Author Affiliations: Geriatric Research, Education and Clinical Center (Drs Vasilevskis and Ely) and Clinical Research Training Center of Excellence (Dr Vasilevskis), Department of Veterans Affairs Medical Center, Tennessee Valley Healthcare System, and Department of Medicine (Drs Vasilevskis and Ely), Center for Health Services Research and Division of General Internal Medicine and Public Health (Dr Vasilevskis), and Division of Allergy, Pulmonary, and Critical Care Medicine (Dr Ely), Vanderbilt University, Nashville, Tennessee.
We in medicine have long wondered if a period spent in delirium during a hospitalization leads to a change in the trajectory of cognitive decline (a phenomenon any busy clinician would vouch he or she has seen many times) or if that accelerated decline is due to other comorbidities. Whereas many hospital-acquired conditions are considered short-term complications (eg, hospital-acquired pneumonia), we increasingly recognize that the effect of hospitalization on patients' functional and cognitive health may be prolonged or permanent.
The long-term effects of hospitalization on cognition are highlighted by a pair of recent studies. In a large prospective community cohort, Ehlenbach et al1 demonstrated a 40% increased risk of incident dementia up to 6 years following hospitalization for a noncritical illness (hazard ratio, 1.4; 95% CI, 1.1-1.7) and an even greater risk among the critically ill (hazard ratio, 2.3; 95% CI, 0.9-5.7). In another cohort of community-dwelling older patients, Wilson et al2 showed that hospitalization was independently associated with a 2.4-fold increase in the rate of global cognitive function decline.
Each study raised important questions about the mechanism by which hospitalizations lead to incident and accelerated cognitive decline. Sepsis, critical illness, and surgical procedures seem to have a role, but they do not tell the whole story.3,4 In each case, the development of delirium has been hypothesized to be an important hospital-acquired condition that may be a critical mediator of the subsequent acquisition of a long-term dementia-like disability. This has been recently supported by important studies5,6 that demonstrated independent relationships between hospital-acquired delirium and short-term and long-term cognitive impairments.
The study by Gross et al7 in this issue of the Archives of Internal Medicine seeks to highlight further the role of hospital-acquired delirium in the progression of cognitive decline. This study examines a unique cohort of 263 hospitalized patients with preexisting Alzheimer dementia. This cohort was followed up for a median of 3.2 years, during which 56.3% developed delirium during their index hospitalization. The authors subsequently found that the development of hospital-acquired delirium was independently associated with cognitive deterioration up to 5 years after hospitalization at a rate 2.2 times greater than that among patients without delirium.
A methodological challenge of studying the long-term effects of hospitalization is appropriately considering the cognitive status before hospitalization. Without prehospitalization information, one is at risk of erroneously concluding that hospitalization, or a hospital-acquired condition such as delirium, independently affects posthospitalization outcomes, when in fact the relationship may be alternatively explained by patients' prehospitalization cognitive function or trajectory. This study has gone to great lengths to overcome this bias by examining a cohort with frequent and robust measures of cognitive function before hospitalization. By limiting the population to those followed up for at least 3 clinic visits, the authors were able to adjust for baseline and trajectory of cognitive decline. As a result, significantly greater certainty exists that the hospitalization event was a major factor in the subsequent cognitive decline.
The authors' primary question was whether delirium was the hospital-acquired event that accounted for rapid progression of cognitive decline. This question is a more challenging one given the complex nature of hospitalization and the limitations of medical record–based diagnosis of delirium. Although delirium has sound theoretical basis as a risk factor for long-term cognitive impairment, there are other important factors not accounted for in this study that may confound the relationship. Factors, such as sepsis, hypotension, and polypharmacy, are a few alternative mechanisms that may contribute independently to long-term cognitive outcomes. Alternatively, it is possible that delirium is the condition through which these other complications may lead to progression of cognitive decline.
Another important consideration of the study surrounds delirium measurement. The authors use a medical record–based method that depends on symptom documentation by clinical providers.8 Notably, hospitalizations occurred across 46 different hospitals. This raises the possibility that hospital-level effects may be related both to the quality of medical record–based diagnosis of delirium and to posthospitalization cognitive outcomes. Most important, the medical record–based method of delirium diagnosis is of moderate sensitivity and specificity. Although the authors state that only 9% of patients would be at risk of a false-positive diagnosis, this applies to general inpatient populations. In a population of patients with dementia, the rate of false positivity is increased. Another important consideration is that 2 other factors associated with false-positive medical record–based diagnosis of delirium include severity of illness and delirium risk (derived in part using severity of illness). Both of these elements of prognosis must be better incorporated in future work.
Regardless of these limitations, the consistency of the study findings with prior research makes the implications clear. This study stands as a stark warning of the potential long-term dangers of hospitalization and delirium. Engel and Romano9 were equally concerned about these dangers in 1959. They wrote: “[N]ot only does the presence of delirium often complicate and render more difficult the treatment of a serious illness, but also it carries the serious possibility of permanent irreversible brain damage. . . . [T]he physician who is greatly concerned to protect the functional integrity of the heart, liver and kidneys of his patient has not yet learned to have similar regard for the functional integrity of the brain.”9(p261) The present study provides another example for increased regard of the “functional integrity of the brain.” This is critical for the growing population with Alzheimer disease, most of whom will be hospitalized at least once during their lifetime.
This study also adds to growing evidence that delirium is a hospital-acquired condition with substantial long-term consequences. We can no longer be blind to the presence of delirium and must implement validated measures that are reliably measured at the bedside.10 We should not wait for delirium to happen but must work to implement proven interventions that prevent delirium.11 Finally, we must continue to find ways to treat delirium that include focus on comorbid disease management, removal of offending medications, and environmental modifications. We must also develop ways to rehabilitate the brain following acute illnesses so as to mitigate the long-term cognitive decline associated with delirium.
Correspondence: Dr Vasilevskis, Department of Medicine, Vanderbilt University Medical Center, 1215 21st Ave S, Ste 6000 Medical Center East, North Tower, Nashville, TN 37232-8300 (firstname.lastname@example.org).
Published Online: August 20, 2012. doi:10.1001/archinternmed.2012.3414
Financial Disclosure: Dr Ely has received grant support and honoraria from Eli Lilly, Hospira, and Pfizer.
Funding/Support: Funding was provided by grant K23AG040157 (Dr Vasilevskis) and grants R01 AG035117-02 and R01 AG 027472-05 (Dr Ely) from the National Institutes of Health and the Veterans Affairs Clinical Research Center of Excellence and by the Tennessee Valley Healthcare System Geriatric Research, Education and Clinical Center (Drs Vasilevskis and Ely).
Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute on Aging, the National Institutes of Health, or the US Department of Veterans Affairs.
Vasilevskis EE, Ely EW. The Danger of DeliriumComment on “Delirium and Long-term Cognitive Trajectory Among Persons With Dementia”. Arch Intern Med. 2012;172(17):1331-1332. doi:10.1001/archinternmed.2012.3414