POCD: The Dirty Secret of Anaesthesiology?

An Essay on What Patients Are Never Told Before Surgery

One in four elderly patients undergoing major surgery under general anaesthesia will experience measurable cognitive decline within a week. One in ten will still have it three months later. These are not fringe estimates from alternative medicine circles. They come from a landmark peer-reviewed study published in The Lancet in 1998, conducted across thirteen hospitals in eight countries, involving 1,218 patients aged sixty and over.¹

The authors of that study — the International Study of Post-Operative Cognitive Dysfunction, known as ISPOCD1 — concluded that their findings “may have implications for the information given to elderly patients before surgery.”¹

That was 1998. Twenty-seven years later, the information still isn’t being given.

---

Support Independent Investigative Journalism and Research

This work remains free because paid subscribers make it possible. If you find value here, consider joining them.

What paid subscribers get: 1-2 free books per month. e.g.

The PSA Trap
Breast Cancer
Chlorine Dioxide
Drilling for Profit
What Your Vet Can’t Tell You
No Virus

Plus: Access to the Deep Dive Audio Library — 180+ in-depth discussions (30-50 min each) exploring the books behind these essays. New discussions regularly added. That’s 100+ hours of content for less than the price of a single audiobook.

[Upgrade to Paid – $5/month or $50/year]

unbekoming@outlook.com

---

What POCD Is

Post-Operative Cognitive Dysfunction is not the grogginess that follows anaesthesia and clears by the next morning. It is measurable, persistent neurological decline: slower processing, impaired memory, difficulty finding words, reduced concentration, motor clumsiness. In some patients it lasts weeks. In others, months. In a subset, it does not resolve.

The ISPOCD1 investigators were rigorous in how they measured it. They administered neuropsychological tests before surgery, then again at one week and three months post-operatively, comparing results against a matched control group. The definition required objective deterioration on standardised tests — not self-report, not clinical impression. By those standards, 25.8% of patients showed cognitive dysfunction at one week. At three months, 9.9% remained impaired, compared with 2.8% of controls.¹

The study also identified which patients were most at risk. Older age was the dominant risk factor for long-term impairment. A second operation carried an odds ratio of 2.7 for early cognitive dysfunction — meaning patients undergoing a second surgical procedure were nearly three times more likely to show cognitive decline at one week than those having a single procedure.¹

That figure matters for what happened to Sue’s person.

---

The Case That Shouldn’t Happen

Sue is a reader. She wrote because someone close to her underwent cataract surgery twice, one month apart, with general anaesthesia each time. After the second procedure, the cognitive symptoms began: slower thinking, difficulty finding words, dropping things, bumping into furniture.

Those symptoms were reported at follow-up. The cataract office said it would clear in a few days.

It didn’t clear. When she returned weeks later, she was told the symptoms had nothing to do with the surgeries. She was referred to a hospital for testing.

The hospital diagnosed a stroke — despite imaging that was not focal and symptoms that were not unilateral, the hallmarks of stroke presentation. The patient was placed on insulin, antibiotics, and anti-inflammatory steroids. She appeared to improve within a week.

Then she was put under anaesthesia a third time, for a biopsy. The testing that generated the stroke workup had also found small lesions in her brain and lung. These were characterised as cancer. She is now scheduled for radiation.

Map that sequence against the ISPOCD1 data. The study identified a second operation as a significant independent risk factor for early POCD, with an odds ratio of 2.7.¹ Sue’s person had two surgeries a month apart under general anaesthesia, then a third before the cognitive effects of the first two had resolved. The symptoms — word-finding difficulty, motor clumsiness, slowed thinking — are precisely the profile POCD produces. The non-focal, bilateral imaging findings are inconsistent with classic ischaemic stroke.

Whether the stroke diagnosis was wrong is a clinical question that requires the imaging. What is not a clinical question is this: the symptoms were reported, and the institution that caused them declined responsibility.

---

The Informed Consent Gap

Informed consent requires that a patient be told about risks that a reasonable person would want to know before deciding whether to proceed. The ISPOCD1 findings meet that standard without qualification. Nearly one in ten elderly surgical patients still cognitively impaired at three months is not a marginal risk. It is a material risk — the kind that might lead a patient to ask questions, seek a second opinion, request a different anaesthetic approach, or delay an elective procedure.

Cataract surgery is elective. Both eyes do not need to be done a month apart. An elderly patient told that a second operation within that timeframe carries an odds ratio of 2.7 for cognitive dysfunction might reasonably say: I’ll wait longer between procedures. Or I’ll do one eye and reassess.

That choice requires information. The information exists — it has existed since 1998 in a major peer-reviewed journal. It is not being given.

The ISPOCD1 authors themselves flagged this. Their conclusion states directly that the findings carry “implications for the information given to elderly patients before surgery.”¹ The Biedler replication, published the following year in Der Anaesthesist, confirmed the same prevalence rates and the same risk profile.² The research is not obscure, not contested, not preliminary. It is two-and-a-half decades old and sitting in the literature.

What is absent is not the evidence. What is absent is the disclosure.

---

Why There Is No Baseline

The primary tool for making POCD invisible is the absence of a pre-operative cognitive baseline. No baseline means no before. No before means no measurable after. No measurable after means no accountability.

This is not an oversight. Neuropsychological screening tools exist that take fifteen minutes to administer. The ISPOCD1 investigators used them across thirteen hospitals in eight countries without difficulty. The tools are not expensive. The logistics are not complex. The reason they are not used in standard surgical practice is not that anyone forgot — it is that a baseline serves the patient and only the patient.

A pre-operative cognitive score followed by a post-operative cognitive score is a document. It makes decline visible. It makes decline attributable. It converts a patient’s subjective complaint — “I’m not thinking clearly since the surgery” — into an objective, dated, signed record of measurable deterioration. That record has legal weight. That record creates liability.

Without it, the system operates exactly as it does in Sue’s case. The patient reports symptoms. The surgeon identifies no surgical complication. The anaesthesiologist has left the building. The GP has nothing to compare against. The symptoms migrate through the diagnostic system looking for a home — landing, in this case, on stroke — while the most obvious candidate goes unnamed.

Anaesthesiology created this blind spot and maintains it. The absence of baselines is not a gap waiting to be filled. It is a wall, built from the consistent absence of incentive to see what is on the other side.

If you or someone close to you is scheduled for surgery under general anaesthesia, request a cognitive baseline assessment before the procedure. Ask your GP to administer the Montreal Cognitive Assessment (MoCA) or Mini-Mental State Examination (MMSE) and place it in your medical record dated before surgery. If the surgical team won’t provide it, get it done independently. A dated pre-operative score is the one tool that makes post-operative cognitive decline impossible to dismiss — which is precisely why the system doesn’t offer it.

---

The Cascade

The misattribution of POCD symptoms to stroke did not end with a corrected diagnosis. It generated a treatment cascade.

A stroke diagnosis triggers a protocol: imaging, blood work, specialist referral, medication. Sue’s person was placed on insulin, antibiotics, and steroids — each carrying its own risk profile, each prescribed in response to a diagnosis that her presentation did not cleanly support. Non-focal, bilateral cognitive symptoms following two surgeries under general anaesthesia, occurring in an elderly patient within the ISPOCD1 risk window, is a presentation that should prompt POCD as the primary differential. It was not named.

The imaging done to investigate the stroke found lesions. Those lesions, characterised as cancer, generated a biopsy. The biopsy required a third anaesthetic — administered to a patient whose cognitive function had not recovered from the first two exposures. And the biopsy findings are now driving a radiation treatment plan.

Each step in this sequence follows logically from the one before it. Each step was initiated by a system that had already declined to name the most likely cause of the original symptoms. The cascade did not begin with an absence of the relevant differential. Before the third surgery, Sue raised POCD directly with the physician — by name, while reviewing the imaging together. He dismissed it and proceeded. The third anaesthetic was administered to a patient whose cognitive function had not recovered from the first two, over the explicit objection of the person who knew her best. What followed was not a system that failed to consider the right diagnosis. It was a system that considered it, rejected it, and continued.

---

The “Routine Surgery” Problem

Cataract surgery is one of the most common elective procedures performed in the developed world, with the vast majority of patients aged sixty-five and over. It is routinely described as low-risk, quick, and safe. That description is not false — for the eye. For the aging brain under repeated general anaesthesia, the risk profile is materially different, and the ISPOCD1 data applies directly.

The “routine” framing does specific damage. It displaces the patient’s risk calculus before the conversation begins. A patient who has been told their procedure is routine is not primed to ask about cognitive side effects, and a clinician operating within that framing has no obvious prompt to raise them. The consent form reflects the framing: infection risk, bleeding risk, anaesthetic reaction — the acute and dramatic risks that can be managed in the surgical suite. Cognitive decline that manifests over weeks and months, in the organ the surgeon never touches, is outside the frame entirely.

This is not an argument against cataract surgery. It is an argument that “routine” is doing work it should not be doing — suppressing a disclosure that the evidence has required since 1998.

---

What the Researchers Said in 1998

The ISPOCD1 paper closes its discussion section with this sentence: “In clinical practice, the findings may have implications for the information given to elderly patients before surgery.”¹

That sentence is careful, hedged in the way academic medicine hedges. But the implication is not ambiguous. A study finding that nearly one in ten elderly surgical patients remains cognitively impaired three months after surgery, with age as the dominant independent risk factor and a second operation carrying an odds ratio of 2.7, is a study whose authors understood that something needed to change in how patients are informed.

It didn’t change. The literature continued to accumulate. The consent forms did not.

The question of whether this constitutes a failure of the profession, a failure of regulatory oversight, or a failure of institutional incentives that make disclosure inconvenient is worth asking. What is not in question is that the gap between what is known and what is told is not the result of ignorance. The knowledge is twenty-seven years old and published in The Lancet.

---

What Should Happen

Pre-operative cognitive baseline testing in patients over sixty-five is feasible, inexpensive relative to surgical costs, and essential for detecting post-operative decline. It does not require new technology. It requires will.

Informed consent for elderly patients undergoing surgery under general anaesthesia should name POCD explicitly — its prevalence, its duration range, the specific risk carried by repeat procedures within short intervals, and the absence of any known preventive or therapeutic intervention. The ISPOCD1 authors noted that no prophylactic or therapeutic factors could be identified.¹ That absence of treatment is itself material information. A patient who knows that if POCD occurs there is nothing medicine can currently do to accelerate recovery may weigh an elective procedure differently than a patient who assumes that any problem will be manageable.

When post-operative cognitive symptoms are reported, POCD should be named as a primary differential, not a diagnosis of exclusion reached only after other conditions have been ruled out and additional procedures have been undertaken.

None of this is radical. All of it follows directly from research that has been in the mainstream literature since 1998.

---

The Dirty Secret

“Dirty secret” implies something hidden. POCD is not hidden in the research literature — it is hiding in plain sight, documented and confirmed and unacted upon. The secret is not that the risk exists. The secret is that the risk exists, is known, is quantified, and is not being disclosed to the patients who bear it.

Sue’s person did not know, before her first surgery, that one in four elderly patients will experience measurable cognitive decline within a week. She did not know that a second operation would nearly triple her risk of early cognitive dysfunction. She did not know that if those symptoms appeared, the institution that performed the surgery would be unlikely to name the cause. She did not know because no one told her — not because the information didn’t exist, but because the information wasn’t given.

Whatever word you choose for that — omission, negligence, institutional self-protection — the data is published and has been for twenty-seven years. It says what it says.

---

References

¹ Moller JT, Cluitmans P, Rasmussen LS, et al. Long-term postoperative cognitive dysfunction in the elderly: ISPOCD1 study. The Lancet. 1998;351(9106):857–861.

² Biedler A, Juckenhöfel S, Larsen R, et al. Postoperative Disorders of Cognitive Performance in Elderly Patients: The Results of the International Study of Postoperative Cognitive Dysfunction (ISPOCD 1). Der Anaesthesist. 1999;48(12):884–895.

---

Postscript — March 2026

After this essay was published, Sue wrote again. The sequence had continued.

Following the stroke diagnosis and biopsy, the patient underwent ten rounds of brain radiation, then a fourth surgery — a port placement in preparation for thirty additional rounds of lung radiation and six concurrent rounds of chemotherapy. Four surgeries in four months.

After the fourth procedure she fell. The cognitive symptoms that had persisted since the first two cataract surgeries went into freefall. She lost the ability to walk, to hold a conversation, to track a page of a catalogue. She could not recall her birthday or her social security number. She was eating one bite of toast a day.

Before the third surgery, Sue had raised POCD with the treating physician directly, while looking at the imaging. He insisted on stroke and dismissed the concern. The patient and her partner trusted the doctor.

She is now in hospice.

At the time of writing she has shown some recovery — she can state her birthday again, find some words, communicate in fragments. Whether she recovers further is not known. What is known is the sequence: two elective cataract surgeries, a misattributed diagnosis, a third anaesthetic administered after POCD was named and dismissed, a treatment cascade that generated a fourth, and an outcome that the ISPOCD1 data, published in 1998, had the statistical architecture to predict.

The gap between what is known and what is disclosed is not abstract. This is what it produces.