There is increasing concern about delirium in the context of surgical interventions. Dr. Harman Chaudhry, a fellow in orthopedic surgery, and Dr. Chris Patterson, a geriatrician, discuss the seminal issues pertinent to evidence-informed decision-making.
A typical clinical scenario could be the following:
You are a hospital physician attending the orthopedic ward at a community hospital when you receive a call from the daughter of one of your patients. The daughter is fearful of her mother’s state of mind following tomorrow’s surgery to repair her mother’s hip fracture. Previously the family had their father experience delirium following a surgical procedure. The daughter wants to know if there is anything that can be done to prevent this from occurring with their mother.
These two clinicians consider the evidence in the context of four key issues related to efficacy of:
- routine screening for risk factors and the diagnostic accuracy of screening tools;
- strategies to reduce post-operative delirium risk;
- interventions to manage delirium post surgically; and
- strategies to enhance the detection of delirium
Drs. Chaudhry and Patterson have a friendly debate about the practical as well as research issues that can help you address the family concerns about your patient in this clinical scenario.
Orthopedic surgeons (and perioperative physicians) are concerned about post-operative success following surgical management of hip fractures. Up to 50% of patients will develop delirium following a hip fracture. The short-term consequences of delirium include agitation interfering with care, pulling out lines, climbing out of bed, resistance to care and longer hospitalization. There is also evidence that these patients are more likely to develop long-term cognitive impairment and decline in function, and have a higher rate of institutionalization and death . We recognize certain patient characteristics that lead more frequently to the development of delirium than others. Studies show that characteristics such as older age, pre-existing cognitive impairment, and polypharmacy increase a patient’s risk of delirium .
Should we routinely and systematically look for these characteristics?
Although all clinicians, on some level, make a clinical judgment regarding a patient’s post-operative course, it is less commonly done in a systematic and valid manner.
It is important that a general exploration of delirium risk factors is made during a patient’s pre-operative medical history, as you mention. Although there is no evidence that this directly improves outcomes, inquiring about these known risk factors can inform pre-operative consultation with medical or geriatric specialists, facilitate post-operative care and discharge planning, as well as allow for accurate disclosure of risk to the patient and family.
Known predisposing factors include:
- age over 75 years;
- re-existing cognitive deficits (especially dementia);
- poor functional status;
- sensory deficits (vision, hearing); and
- polypharmacy and excessive alcohol intake.
There are a number of short validated instruments that can be administered by the surgeon or other members of the healthcare team, and that can facilitate this assessment, such as the DEAR instrument  and the model proposed by Kalisvaart et al .
In each of these scales, increasing number of predisposing factors raises the risk of post-operative delirium.
Having identified those patients at high risk for delirium, what preventive measures can be taken to reduce the risk of post-operative delirium in both the elective and non-elective surgical patient?
I am aware of several randomized clinical trials evaluating potential pharmacologic interventions to prevent delirium in hip fracture and other hospitalized patients. Obviously, such “magic bullet” preventive agents would be the ideal solution, although I recognize these are few and far between.
However, one promising pharmacologic is haloperidol. A recent trial demonstrated that it significantly reduced the incidence of delirium in patients undergoing noncardiac surgery , while a previous trial demonstrated a reduction in delirium duration and severity among hip surgery patients .
Melatonin is another such agent that has been postulated to reduce the incidence by acting, at least partially, to normalize the sleep-wake cycle. In medical inpatients, melatonin was found to reduce the incidence of delirium from 31% to 12% (p=0.014) . Unfortunately, a recent randomized trial conducted in the Netherlands was unable to demonstrate a significant benefit of its use in hip fracture patients .
Most orthopedic surgeons have been cautious in the uptake of this evidence, and this is not widespread practice. However, is routine haloperidol prophylaxis something that should be routinely implemented for hip fracture patients and others at high risk for delirium?
Strategies to prevent delirium fall into two major categories: pharmacological and non-pharmacological.
Various pharmacological strategies have been shown to reduce post-operative delirium. A systematic review by Zhang and colleagues summarized the evidence, although the review did not focus specifically on hip fracture surgery .
- Dexmedetomidine sedation was preferable to other drugs.
- Typical and atypical antipsychotics were more effective than placebos.
- Despite “common wisdom”, there was no difference between general and spinal anesthetics.
In terms of haloperidol, I agree that it is a promising agent in the prevention of delirium. However, I would be cautious and tempered in its use. First of all, only a single trial has definitively demonstrated a reduction in delirium incidence. Because haloperidol has a potential for adverse events, especially if used at high doses, I would not advocate for widespread dissemination of this practice until the results are reproduced in multiple trials, and confirmed through meta-analysis of these results.
In terms of melatonin, despite the negative findings of a single trial, I would certainly not write off melatonin completely. Melatonin is a relatively benign agent, and if it were effective it would be a great agent to use routinely. Further evidence is needed.
In my opinion the jury is still out on the use of effective pharmacologic intervention. The evidence for non-pharmacologic intervention is far more convincing . The most commonly employed strategy is a multi-component preventative strategy consisting of optimizing management of acute medical conditions, correcting electrolyte abnormalities, orientation to time and place, medication reduction, and addressing pain and anxiety. Early involvement of a geriatric specialist, multi-disciplinary team, or admission to a dedicated geriatric unit has also been shown to be effective , likely through the implementation of the aforementioned strategies.
A cautious uptake of pharmacologic delirium prevention is certainly warranted, and the reasons you delineate are likely why these strategies have not been widely adopted. The non-pharmacologic strategies that you discuss result in no harm to the patient, and I agree that they should be widely implemented in all hip fracture wards and other wards where patients are at high risk for delirium.
Yet despite advances in preventive management, no single method or combination of methods has yet proven to be anywhere near 100% effective. Patients continue to develop delirium post-operatively. What are the key principles of management for delirium?
Unfortunately, there is no ‘cure’ for delirium, so this is a good example of where prevention is far better than treatment. Management is largely supportive.
In general, the strategy involves three principles:
- avoid things which make it worse;
- address the precipitating cause(s); and
- prescribe specific measures to control agitation, aggression, hallucinations and other psychotic features.
Numerous medications can aggravate or cause delirium. Many drugs have anticholinergic effects, and should be avoided if at all possible. Examples include dimenhydrinate/Gravol and certain opiates (e.g. meperidine/Demerol). A good list can be found at: http://www.ncbi.nlm.nih.gov/pubmed/26446832.
Acute medical conditions, which may have precipitated the delirium, should be treated promptly. Infections, heart failure, metabolic disturbances and electrolyte abnormalities are among the most frequent. The patient’s surrounding environment should also be optimized to provide for a calm and reassuring atmosphere. Use of physical restraints should be avoided. Family members or other familiar people at the bedside may help to re-orient the patient.
Although there is limited evidence for pharmacologic interventions, it may be necessary to sedate individuals with hyperactive or mixed delirium. Most experts recommend small doses of antipsychotics with a lower potential for anticholinergic side effects (e.g. haloperidol, risperidone). It is best to avoid benzodiazepines (BZP) except when the delirium is due to alcohol or BZP withdrawal, or when the patient has Parkinson's disease, which may be aggravated by antipsychotics.
Consultation with a geriatrician, internist, or dedicated geriatric team should be considered to facilitate management.
Excellent summary, Dr. Patterson! Finally, I would like to discuss some of the difficulties in identifying patients with delirium. Despite frequent daily contacts, the diagnosis of delirium can be missed or overlooked entirely. This is a theme we find consistently in the literature; active monitoring of patients reveals rates of delirium which are much higher than those with passive monitoring . Delirium can present in one of three forms: hyperactive, hypoactive or mixed. Hyperactive delirium is the most familiar and identifiable: these patients are agitated, aggressive and restless. Up to 70% of cases actually present as a hypoactive subtype . These patients are lethargic and withdrawn, and may not be identified without directed case finding. A mixed subtype presents with characteristics of both hyperactive and hypoactive delirium.
Missed diagnoses can lead to patients missing out on important management interventions and, in some cases, being discharged home or back to a nursing home while still suffering from delirium. We need a way to accurately identify patients with delirium despite busy clinician schedules, suboptimal nurse-to-patient ratios, and resource limitations in general.
The most reliable and validated instrument available for delirium screening is the Confusion Assessment Method (CAM) instrument . This is a highly sensitive and specific questionnaire that takes approximately 10 minutes to administer at the bedside. Any health care personnel with appropriate training can administer the CAM. This includes orthopedic surgeons, internists, nurses and physiotherapists.
The CAM seeks 4 items:
- acute onset and fluctuating course; PLUS
- inattention; AND
- disorganized thinking
OR decreased level of consciousness.
Active daily monitoring of patients with the CAM will identify almost all patients with delirium. However, whether increased monitoring and identification translates into better patient outcomes has not been demonstrated definitively in the literature.