‘It was a great brain, and I miss it’: lay perspectives on postoperative cognitive dysfunction

Abstract

Background

Postoperative cognitive dysfunction (POCD) is a concern after anaesthesia and surgery, but preoperative discussion of neurocognitive risks with older patients rarely occurs. Anecdotal experiences of POCD are common in the popular media and may inform patient perspectives. However, the degree of alignment between lay and scientific perspectives on POCD is not known.

Methods

We performed inductive qualitative thematic analysis on website user comments publicly submitted under an article entitled, ‘The hidden long-term risks of surgery: “It gives people's brains a hard time”’, published by the UK-based news source The Guardian in April 2022.

Results

We analysed 84 comments from 67 unique users. Themes that emerged from user comments included the importance of functional impact (‘Couldn't work … even reading was a struggle’), attribution to a range of causes but particularly the use of general, rather than consciousness-preserving, anaesthesia techniques (‘side effects aren't fully understood’), and inadequate preparation and response by healthcare providers (‘I would have benefited by being warned’).

Conclusions

There is misalignment between professional and lay understandings of POCD. Lay people emphasise subjective and functional impact of symptoms, and express beliefs about the role of anaesthetics in causing POCD. Some patients and caregivers affected by POCD report feeling abandoned by medical providers. In 2018, new nomenclature for postoperative neurocognitive disorders was published, which better aligns with lay perspectives by including subjective complaints and functional decline. Further studies based on newer definitions and public messaging may improve concordance between different understandings of this postoperative syndrome.

Editor's key points.

• •Postoperative cognitive dysfunction (POCD) is an important problem with a lasting impact for patients.
• •Although the biological mechanisms of POCD are being actively researched, we know much less about the patient experience of this condition.
• •This study sheds light on the importance of doctors' understanding of the patient experience of POCD as part of shared decision-making about surgical treatments.

Durable cognitive change is a feared outcome for older adults after surgery and anaesthesia, prompting high-engagement articles in the popular media and specialty-level initiatives to promote cognitive health after surgery. Practising anaesthetists will likely have had the experience of navigating a patient's or family member's preoperatively expressed worry about the potential for adverse cognitive outcomes, including postoperative cognitive dysfunction (POCD). Nearly 40% of respondents to a survey on intraoperative awareness reported feeling ‘very fearful’ of suffering permanent cognitive deficits after anaesthesia,¹ and patients who experience subjective cognitive decline after surgery are likely to view their anaesthesia care negatively.² In 2018, a redefinition of postoperative neurocognitive disorders newly included functional impact and subjective impairment as descriptors³; however, it is still unknown how the subjective experience of cognitive change may impact our understanding of the POCD symptom complex. Further elements of the lay perspective on symptoms, precipitants, and beliefs about subjective cognitive change experienced after surgery have not been elicited.

In April 2022, the UK-based news source The Guardian's Sunday paper, The Observer, published online an article entitled, ‘The hidden long-term risks of surgery: “It gives people's brains a hard time”’.⁴ Written by Dr David Cox, a freelance journalist with a PhD in neuroscience, the article described patient anecdotes and interpreted recent and historical examples of published peer-reviewed literature on postoperative cognitive change (including in children). After the article's publication, The Guardian opened a moderated comments period, during which 84 comments were submitted and published online. We conducted qualitative thematic analyses on these submitted comments to identify themes reflecting the lay perspective on cognitive change after surgery, focusing on adults.

Methods

Data collection

Data were obtained from the comments section of ‘The hidden long-term risks of surgery: “It gives people's brains a hard time”’, which was posted at 05:00 EDT on April 24, 2022. The comments section was opened on April 24, at 05:14, and the last comment posted on May 7, 2022; the section was subsequently closed to further comment. Comments were obtained for analysis in June 2022 and were given an alphanumeric code to maintain structure (i.e. comments submitted in response to other comments) and chronological order while anonymising username and submission date/time.

Ethics and approval

Comments submitted to The Guardian are published under a user-selected semi-anonymous username which cannot be publicly linked to any identifying information. The Human Research Protection Program Institutional Review Board at the University of California, San Francisco, regarded this study as exempt from the need to seek participant consent (IRB#22-36952). The British Psychological Society's 2021 Ethics Guidelines for Internet-mediated Research hold that consent is not needed if online data are considered ‘in the public domain’.⁵ The Guardian explicitly notes, in its privacy policy, that user-submitted comments are publicly accessible, are identified by username, and that comments and other personal data a commenter chooses to post ‘can be viewed online and collected by other people’.⁶ We sought and obtained permission to analyse these data from the original article's author.

Researcher reflexivity statement

LL is a pre-clinical medical student. AKS is an academic geriatrician and palliative medicine physician with a research programme focused on prognostic modelling for older adults with and without dementia. DD is a medical sociologist who studies medical culture and its role in advancing health equity, particularly for older adults and those with dementia. ELW is a clinically practicing academic anaesthesiologist with a research programme in long-term cognitive change after surgery.

During the manuscript review process, the clinicians (LL, AKS, and ELW) were asked to reflect upon their pre-existing biases around POCD to offer additional context to how they approached the inductive analysis. LL was not familiar with POCD before engaging in this research. AKS felt that surgery, anaesthetics, and potentially postoperative delirium all have negative short-term implications for cognition in older adults, but felt uncertain about the causal role of those elements in long-term cognitive decline. ELW felt that the overall evidence for anaesthetics or surgery as a necessary or sufficient cause of long-term POCD was weak, and yet was aware that credible anecdotes of durable cognitive change after surgery exist. The present inductive analysis of these comments (many of which are anecdotes) was thus conceived by ELW as a strategy to reconcile the anecdotal, lay perspective, or both with experimental/observational research designed to elicit the causal role of surgery/anaesthesia in later cognitive decline.

Data analysis

Because little to no data exist describing the lay perspective on POCD, this analysis was conducted in an ‘exploratory’ qualitative health research style, without a priori hypotheses and an unstructured, adaptable analysis strategy using inductive, iterative coding.⁷ Two researchers (LL and ELW) analysed the comments using inductive thematic analysis, which was selected because it allowed the data itself to drive generation of key themes. First, comments were read and re-read individually (i.e. immersion), and then iteratively grouped into preliminary themes, with key examples being pulled out for closer analysis, separately by both LL and ELW. Then, preliminary themes were reviewed collaboratively and a set of themes and settled upon for further coding. Memoing was used throughout the analysis among the coders to detail thoughts on, and potential relationships among, the comments and the themes. In further meetings, including in collaboration with co-authors AKS and DD, themes were discussed, finalised, and placed within the context of a broader conceptual framework informed by the research literature. The themes and sub-themes, as defined and agreed upon by the authors, are presented in the Results section, along with illustrative quotations. When, post hoc, comments were considered chronologically, we observed that no themes emerged newly from the last 10 comments to be published, arguing that thematic saturation was achieved.

Post hoc robustness analysis

We conducted a post hoc robustness analysis of comments submitted under an additional eight articles identified by a structured search strategy of four news databanks (NewsBank Access World News, LexisNexis Universe News, ProQuest US Major Dailies, and NewsBank America's News Magazines) from 2010 to present. Via review of >200 additional reader-submitted comments, we assessed whether themes identified in comments submitted under the article in The Guardian were shared in similar comments sections under other popular media articles about POCD. This post hoc analysis is presented in more detail in online-only Supplementary material.

Results

Over a period of 13 days, 84 user comments were submitted under 67 unique usernames. In our initial inductive analysis, we distinguished comments that described anecdotal experiences of POCD from comments that expressed attitudes and beliefs about POCD. Ongoing analysis, however, caused us to reconsider that dichotomous categorisation. Some attitudes and beliefs responded to article content, but others referenced anecdotal experience. In many comments, content related to article content and objective data proved inextricably intertwined with descriptions of subjective or anecdotal experience. Our final themes, therefore, do not explicitly distinguish between anecdotal and non-anecdotal derivation, although some themes (e.g. functional impact) were substantially more accessible via anecdote. In our results, we include brief representative content in the text. We present full quotes in Table 1.

Table 1.

Representative quotes, by theme and sub-theme. Comments have been assigned numbers to identify whether selected excerpts come from the same submitted comment; they are numbered in the order of being listed in the table. POCD, postoperative cognitive dysfunction.

Themes	Comments
Functional impact
Deficits were described in functional terms	1: ‘Couldn't work, concentration was woeful, even reading was a struggle’.
	2: ‘I really cannot multi-task anymore’.
	3: ‘I found I could no longer read … I have trouble building up an overview of the argument being presented. My mind seems to miss key pieces of information as I go’.
Deficits were sometimes subtle but subjectively noticeable	4: ‘Until the operation I had a 360∗ memory, which could retain the multi-layered information & awareness [professionals] require … I can retain the primary info well enough, but I find I lack the circuitry to process the secondary or contradictory without effort … In all other regards my career continues, but I have to make conscious mental notes instead of being able to recall them effortlessly’.
Deficits observed in others could be extreme	5: ‘ … lost his ability to make new memories … [can't] hold meaningful conversations … take medication, feed himself, wash himself’.
Deficits do not always occur	6: ‘It was very successful … her severe pain is gone and she is very happy about that’.
	7: ‘I have had procedures involving general anaesthetic before and sailed through the experience’.
Emotional/psychiatric impact
Sense of loss with cognitive change	Cognitive symptoms were accompanied by pejorative terms: (1) ‘woeful’, ‘struggle’, (8) ‘foggy mess’.
	9: ‘My own light dimmed … I was diminished intellectually’.
	10: ‘I just want my brain back, it was a great brain, and I miss it’.
Psychiatric symptoms as part of POCD	11: ‘He recovered well, but his personality changed markedly, becoming less tolerant and had great spells of anger’.
	12: ‘I was like a different person when I came out, I had a strange sort of depression and I cried for hours every day … this quiet voice that was saying something isn't right, and a much louder one which said nothing mattered any more’.
	13: ‘I was prepared for the physical pain, but I did not know, that my whole psyche would feel different, too. It was like everything was painted in a darker color, as if I had a weaker shell than I usually had and therefore was threatened in a way I usually was not … To blame for our missing knowledge about this, is this stupid division, that we have into physical and psychological’.
Psychiatric symptoms may resolve with appropriate treatment	14: ‘He had depression afterwards, which was only treated after about 2 years and he was back to normal after another 12 months or so … we really struggled with this at the time and he retired early. If we'd understood this possibility it would have been easier to understand and take action sooner’.
Perception of causes
Anaesthetics are implicated	15: ‘When we still don't really know how general anaesthetics work it's hardly surprising that their side effects aren't fully understood. What is surprising is that people still opt for elective surgeries that require anaesthesia’.
	16: ‘My suspicion was that it was the particular drugs they used in the anaesthesia’.
	8: ‘After a really long op (nearly 4 hours) with a spinal anaesthetic … I felt mentally fine ….With a much simpler op (probably less than an hour) done under a general anaesthetic I was a foggy mess for a couple of days afterwards’.
	17: ‘This 2007 paper states that the inhaled general anaesthetic, isoflurane, induces a vicious cycle of apoptosis (cell death) … ’
Surgical trauma is inseparable	18: ‘What the primitive parts of the brain know is that you have been poisoned unconscious and cut open’.
	19: ‘Major surgery is a traumatic event, for both the body and mind’.
Role of other facets of hospitalisation	20: ‘My experience is that ‘granny’ with her broken hip ends up often immobile in a hospital bed for an extended period … Add to that the emotional stress of a strange environment and an absence of mental stimulation, not to mention the far-from-nutritious hospital food, and it's little wonder there's post-operative physical and cognitive decline’.
Counter-perspective: POCD after medical hospitalisation	21: ‘She achieved the brain fade noticed after having an operation, but without the actual operation’.
Inadequate response by medical providers
Healthcare providers may be dismissive	22: ‘I knew something was different post surgery, but it has taken years of saying something to doctors for someone to listen to me’.
Sense of abandonment	23: ‘I lost three days which was met with a shrug of the shoulders from my surgeon and cardiologist … [I was] left to fend for myself [for symptoms lasting at least a year]’.
Missed opportunity to discuss	14: ‘Before surgery, there was a half day course for the patient and caregivers about what to expect … Nothing was said about the potential effect on the brain’.
	23: ‘I would have benefited by being warned about pump head and had postoperation counselling’.
Doctors may not be aware of the risks	24: ‘We do not blame the doctors for that [hastening the onset of Pick's disease]. This phenomenon simply wasn't on their radar at the time’.
Importance of sharing information
Articles are validating	9: ‘This article helps my experience feel more real. Thank you’.
Anaesthesia and surgery are more dangerous than we appreciate	1: ‘[my previous POCD symptoms] would certainly make me think long and hard about having surgery in the future. I think it would have to be a life or death situation to persuade me to go under again’.
Risks must be balanced with anticipated benefits	25: ‘ … of course for many surgeries the risk of not proceeding far outweighs the risk of surgery itself’.
Media sensationalism may drive skewed reporting	26: ‘ … it's good to be very wary of these articles, esp when research is so new. Beware clickbait’.
Research in this area matters	5: ‘I feel it's too late for Dad, but I find some comfort knowing that others may be able to benefit from the preventative measures discussed’.

Functional impact

Individual anecdotes centred the experience of POCD in functional terms. Dysfunction in elements of daily cognitive load – ‘cannot multi-task’, ‘couldn't work’, ‘I could no longer read’ – were commonly cited, indicating that symptoms thought to be consistent with POCD are sometimes truly disabling. Commenters described specific skills, such as ‘a 360∗ memory’ or ‘a partially photographic memory’, they possessed preoperatively which were lost; one comment implied that, with effort, their cognitive loss could be compensated for by deliberate effort (‘I have to make conscious mental notes’). Functional impact described in anecdotes about a loved one could be extreme and severe, including deep impacts to formerly independent activities of daily living such as eating and bathing which would, as described, meet criteria for major neurocognitive disorder (formerly dementia) as defined in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V). Thus, a spectrum of functional change from subtle deficits which could be deliberately compensated for, to frank dementia, was felt to be consistent with POCD. In contrast, several comments were submitted which denied subjective cognitive impact after surgical exposures (‘sailed through the experience’).

Emotional/psychiatric impact

The emotional experience of POCD could include both an individual's emotional response to their observed deficits and new problems with psychiatric health which occurred after surgery. Commenters chose terms with negative emotional valence, such as ‘struggle’, ‘woeful’, and ‘diminished’ to describe their cognitive experiences, and felt a sense of explicit loss, as in the titular quote. In addition to emotional experiences of the described cognitive deficits, several anecdotes also listed symptoms of psychiatric disease, such as ‘spells of great anger’ and ‘a strange sort of depression’, as part of the POCD symptom complex. These psychiatric experiences were highly distressing, and as in the case of one who ‘cried for hours every day’, probably also functionally impactful.

Perception of causes

Both anecdote and non-anecdote comments addressed potential causes of POCD. Many comments attributed POCD symptoms explicitly or obliquely to general anaesthetic medications or their ‘side effects’, particularly those who experienced cognitive changes after procedures perceived as ‘not a particularly complicated or invasive surgery’. The experience of recovering from a general anaesthetic, a ‘foggy mess’, was contrasted negatively with the ‘mentally fine’ recovery from neuraxial anaesthesia. One comment cited scientific justification for the direct causal role of isoflurane in neuronal apoptosis. However, despite the plurality of comments implicating anaesthesia, the inseparability of surgical trauma was also acknowledged, as ‘major surgery is a traumatic event, for both the body and mind’. Furthermore, several comments described other facets of surgical hospitalisation, such as ‘far-from-nutritious food’, ‘absence of mental stimulation’, and being ‘immobile in a hospital bed’, which could have been implicated in POCD symptoms. One comment observed cognitive changes consistent with the POCD symptom complex which occurred in her mother after a medical hospitalisation (i.e. with no anaesthetic medications used), offering a counterexample to those who attributed symptoms to general anaesthetic medications.

Inadequate response by medical providers

Several commenters with personal or close experience with POCD symptoms described dismissive encounters with healthcare providers (‘it has taken years of saying something to doctors for someone to listen to me’) leading to feelings of abandonment and ‘being left to fend for myself’. There were important missed opportunities to improve care for POCD sufferers; providers, for example could have discussed the risk of cognitive symptoms during a ‘half-day course’ provided as preoperative education. Instead, the commenter's loved one was left with years of adverse symptoms, and the commenter felt that ‘if we'd understood this possibility it would have been easier to … take action sooner’.

Importance of sharing information

Because healthcare providers provide inadequate support, reading a popular media article about POCD may therefore be validating and help one's ‘experience [with POCD] feel more real’. Some people with personal experience with the POCD symptom complex submitted strongly worded comments emphasising global caution when considering anaesthesia and surgery, such as ‘it would have to be a life or death situation to persuade me to go under again’. This was tempered by a counter-perspective recognising the importance of surgery for correcting serious conditions, and others warning about sensationalism in the media (‘clickbait’) as a prompt for articles such as this. Commenters addressed the human importance of additional research into causes and preventive strategies, for example ‘I feel it's too late for Dad, but I find some comfort knowing that others may be able to benefit’. Articles such as this one, therefore, serve to disseminate lay information about a perioperative complication viewed as highly clinically significant by the commenters, which is underappreciated by the medical system, and which may be amenable to better management via, at a minimum, effective communication.

Post hoc robustness analysis

Comments submitted under eight unique popular media articles identified by a structured search strategy offered additional support for the themes described above (Supplementary material).

Discussion

Qualitative analysis of reader comments submitted under a popular media article about POCD offers insights into the lay perspective on this syndrome. Taken together, the comments suggest that POCD is perceived as a functional and emotional syndrome which is underappreciated or outright dismissed by healthcare providers, and that consciousness-preserving, rather than general, anaesthetic techniques are implicated in many instances. Counter-perspectives include appreciation among many commenters that other potentially cognitively impactful events – including surgery, surgical complications, and sleep disruption, lack of stimulation, and changes in routine and diet as part of hospitalisation – occur and may be inseparable from anaesthetic medications, and that articles in the popular media have an important role in validating the experience of, spreading information about, and improving care for patients experiencing or at risk for POCD. These themes were transferable to other user-submitted comments on popular media POCD articles.

There have been specialty-wide calls among anaesthesiologists to discuss postoperative neurocognitive disorders as a potential outcome of anaesthesia and surgery for older adults^8,9; however, adherence to this guidance is poor.¹⁰ We show that functional and emotional/psychiatric impact of POCD is critical to the understanding of anecdotes described by laypeople. Functional and emotional elements are not well captured by neuropsychiatric testing-based definitions used to diagnose POCD before the 2018 redefinition,³ which may exacerbate a disconnect between a clinician's need to describe POCD according to neuropsychiatric definitions, and the patient's intrinsically functional and emotional experience of the syndrome. However, in the absence of concrete discussion, or after outright dismissal, by surgery or anaesthesia providers, patients are unprepared for adverse cognitive symptoms and have difficulty accessing guidance for seeking further care. Strategies to bridge the gap between what is known about postoperative neurocognitive disorders, and how and by whom they are discussed with patients, should be the focus of further study.

There remains a strong perception among those who submitted comments that general anaesthesia is plausibly implicated in, and alternative anaesthesia strategies plausibly protective against, adverse cognitive recovery experiences. Nonetheless, recent large randomised trials^11,12 have found no difference in delirium rates between spinal with or without sedation vs general anaesthesia in hip fracture surgery for older adults. Long-term neurocognitive outcomes remain understudied, and lack of group-average difference does not preclude some individuals from experiencing outsize cognitive decline,¹³ potentially in response to a given anaesthesia technique; identifying these individuals must be, and is, a focus of ongoing work. Anecdotes here offer descriptions of subjectively better neurocognitive recovery after spinal anaesthesia, which must be reconciled with the scientific understanding to present a unified perspective which accurately captures areas of certainty and uncertainty.

Qualitative analysis is often used as a first step when no data exist on a phenomenon; however, it has significant limitations. First, qualitative analysis is inherently not generalisable, and this analysis was further limited because comment submission was voluntary (i.e. not derived from a purposive sample). Individuals who comment on news articles typically represent a minority of readers, overrepresenting those who have a particular interest in the topic.^14,15 Responses therefore are neither a random nor representative sample of popular views on this topic; they may be most narrowly thought of as a sample of extreme views which is, nonetheless, easily accessible to interested prospective surgical patients. There is no way to ascertain demographic or other details about those who chose to respond, so it is unclear to what population these responses may be transferrable. We cannot establish whether the subjectively reported cognitive deficits were objectively measurable (e.g. with neuropsychological testing); the relationship between subjective and objective cognitive decline after surgery is known to be unreliable.¹⁶ Finally, responses to only one article on POCD were formally considered. Post hoc analysis suggests the themes were robust to article source but not exhaustive; it is not known how the content in the index article may affect who responds, and in what way. Further work – for example leveraging purposive sampling strategies to obtain a broader theoretical understanding; interviewer-initiated probing questions to elicit details which are here underexplored; more details about the surgery, anaesthetic, and postoperative course which are inherently interesting to clinicians and may have implications for the ‘lived experience’ of POCD – is necessary to advance our understanding of patient experiences of POCD.

Scientific journalism has an important role in disseminating scientific findings to the public,¹⁷ and conversely, examining public reaction to popular media articles may offer novel insights on public relevance of scientific findings. Qualitative thematic analysis of comments submitted in response to a popular media article on POCD offers insight into public perception of symptom complexes, causes, and challenges the medical community may face when discussing this syndrome with patients and their care partners. Public messaging about plausible causes of POCD and supporting perioperative healthcare providers' role in recognising and validating the symptom complex offer areas to target for potential improvement.

Authors’ contributions

Conception and design of the study: all authors.

Analysis and interpretation of data: all authors.

Acquisition of data: ELW.

Drafting of the initial version of the article: LL, ELW.

All authors were involved in critical revision of the manuscript and approval of the final published version, and agreed to be accountable for all aspects of the work.

Declaration of interest

ELW is a member of the associate editorial board of the British Journal of Anaesthesia. The other authors have no potential conflicts of interest to declare.

Funding

National Institute on Aging of the US National Institutes of Health (grant numbers K07AG066814 to DD; K24AG068312 to AKS; and R01AG079263 to ELW); and the UCSF Department of Anesthesia & Perioperative Care. The study sponsors had no role in the design, analysis, interpretation of data, writing, or decision to submit for publication.

Handling editor: Rupert Pearse

Appendix A. Supplementary data

The following is the Supplementary data to this article.