Introduction

An important part of anesthetic practice is obtaining informed consent for the anesthetic procedure. Although institutional and jurisdictional legal practices vary according to local and international requirements the basic tenets of informed consent are broadly applicable across anesthesiology practice worldwide. The manner in which consent is obtained may range from a simple documentation in the patient chart to a more detailed elaboration of the details discussed. Many institutions have developed customized forms, which list the specific risks of anesthesia. Whatever model is used, it must be underpinned by a frank explanation and discussion of the anesthetic plan with the patient, including risks and benefits to fulfill both ethical and legal obligations. The purpose of this chapter is not to delve into the ethical and legal implications of consent, which have been well addressed in the literature, by professional societies and by medical indemnity organizations, but to specifically focus on how cognitive decline in the elderly directly impacts on the process of undertaking and obtaining such consent.

The issue of cognitive capacity may have been seen as a relatively isolated issue in the past being relegated to the clinical extremes of cognitive impairment; however, developed and developing countries are witnessing a massive increase in the aging demographic. The effects of the aging population will impinge on all aspects of society – altering both the way we live and work, with vast social and economic consequences. According to the World Bank, the average life expectancy in the United States has increased from 70 years in 1960 to 79 years in 2012 and projections predict this will rise even further in the future. The elderly utilize health services to a much greater extent than those who are younger. For example people aged > 65 years consume four times as many health services as those who are younger. It is sobering to recognize that the fastest growing segment of the population in the United States is the group 80 years and older [1].

One of the results of the aging demographic is the rapidly increasing numbers of elderly presenting for anesthesia and surgery. Although a myriad of health issues beleaguer the elderly, many eventually lead to either some form of invasive surgery or some less invasive procedure (e.g., coronary artery stents, colonoscopy). For example, in 2010 in the United States, more than 19 million patients over 65 years (representing 37% of all anesthetics) presented for surgery [2]. In Australia, projections are that by 2050, 50% of all anesthetics will be administered to those over 65 years old [3]. These elderly patients present challenges in anesthetic management because of associated systemic illnesses such as hypertension, cardiovascular disease, and diabetes. However, cognitive decline, which increases markedly with age, threatens to become an overwhelming issue.

Preoperative Cognitive State

The prevalence of cognitive decline before surgery has been reported as high as 32% for noncardiac surgery [4], and from 25% to 45% before cardiac surgery [5–7]. Other sections of this book deal with this in more detail but these figures demonstrate that even before surgery commences, a substantial proportion of elderly patients suffer from cognitive impairment to some degree. The prevalence identified before surgery is higher than that seen in nonsurgical population studies of the elderly, where mild cognitive impairment (MCI) has been identified in 14%–18% of those aged over 70 years in well-conducted population studies [8].

This disparity in the prevalence of cognitive impairment reported in population studies compared with that reported preoperatively relates to both the criteria used in attribution and possibly the patient population itself. The definition of subtle cognitive impairment used in population and clinical studies differs from the definition used in studies of anesthesia and surgery. Neurologists and old-age psychiatrists have used the term mild cognitive impairment or, more recently, mild neurocognitive disorder (DSM-5) to classify individuals with subtle cognitive decline [9]. The criteria for classification differ from the term preexisting cognitive impairment (PreCI) used in many studies of anesthesia and surgery. These differences in the rules for classification may account in part for the differences in reported prevalence of cognitive impairment in population and anesthesia studies. In one study reporting the prevalence of cognitive impairment before elective hip arthroplasty, 22% of patients were classified as MCI and 20% as PreCI but only 30% of the patients with PreCI were classified with MCI, indicating the different definitions may detect different patients with cognitive impairment [10].

Another contributing factor to explain the increased prevalence of cognitive impairment in those undergoing surgery is that these individuals have ongoing systemic pathology. Systemic effects of illness and chronic inflammation are present whether it be cardiovascular in nature (e.g., cardiac and vascular surgery) or some other pathology (e.g., cancer) which has brought them to require surgery. Certainly the presence of cardiovascular disease predisposes to cognitive decline [11]. Regardless of the reason, we need to acknowledge that up to a quarter of patients over 65 years who present for anesthesia and surgery will harbor some degree of cognitive decline.

The diagnosis of MCI requires that individuals are not limited in their daily life by cognitive decline and are not demented, but have a diminished capacity on neuropsychological testing and a complaint in their cognitive ability by either themselves or an informant [12]. While a few of these patients may present for surgery with such a diagnosis known, the majority will be unidentified because of the subtle nature of the condition. MCI is already present in over 14%–18% of those over 70 years, and increases with age, indicating that many patients presenting for anesthesia and surgery will be subtly cognitively compromised.

It should be recognized that cognitive decline forms a continuum and patients with more advanced forms of cognitive impairment may also present without a diagnosis. In a simple brief perioperative consultation even advanced dementia may not be apparent without careful and specific questioning of the patient or consultation with an informant who knows about the patient.

Postoperative Cognitive State

Cognitive decline is not only present before anesthesia and surgery but is itself a well-established consequence of the anesthesia/surgical process. Postoperative cognitive dysfunction (POCD) historically has referred to a decrement in cognition at various times after anesthesia and surgery. It has been most commonly defined by a reliable change index (RCI), using a control group to identify change [13]. It occurs not only after cardiac and general surgery but also after noninvasive procedures with sedation, such as coronary angiography [14]. As with preoperative cognitive impairment, POCD has occurred most commonly in the elderly. For example, Newman et al. reported that cognitive decline was evident in 53% of cardiac surgical patients at discharge, 36% at 6 weeks and 24% at 6 months postoperative [15].

The incidence of POCD is high when measured 7 days after anesthesia and surgery, and the new nomenclature, as recommended by the Nomenclature Consensus Working, Group [16], would term this delayed neurocognitive recovery (dNCR), as most patients do indeed recover. For noncardiac surgery the incidence is 17%–25% [14,17] and appears to be even higher at 7 days following cardiac surgery, 43%–53% [14,15]. At 3 months, when the new nomenclature adopts the DSM-5 terminology of either major or minor neurocognitive disorder (postoperative), there remains a significant number of patients with cognitive impairment after both noncardiac and cardiac surgery and even noninvasive procedures – this has been reported as a POCD incidence of 16% after both cardiac surgery and elective hip replacement and an incidence of 21% after coronary angiography [14]. Whether these decrements in cognition continue for longer than 3 months or even permanently is less well established. For noncardiac surgery, the evidence seems to support that the incidence of cognitive impairment at 12 months does not differ from nonoperative controls [4,18]. For cardiac surgery, the evidence of long-term cognitive impairment associated with the procedure is equivocal, with some studies supporting prolonged cognitive decline, while others have failed to substantiate this (see Table 15.1).

In summary, many patients over 65 years of age will present for surgery with preexisting cognitive impairment, which may be subtle (but detectable on neuropsychological testing), or more advanced (many of whom remain undiagnosed). Furthermore there is sound evidence that many of these patients will suffer further decline as a result of anesthesia and surgery [4]. The most prominent feature associated with both preoperative and postoperative cognitive decline is increased age. It is now well established that, at least in noncardiac surgery, preoperative cognitive impairment is a major risk factor for postoperative cognitive dysfunction [4]. Remarkably, the impact of these cognitive issues on the daily conduct of anesthesia and surgery has only recently received attention.

Informed Consent and Cognition

Informed consent presumes that an individual fully understands and comprehends the information presented to them and is competent to make an appropriate decision to accept or modify the risks and benefits so explained. Any decrement in cognitive ability has the potential to interfere with the informed consent process. The implications of impaired cognition, both preoperatively and postoperatively, have been largely ignored, likely due to factors such as: the presumed inability to modify risk, denial of the existence of perioperative neurocognitive disorder (PND), a desire to avoid procedural delays, or simply a desire to avoid confrontation and stigmatization.

Informed consent is a fundamental part of medical practice in which the physician explains the planned procedures, and possible alternatives with attendant risks and benefits. In anesthesia and surgery, cognitive decline impacts on consent in three key areas: (1) the competence and capacity of an individual with cognitive decline to understand and provide informed consent for a procedure; (2) the information provided by the anesthesiologist to the patient regarding the likelihood and consequences of suffering cognitive decline after anesthesia and surgery; and (3) when participating in research projects, the responsibility to inform participants of their test outcomes. Consent for research has been considered in depth by the Presidential Commission for the Study of Bioethical Issues [23] and will not be discussed in detail here. The importance of item (1) above relates not only to the presentation for elective surgery, but also the need to consent for emergency procedures or for procedures subsequent to an original intervention, during which time PND may exist.

Competence and Capacity

Valid informed consent requires that the anesthesiologist makes a judgment about an individual’s ability to undertake an informed choice concerning the anesthetic. Competence is defined as the capacity to understand the general nature and effect of the proposed treatment [24]. Most discussions of competence divide individuals into two groups. Either the individual is competent to make informed consent or they are not. Every person is presumed to have the capacity to consent or refuse medical treatment unless and until that presumption is rebutted [25]. In general, it has been assumed that normal doctor-patient interactions should expose any deficiency on the part of the patient to provide informed consent.

Although there is little doubt that patients with frank dementia or severe cognitive impairment would not have the mental competence to understand medical explanations, these patients are not always diagnosed on presentation and they may mask deficiencies quite well on interview. Elderly patients with mild cognitive decline are also likely to show no overt signs of cognitive deficiency. When it is obvious that the patient does not have the capacity to provide informed consent, specific alternatives exist. These include relatives acting on behalf of the patient (e.g., with medical power-of-attorney) or in their absence, guardianship can be provided by the state or relevant authority. However, such cases represent a minority of patients presenting for surgery. As explained above, anesthesiologists will be presented with increasing numbers of elderly patients who are able to function normally in daily life, yet harbor cognitive deficiencies which become revealed only when tested with suitably sensitive tests of cognition. Our own experience in running a CSF clinic for the diagnosis of Alzheimer’s disease has reinforced our assertion that assessment of these patients’ cognitive abilities during a standard preoperative consultation is highly insensitive. Many patients with frank dementia are able to partake in social conversation with minimal indications that they are impaired. It is only after extracting a history from a suitable informant, comparing to known facts (e.g., a medical record), or referring to previous cognitive assessments that one becomes aware of the magnitude of cognitive decline. Many of these patients have provided a history on admission which, on reflection, could have been total confabulation, yet they may be invited to sign a “consent for treatment” form.

Consent requires that the anesthesiologist (and surgeon) decide whether the individual has the capacity to make an informed decision. This is known as medical decision-making capacity. The definition of a person incapable of giving consent has been defined as “incapable of understanding the general nature and effect (i.e., consequences) of the proposed procedure or treatment, or incapable of communicating consent or refusal” [24]. Disorders which cause acute or chronic decreases in cognition such as intellectual disability, head injury, dementia, delirium, and psychosis are well known to affect a person’s capacity – as do some medications. These disorders are usually self-evident and have been well addressed from both an ethical and legal perspective. Most hospitals and jurisdictions have well-defined processes for obtaining informed consent from surrogates whether they be family members, hospital administrators, or legal appointees.

The importance of the assessment of capacity to consent cannot be underestimated. It is demeaning to label a competent person as incapable of informed consent. Conversely to accept informed consent from an individual who does not have capacity is ethically untenable. Currently in the law there is a presumption of capacity in anyone over a specified age, typically 18 years. This presumption may be rebutted by evidence to the contrary. The law regards capacity to be assessed in the context of a person’s cognitive abilities in relation to the decision being made. For example, in legal matters, a person with cognitive decline may be competent to make a will but not enter into a contract.

As with most professional organizations, the Australian and New Zealand College of Anaesthetists, the Association of Anaesthetists of Great Britain and Ireland, and the American Society of Anesthesiologists guidelines for consent also presume adults to be competent unless there are reasonable grounds for believing otherwise [26–28]. These recommendations, supported by legal advice, disregard both the prevalence and hidden nature of cognitive impairment in the elderly.

A problem arises where individuals do not overtly appear incapable of giving informed consent, but may harbor cognitive impairment which in fact makes informed consent questionable. Such a situation may be very common in those with subtle cognitive impairment or unrecognized cognitive decline. An excellent example of this is demonstrated in a report by Marcucci et al. [29]. They describe two cases where individuals were able to understand and consent to the surgical procedure which involved simple concepts of surgery and removal of a tumor because surgery is a fairly tangible process. However, they were not able to understand the more abstract concepts of anesthesia. The situation was resolved by allowing the patients to consent to the surgery and obtaining surrogate consent for the anesthesia. In both cases, rudimentary cognitive testing demonstrated cognitive deficits which were not overtly apparent.

This report highlights the difficulties in obtaining informed consent from those with subtle or concealed cognitive impairment. These individuals often appear “normal” during casual conversation. There is no doubt that many elderly individuals we see every day if properly questioned, would not have the acuity to provide informed consent for anesthesia.

Okonkwo et al. recruited patients with Parkinson’s disease to study medical decision-making capacity and MCI [30]. Five core consent standards were used to assess capacity using hypothetical medical scenarios: expressing a treatment of choice; appreciating the consequences of the treatment choice; providing rational reasons for a treatment; understanding the risks and benefits; and making the reasonable treatment choice. Patients with MCI performed comparably well to controls on minimal consent standards requiring expression of choice or making the reasonable treatment choice, but significantly below controls on the three clinically relevant standards of appreciation, reasoning, and understanding. They concluded that patients with MCI demonstrate significant impairments on clinically relevant abilities associated with capacity to consent to treatment.

It is likely that the inability to provide informed consent for both surgery and anesthesia will dramatically increase in concert with the increasing numbers of elderly presenting for invasive or diagnostic procedures. The current standing of the law defaulting to competence in those over the age of 18 years may well be inappropriate for some elderly. This is a concerning situation for both surgeons and anesthesiologists, who currently assume that individuals have the capacity to consent, if no overt evidence to the contrary exists. Such an approach is currently supported by the courts [25]. Yet there is no doubt that such an approach completely ignores the discrepancy between superficial social interactions and inability to comprehend the more abstract concepts of anesthesia [29]. Appelbaum and Grisso [30] have refined the requirements of competence into four major categories:

• 
  

  Communicating choices: the inability to make a choice is a self-evident indication that the individual is unable to make informed consent.

  
  
• 
  

  Understanding relevant information: beyond the mere reception, storage, and retrieval of information, patients must be able to understand the fundamental meaning of information; to test understanding, it is best to ask that patients paraphrase the information.

  
  
• 
  

  Appreciating the situation and its consequences: patients must not only understand the information but must grasp what it means for them.

  
  
• 
  

  Manipulating information rationally: involves ability to reach conclusions which are logically consistent with the starting premises; this may be tested by asking patients appropriate questions.

There is no general agreement on how to measure any of these categories, although there have been several attempts to design assessment tools [32]. To understand and appreciate a procedure a patient must encode, retain, and process information which involves multiple cognitive domains such as attention, memory, and cognition [33]. Incorporating feedback from patients on their level of understanding (beyond simple recall), as part of the consent process, will help to confirm competence. To test understanding, it is best that patients paraphrase the information; to test appreciation of the situation requires eliciting their conception of their illness, need for treatment, and likely outcomes; to test rational manipulation of information one must ask patients for the reasons underlying their choices [31].

In reality, the situation is even more complex. Comprehension of different procedures may require different levels of cognitive ability. In the cases described above by Marcucci et al., the degree of patient capacity was directly related to the conceptual complexity. Simple anesthetic concrete concepts may be more accessible than more abstract concepts such as being unconscious or providing postoperative pain relief. Informed consent for research poses even further problems as many of the concepts are difficult for patients to comprehend. Research subjects need to understand the distinction between treatment and research. Subjects may need to understand that they may be randomized to a particular treatment arm that provides no direct benefit to them. Research subjects often believe that therapeutic decisions are always made for their benefit, a misconception that Appelbaum has labeled, “the therapeutic misconception” [34]. Finally, to make the situation even more complicated, we know that cognition is not a stable condition but varies over time [35].

The logical conclusion is that even though there may be no discernible reason to suspect that informed consent is not obtainable in the elderly, the current default that all elderly patients are competent to sign informed consent should be questioned. Although the law currently supports obtaining consent from the elderly who appear “normal” at face value, we now know that many suffer from subtle cognitive impairment and are incapable of such a process. Certainly from an ethical point of view, it is no longer acceptable to explain the surgery and anesthetic plans and obtain a signature as confirmation of informed consent. In the future a comprehensive confirmation that the four categories of competence as outlined above should in some way be reaffirmed before such a signature is obtained. This is important and the time involved should not be used as a reason to avoid confirmation. Until such a feedback loop is added to the process, both surgeons and anesthesiologists may be acquiring signatures for patients’ assent (agreement without understanding) rather than valid informed consent. The presence of PND may also make consent for reoperation for immediate complications problematic as noted above. As the age of surgical population inexorably increases, these issues will become more prominent, and we hope these ethical and legal issues can be suitably addressed.