoms are short-term memory impairment, which then tends to gradually progress to disorientation, confusion, impaired judgement, behaviour changes, apathy, depression, difficulty speaking, swallowing and walking.
Vascular dementia (previously known as multi-infarct or post-stroke dementia): this type includes 20–30 per cent of all dementias with initial symptoms such as impaired judgement and inability to plan steps needed to complete a task.
Dementia with Lewy bodies (abnormal aggregations of the protein alpha-synuclein in the cortex): patients with this disorder tend to experience memory loss and decision deficits common in AD, but some of the early symptoms also include sleep disturbances, hallucinations, muscle rigidity or other Parkinsonian features.
Fronto-temporal dementia (FTD): includes dementias such as behavioural variant FTD, primary progressive aphasia, Picks disease and progressive supranuclear palsy. Typical symptoms include changes in personality and behaviour with difficulty in using language.
Other causes include alcoholism, vitamin deficiencies, HIV infection, Huntington’s disease, chronic traumatic brain injury and prion disease.
More than one type of dementia may coexist in a single patient (e.g. Alzheimer’s disease and vascular dementia).
Apart from patients with dementia (known or undiagnosed), there are many patients who are at risk of post-operative cognitive problems even though they may exhibit only minor symptoms in everyday life. Syndromes recognised relatively commonly in the post-operative period are post-operative cognitive dysfunction (POCD) and delirium. POCD is one of the most well-recognised neuropsychological consequences of anaesthesia. It is a mild cognitive disorder separate from delirium and dementia, and characterised by subtle impairments in memory, concentration and information processing. There are no formal diagnostic criteria to describe this diagnosis yet. Although usually transient, some patients may experience long-term dysfunction, which may become permanent. POCD is particularly prevalent in patients over 60. Moreover, patients who develop POCD or post-operative delirium may be at higher risk for cognitive decline/dementia later in life.
Delirium is an acute state of mental confusion that occurs in the context of an inter-current illness. It can affect up to 15 per cent of patients admitted to hospital, particularly the elderly. Patients are disorientated and often unsure of what is going on with and around them. They can find it difficult to follow instructions and may be frightened, anxious, irritable, restless and agitated. The confusion often varies over time and is usually worse at night. Treatment is aimed at the underlying cause. Patients with an underlying dementia syndrome are likely to suffer a more severe or prolonged delirium syndrome and there is an increased risk of progressing to dementia following an episode of delirium. Patients exhibiting POCD or post-operative delirium have a significantly higher mortality than control patients. When optimising patients for surgery, assessment and planning strategies for prevention of POCD or delirium need to be part of the pre-assessment preparation (American Geriatrics Society Expert Panel on Postoperative Delirium in Older Adults, 2015; Guenther, Riedel and Radtke, 2016).
Risk Factors and Pathogenesis
The neuropsychological pathogenesis leading to POCD following anaesthesia, surgery and critical illness is not fully understood, but likely to be multifactorial. The risk factors are similar to those for post-operative delirium and summarised in Table 9.2.
Anaesthetic Pre-assessment Evaluation of Cognitive Impairment
Post-operative cognitive complications are common but difficult to predict. It is worthwhile using patient evaluation tools in the pre-assessment clinic to provide a baseline cognitive status for comparison in the post-operative period and beyond. This baseline is invaluable for recognising post-operative delirium and cognitive deterioration and for post-surgical discharge planning. It is often beneficial to assess the patient with a relative or carer present to act as a collateral.
Common symptoms and signs to look for include:
Impaired short-term memory
Failure to manage complex tasks (e.g. household, finances)
Loss of orientation or spatial abilities (e.g. getting lost when out alone)
Linguistic difficulties (e.g. inability to find words)
Change in behaviour (depression, agitation, aggression)
Other medical conditions such as hypothyroidism, vitamin B12 deficiency and depression may present with similar symptoms and should be considered before diagnosing dementia.
Cognitive Testing. In patients with suspected or confirmed dementia, the pathway described in what follows (Figure 9.1) can help to clarify the patient’s preoperative and post-operative status and guide further specialist input.
Various scales and scores can routinely be used instead of comprehensive neuropsychiatric testing. These tests are not diagnostic, but they allow a baseline to be established and comparison to be made with previously identified cognitive deficits.
The Mini-Mental State Examination (MMSE) is the most extensively used scale and assesses cognitive performance in a range of domains to generate a 30-point score. A score of more than 25 is normal, 21–4 is consistent with mild, 10–20 is moderate and less than 10 indicates severe cognitive deficits. This examination is, however, not suitable for patients with sensory impairments or linguistic difficulties. The MMSE is also poorly sensitive for mild cognitive impairment, particularly in highly educated populations. The MMSE was freely available until 2000 when the copyright for it was transferred to a company. In 2011 enforcement of the MMSE copyright was initiated. This has led to the use of alternative screening tools for assessment of cognitive impairment.
The Montreal Cognitive Assessment (MoCA) is a 30-point scale with higher sensitivity for mild cognitive impairment than the MMSE. It includes a clock-drawing test and five-point recall. It is validated for use in multiple languages and in patients with visual impairment. Scores between 18–26 indicate mild cognitive impairment, and scores below 18 point to likely dementia. It is freely available at www.mocatest.org.
Patients identified to have cognitive impairment or dementia in preoperative assessment should ideally be referred for specialist evaluation and management in a memory clinic. Additionally, multidisciplinary assessment and support can be initiated to enhance independence and quality of life. If surgery is relatively urgent, this need not necessarily take place before the planned surgery and can be arranged before discharge.
A risk assessment for delirium should take place in all patients with suspected or confirmed cognitive dysfunction. A simple screening tool for this is presented in Table 9.3.
The tool is currently in use at Imperial College Healthcare NHS Trust, London, UK. Adapted with permission.
Perioperative and Anaesthetic Planning
Anaesthetic planning should aim at providing optimal perioperative conditions and environment by avoiding or minimising the use of drugs and anaesthetic techniques, which increase the risk of delirium and POCD.
Patients with dementia are likely to be elderly and suffer from frailty and other medical and mental co-morbidities such as chronic obstructive pulmonary disease (COPD), diabetes, hypertension, heart disease and depression. These conditions need to be assessed and optimally managed.
Nutritional and hydration status should be evaluated and, if possible, optimised preoperatively.
Routine drug regimens need to be reviewed and scrutinised for side effects. For instance patients may be taking the centrally acting acetylcholinesterase inhibitors donepezil or rivastigmine as treatment for early mild to moderate stages of AD. These drugs have systemic parasympathomimetic cholinergic side effects such as bradycardia, increased salivation, bronchial secretions and bronchospasm. They also stimulate gastric secretions’ intestinal motility and may cause nausea, vomiting and diarrhoea. The antipsychotic drug risperidone is licensed for treatment of severe aggression associated with dementia, and its side effects include rigidity and Parkinsonian symptoms.
Should there be a suspicion of excess alcohol use, usual consumption should be quantified and documented. Acute withdrawal could lead to delirium tremens, and these patients are also at risk of Wernicke’s encephalopathy, which can be prevented and managed with chlordiazepoxide at reducing doses and high-dose parenteral vitamin B and C (Pabrinex™).
Information regarding the operation and other related details should be available in layman’s terminology for patients and carers to read.
Adjustments of perioperative hospital surroundings should be made allowing personal belongings and photos at the patient’s bedside and family, friends or usual carers to accompany the patient on the ward, as well as in the anaesthetic and recovery areas.
The patient’s contact with his or her environment should be optimised by providing good lighting, as well as visual and hearing aids.
Maintenance of day-night routine should be ensured to stimulate normal sleep patterns and to minimise pharmacological ‘night sedation’.
Repetitive orientation strategies should be made available, such as visible calendars, clocks, signposts, letter boards or pictures with repetition of factual information the patient may be inclined to forget.
Careful consideration should be made regarding the type, technique and length of surgery, for instance, benefits and risks of open versus laparoscopic techniques; reducing the length of surgery may be of benefit.
No or minimal use should be made of benzodiazepines, long-acting opioids and antiemetics with psychotropic and/or dyskinetic side effects, such as metoclopramide or droperidol.
Monitoring of anaesthesia depth may be of benefit (Chan et al., 2013).
Maintain normotension and optimal oxygen delivery during major surgery and post-operatively, in a high-dependency or intensive care unit. preoperative anaemia should be assessed and possibly treated, and significant blood loss adequately replaced.
A multimodal analgesic plan should be carried out, including local and regional analgesia where possible, with early involvement of an acute pain management team.
Vigilance should be taken and early treatment provided for complications contributing to delirium, such as pain, infection, electrolyte disturbances, dehydration, urinary retention and constipation.
Small, titrated doses of haloperidol for severe agitation and delirium with agitation can be considered. However, patients with Lewy body dementia or PD and dementia are at high risk of developing Parkinsonian symptoms and even neuroleptic malignant syndrome when antidopaminergic neuroleptic drugs such as droperidol and haloperidol and the antiemetic drug metoclopramide are given.
Parkinson’s disease (PD) is the most common movement disorder, affecting 1–3 per cent of elderly patients. The classic features of the disease include resting tremor (often asymmetrical), reduced spontaneous movements (bradykinesia), ‘cogwheel’ muscle rigidity and gait disturbance. Involvement of facial, oropharyngeal and laryngeal muscles may affect speech, swallow, gag and cough.
The main pathophysiology of PD is reduced dopamine secretion from the substantia nigra to the basal ganglia. This is also associated with an increased activation of NMDA receptors in the striatum, which bind glutamate released from the cerebral cortex (Wright, Goodnight and McEvoy, 2009). PD can also affect cholinergic, noradrenergic and serotoninergic transmission and thus cause degeneration of central olfactory, cortical, brainstem and spinal, as well as peripheral autonomic neurons. This can lead to a wide variety of associated non-motor symptoms, which do not respond to dopamine substitution. These include anosmia, fatigue, sleep and dream disorders, sensory disturbances, hallucinations, neuropathic pain, mood disorders (mainly depression), psychosis and autonomic dysfunction. In more advanced stages this autonomic dysfunction can affect the cardiovascular and gastrointestinal systems, bladder and thermoregulation. A majority of patients with PD develop dementia late in their disease course (Hudson and Greene, 2015; Nicholson, Pereira and Hall, 2002; Probasco et al., 2013).
The treatment principles of PD are pharmacotherapy and, less commonly, high-frequency deep brain stimulation (DBS). Pharmacotherapy is mainly centred on increasing cerebral dopamine activity by administration of levodopa, dopamine agonists and/or reduction of dopamine metabolism by inhibitors of monoamine oxidase (MAO) B or catechol-O-methyltransferase (COMT). High-frequency DBS is focussed on the subthalamic nucleus or globus pallidum.
Preoperative Assessment and Planning
Preoperative assessment and planning should focus on the following potential problems related to the disease and its treatment (Brennan and Genever, 2010; Mariscal et al., 2012; Nicholson et al., 2002).
Rigidity can affect mouth opening, jaw and neck motility and can lead to difficulties in airway management.
Impaired gag, swallow, cough and autonomic dysfunction of the oesophagus and stomach increase the risk of bronchopulmonary aspiration.
Obstructive and/or restrictive respiratory problems can be caused by airway and chest wall muscle rigidity.
Worsening brady- or akinesia may be associated with perioperative stress and difficulties administering anti-Parkinsonian drugs.
Severe tremor and dyskinesia can make regional anaesthesia difficult or unsafe.
Autonomic dysfunction can trigger severe hypotension under general and neuraxial anaesthesia. Other important aspects of autonomic dysfunction that can cause perioperative problems and complications include excessive salivation, oesophageal dysmotility, gastric stasis and bladder dysfunction, as well as impaired temperature regulation.
Anxiety and depression can be exacerbated in the perioperative period.
Perioperative prophylaxis and treatment for nausea and vomiting based on antidopaminergic drugs, such as metoclopramide, phenothiazines (e.g. prochlorperazine) and butyrophenones (e.g. haloperidol, droperidol), must not be used in patients with PD as they can exacerbate Parkinsonian symptoms. Domperidone is a peripherally acting dopamine antagonist occasionally used to treat levodopa-induced nausea and vomiting and can be considered as an antiemetic or prokinetic perioperatively.
It is paramount that an individual patient’s normal treatment and care routine is maintained as near as possible as usual during a hospital stay and in the perioperative period. Levodopa and other anti-Parkinsonian medications should be administered according to the individual patient’s routine schedule, which should not be changed for the convenience of ‘drug rounds’ on hospital wards. Some patients take extra doses of levodopa in between scheduled times for symptom control. This must be explored with them or their carers and included in the medication they receive in hospital. An enteral feeding tube may need to be inserted in order to ensure that administration of medications is not interrupted perioperatively, including during prolonged operations. Delay or cessation of administration of all types of dopaminergic medication can lead to Parkinsonian hyperpyrexia, similar to neuroleptic malignant syndrome. In addition to fever, its manifestations include severe tremor, muscle rigidity, laryngospasm, impaired ventilation and rhabdomyolysis, as well as delirium and autonomic instability. This is more common in patients with advanced severe PD and on high-dose L-dopa.
Preoperative planning for patients with PD should be in close collaboration with their neurology team and specialist nurse. Scheduling a patient with PD first on an operating list may also help in planning perioperative drug administration and early re-involvement of the neurology team post-operatively (Brennan and Genever, 2010).
The multisystem impact of PD in the perioperative period ideally warrants high-dependency care, particularly after major surgery.
Perioperative Implications and Management of Specific Medications in Parkinson’s Disease
Dopamine supplementation is provided by enteral administration of levodopa with a peripheral decarboxylase inhibitor, e.g. carbidopa (Sinemet™) or benserazide (Madopar™). These combinations prevent the peripheral metabolism of levodopa to dopamine and reduce nausea and vomiting from dopaminergic stimulation of the area postrema, which lies outside the blood brain barrier.
Levodopa has a short elimination half-life. Withholding it increases the risk of worsening Parkinsonian symptoms or Parkinsonian hyperpyrexia. A patient’s normal levodopa administration should be maintained as near as possible to the time of surgery and restarted as soon as possible post-operatively. In prolonged operations regular administration of levodopa should be continued intra-operatively by gastric tube. Thus an enteral feeding tube may have to be inserted for drug administration, even if it is not indicated by the type of surgery. When changing from slow-release to dispersible formulations of levodopa with carbidopa or benserazide, a 30 per cent dose reduction is necessary due to increased bioavailability (Brennan and Genever, 2010).
The ergot dopamine agonists bromocriptine, cabergoline and pergolide have been associated with pulmonary, retroperitoneal and cardiac fibrosis, which can lead to aortic and/or mitral regurgitation. Thus patients whose past or present treatment regimens have included these drugs need careful respiratory, cardiac, abdominal and renal assessment. These side effects have largely led to this group of drugs being replaced by non-ergot dopaminergic agonists such as pramiprexole, ropinirole and rotigotine.
Skin patches of rotigotine (Neupro™) are available for once-daily administration.
Side effects of all dopaminergic agonists include sedation, hallucinations, cognitive impairment and impulse control disorders.
These drugs increase dopamine availability by reducing its breakdown and they may have a disease-modifying effect by reducing the oxidative metabolism of dopamine. They also decrease the metabolism of serotonin. This leads to a risk for serotonin syndrome with co-administration of selective serotonin reuptake inhibiting (SSRI) antidepressants and other drugs with serotonergic effects, e.g. pethidine, fentanyl and tramadol. Some authors recommend discontinuing these drugs 3 weeks prior to elective surgery. However, as discontinuation of Monoamine Oxidase Inhibitors (MAOI) could lead to a deterioration of PD, a preferable approach may be to continue MAOI medication and to avoid any perioperative drugs with serotonergic effects. MAOI may also affect liver function, particularly in patients with pre-existing liver disease.
The peripheral inhibition of decarboxylation by carbidopa or benserazide results in dopamine mainly being metabolised by cerebral Catechol-O-Methyltransferase (COMT). The COMT inhibitors increase the cerebral availability of levodopa and the duration of its action. This can also lead to an increase in dopamine side effects such as nausea, vomiting and dyskinesia, which may require a dose reduction of levodopa. Fatal liver toxicity has been reported with tolcapone and liver function needs to be closely monitored in patients taking this drug. Stalevo™ is a drug combining levodopa with carbidopa and the COMT inhibitor entacapone.
Amantadine is a direct dopamine agonist, and possibly also an NMDA antagonist. It is used relatively rarely to reduce dyskinesias in advanced PD. It should not be discontinued abruptly as this may lead to withdrawal with worsening symptoms.
Centrally acting anticholinergics like trihexyphenydil and benztropine are prescribed to diminish Parkinsonian tremor, but are rarely used in current practice. They can be associated with worsening glaucoma, urinary retention, central anticholinergic confusional syndrome and cognitive impairment. These side effects could be potentiated in the perioperative period and it may be safer to discontinue these drugs before elective surgery.
PD patients with Lewy bodies suffering with hallucinations and psychosis or dementia are usually treated with the atypical antipsychotics clozapine and quetiapine, because antidopaminergic phenothiazines and butyrophenones need to be avoided. Sudden cessation of clozapine perioperatively should be avoided. This drug can lead to severe dyskinesia and decompensation of psychosis. Besides, if clozapine is discontinued for more than 48 hours, it must be restarted at a low dose and slowly increased. However, it also has to be borne in mind that side effects of clozapine include severe gastrointestinal hypomotility, which can lead to gastric stasis, ileus, constipation, faecal impaction and toxic megacolon. This risk may be greater in the perioperative period. Rarely, clozapine may be associated with a severe cardiomyopathy or agranulocytosis and routine monitoring of white cell and neutrophil counts is mandatory.
Perioperative Conversion of Enteral to Parenteral Drug Treatment for Parkinson’s Disease
If perioperative enteral administration of anti-Parkinsonian medication is complicated by abnormal gastrointestinal function, a rotigotine skin patch can be used as a temporary monotherapy alternative. Rotigotine skin patches (Neupro™) are available for once-daily administration with a dose range of 2–24 mg in 24 hours. Eight milligrams of rotigotine is approximately equivalent to 300 mg of levodopa. Another alternative is subcutaneous injections or an infusion of the short-acting dopamine agonist apomorphine. The latter is strongly pro-emetic, and should be accompanied by routine administration of antiemetics such as ondansetron or domperidone (Brennan and Genever, 2010; Mariscal et al., 2012; Nicholson et al., 2002).
The algorithm in Figure 9.2 provides detailed information on exact dose conversion from enteral to parenteral anti-Parkinsonian therapy.
N-methyl D-aspartate (NMDA) antagonism with low-dose ketamine (10 mg increments, up to 0.1–0.5 mg/kg) has been reported to abolish tremor, dyskinesia and dysarthria perioperatively (Wright et al., 2009). This may be helpful for placement of lines or regional blocks, airway management and as bridging replacement therapy when the routine administration of L-dopa is hampered in the perioperative period. The analgesic effect of low-dose ketamine may be of additional benefit and may reduce the need for opioids, which can be problematic and exacerbate rigidity in PD patients and cause nausea, vomiting, gastric stasis and constipation.
Anaesthetic and Procedural Issues in Patients with Deep Brain Stimulation
The deep brain stimulation (DBS) lead is connected to a generator implanted subcutaneously in the chest wall, usually in the subclavicular area. Anaesthetic concerns associated with DBS are similar to cardiac pacemakers with regard to magnetic resonance imaging (MRI), surgical diathermy, electrocautery, defibrillation and cardioversion (Dobbs, Hoyle and Rowe, 2009).
MRI. The electromagnetic field generated by MRI scanners can lead to heating or dislodgement of the DBS lead with damage to the surrounding brain tissue. It can also induce stimulation, reset the lead or turn it off. Nevertheless, there are published case series of safe MRI in patients with DBS. Manufacturers such as Medtronic have detailed Internet resources offering advice to provide safe MRI scans for patients with DBS.
Diathermy. Unipolar diathermy can suppress, reprogram or damage DBS leads, and the best option is to use bipolar diathermy. The safest approach, particularly if unipolar diathermy cannot be avoided, is for the DBS system to be temporarily deactivated preoperatively by a specialist with an external handheld device. If this is not possible in urgent surgery, the diathermy current pathway to the ground plate should be placed as distant as possible from the stimulator in the chest wall and the lead, which runs subcutaneously through the neck and behind the ear to the brain. The lowest voltage and power settings should be used for unipolar diathermy.
Defibrillation and cardioversion. The defibrillator pads need to be positioned as far away as possible from the DBS generator in the chest wall and perpendicular rather than parallel to the lead pathway. It is prudent to use the lowest appropriate energy level.
Epilepsy is a cerebral disease manifesting as recurrent unprovoked seizures. It can be secondary to a structural abnormality in the brain, a variety of metabolic and infective conditions or idiopathic. It is the most common neurological illness in all age groups with a prevalence of 0.5–2 per cent (Carter and Adapa, 2015; Hudson and Greene, 2015; Perks, Cheema and Mohanraj, 2012; Probasco et al., 2013).
Seizures can be classified in the following manner:
Generalised: affecting both cerebral hemispheres, usually tonic-clonic and with loss of consciousness;
Focal: affecting part of one cerebral hemisphere
with interruption of consciousness: complex focal
without loss of consciousness: simple focal.
Complex partial seizures and non-convulsive status may sometimes be difficult to recognise, particularly in elderly patients. They may present as acute confusion or change in mental state rather than classic epileptic features. If no other obvious cause of altered consciousness or another acute change in mental state can be found, these conditions should be excluded and a neurological consultation and EEG considered (Beyenburg, Elger and Reuber, 2007).
Patients with recent onset of or poorly controlled epilepsy require careful general and neurological review prior to elective surgery. This assessment should include diagnosing or excluding haemorrhagic or ischaemic stroke (the most common cause of epilepsy in adults older than 35), brain tumour, meningo-encephalitis (infectious or autoimmune), hypoglycaemia and severe hyponatraemia, as well as drug toxicity and withdrawal (alcohol and benzodiazepines).
Anaesthetic pre-assessment of patients with epilepsy must include a meticulous history, and the input of relatives and carers can be very useful. Important details to be ascertained and recorded include:
the type and frequency of seizures;
known provoking factors for seizures, such as sleep deprivation or menstruation;
the nature of auras and other associated phenomena;
how long ago the last seizure occurred;
the dose and dosing intervals of antiepileptic medication and a review of side effects and possible interactions with anaesthetic and analgesic drugs of the individual patient’s antiepileptic medications;
the use of alcohol and other recreational drugs;
anticonvulsant levels (e.g. phenytoin, valproate) should be measured if seizure frequency has recently increased or there are signs of drug toxicity.
It may also be helpful to review recent brain imaging and EEG reports.
If a patient is on a ketogenic diet to control epilepsy refractory to drug therapy, the exact details of this diet and how it is best maintained in the perioperative period need to be ascertained. Preoperative carbohydrate drinks and perioperative dextrose infusions should be avoided in these patients. In addition to routine blood result screening preoperatively, it is wise to check the patient’s blood glucose and ketone levels.
The presence of non-epileptic attacks, which can coexist with true epileptic seizures, may need to be explored. Patients with non-epileptic attack disorder could have a history of multiple hospital and intensive care admissions as well as other medical and psychological problems. They may also report deterioration with anticonvulsant treatment. Non-epileptic attacks tend to last longer than epileptic seizures, usually more than 90 seconds, with asynchronous head and limb movements, no cyanosis and eye closure resistant to opening. Differentiating between non-epileptic attacks and true seizures is important in the perioperative period, so that the harmful side effects of sedation and anticonvulsants can be avoided (Reuber, Enright and Goulding, 2000).
Perioperative management should include avoidance of factors precipitating seizures such as sleep deprivation, hypoxia, hypoglycaemia, hyponatraemia, hypocalcaemia, hypomagnesaemia or diet change in patients normally on a ketogenic diet. Anticonvulsant medication must be continued where possible pre- and post-operatively, and this is best planned with the treating neurology team. This may necessitate administration via an enteral feeding tube. Parenteral antiepileptic treatment must be considered in surgery with post-operative gastric stasis and ileus. Antiepileptic drugs available for parenteral administration include: diazepam, lorazepam, phenytoin, phenobarbitone, valproate, levetiracetam and lacosamide.
It would be prudent to have an emergency supply of rectal diazepam or buccal midazolam and/or intravenous benzodiazepines easily accessible in the perioperative period.
Should seizures occur, first-line management is usually with a benzodiazepine such as lorazepam 2–4 mg i.v. Reversible causes such as hypoxia, hypotension, hypoglycaemia, hyponatraemia, hypocapnoea, alkalosis, hypocalcaemia and infection must be excluded and treated appropriately if necessary.
It is important that epileptic seizures are distinguished from abnormal movements such as myoclonus, and rigidity associated with anaesthetic or analgesic drugs, such as propofol, etomidate or fentanyl or post-operative shivering.
Antiepileptic therapy may cause cerebral side effects, particularly sedation and reduced cognitive ability. Some anticonvulsants affect anaesthetic drug requirements through tolerance and enzyme induction. Several antiepileptics can be associated with elevation of liver enzymes and blood dyscrasias.
Some specific issues relating to commonly prescribed anticonvulsants are summarised next:
Barbiturates and carbamazepine: enzyme induction may lead to increased metabolism and reduced plasma levels of these drugs and other drugs metabolised by the same enzymes. These drugs include sedatives, beta-blockers, calcium channel blockers, amiodarone, warfarin and some antibiotics.
Phenytoin: enzyme induction and its reliance on high protein binding may lead to changes in blood levels including other drugs. Phenytoin has saturable elimination kinetics, and adult patients on doses above about 500 mg/day may experience a linear rise towards toxic plasma levels with each further dose. There is also a high inter-individual variability in the pharmacokinetics. Important side effects of this drug include cardiac arrhythmias and conduction block, liver dysfunction, blood dyscrasias (e.g. megaloblastic anaemia, leukopenia), ataxia, sedation and coma, particularly with toxic levels. Changes in blood albumin level in the perioperative period may lead to increases in free drug, leading to toxic levels. The corrected phenytoin level can be calculated using the following formula: