Long term cognitive dysfunction in older people after non-cardiac surgery

Half of all people reaching the age of 65 subsequently have one or more operations,¹ but despite substantial research on short term cognitive dysfunction within the first week after the operation little research has been undertaken into the potential long term effects on cognition. The exception is cardiac surgery, where cognitive dysfunction has been well documented and has usually been attributed to the adverse effects of cardiopulmonary bypass on the brain.²,³ Various risk factors for long term (defined as three months or more) postoperative cognitive dysfunction have been investigated, including type of anaesthetic agent, general versus regional anaesthesia, use of anticholinergic agents such as atropine, or the physiological effects of the anaesthetic such as hypoxia, hypotension, or hyper-ventilation.

The multicentre International Study of Post-Operative Cognitive Dysfunction (ISPOCD1) compared 1218 patients aged 60 years and older undergoing major surgery with a control group (n = 321) of a similar age.⁴ At three months after the operation cognitive dysfunction was found in 9.9% (94/1218) of patients compared with 2.8% (5/321) of controls. However, when a subset was re-tested one to two years later, 10.4% (35) showed cognitive dysfunction compared with 10.6% (5) of controls, which implies no long term effects, although the small control group (n = 47) may have been inadequate.⁵ No relation was found with hypoxaemia or hypotension, but a logistic regression analysis showed that higher age (P = 0.002), infectious complications in the postoperative period (P = 0.045), and cognitive dysfunction one week after surgery (P = 0.006), but not at three months, were significant risk factors for long term cognitive dysfunction one to two years after surgery.

Several studies involving patients who had undergone orthopaedic surgery have been carried out. Jones et al studied 146 patients aged over 60 years, who were randomly allocated to general anaesthesia or regional anaesthesia groups and compared with a control group of patients on the waiting list for surgery.⁶ The results showed no cognitive dysfunction after three months and no statistically significant differences between the groups.

Williams-Russo et al conducted a randomised prospective study of epidural versus general anaesthesia on the incidence of long term cognitive dysfunction in 262 adults (134 receiving epidural anaesthesia, 128 general anaesthesia) aged over 40 years (mean age 69 years).⁷ At six months after surgery cognitive dysfunction was found in 6% (7) of the epidural group compared with 4% (5) in the general anaesthesia group.

Ancelin et al investigated the incidence of cognitive dysfunction in 140 people over the age of 64.⁸ At three months 56% (78) had notable deterioration of more than one standard deviation on one or more of the test scores. However, given the large number of cognitive tests used, the likelihood of type 2 errors occurring was increased. Those showing the greatest degree of deterioration tended to be the most elderly patients, those with the lowest educational level, and those with a history of cognitive decline before surgery. Nevertheless, both the Williams-Russo and the Ancelin studies did not include a control group comprising patients who had no surgery.

Several possible explanations exist for why such different outcomes have been seen. All the studies used different measures for cognitive assessment, and the measures used by Jones et al⁶ may be less sensitive to cognitive change than those used in the other studies. Also, the ISPOCD1 study found no difference between the control and surgery groups after one to two years.

Scant evidence exists about what may contribute to long term postoperative cognitive dysfunction even if it does exist. The two studies comparing general and epidural anaesthesia found no difference in outcome,⁶,⁷ and the ISPOCD1 study found no link between long term cognitive dysfunction and either hypoxaemia or hypotension.⁴ Four out of the five studies found that increasing age was a statistically significant risk factor in the development of long term postoperative cognitive dysfunction. However, higher age also increases the risk of developing dementia, emphasising the need for studies with adequate control groups. Other factors included a low educational level, a history of cognitive dysfunction before surgery, and cognitive dysfunction at one week after surgery. Even so, most patients showing cognitive dysfunction after one week recovered after several months. The only indication of a possible preventive measure would be to reduce postoperative infection rates in surgical wards, and such measures are already standard practice.

Whether or not major surgery or general anaesthesia increases the risks of long term cognitive dysfunction remains unclear. The research so far has had methodological problems, and so it is not possible to draw conclusions. Future research needs to include validated, reliable, and sensitive cognitive assessments and well matched control groups to take into account the possible influences of disability, pain, and depression on cognitive function. Until such studies have been conducted and sufficient evidence is available it will be difficult to provide older patients with informed advice about the potential long term risks of surgery.