Frailty in the ICU
[posted earlier on old neuroccm wordpress site]
It is no secret that ICUs across the world are admitting a greater proportion of older patients – some of these patients have poor baseline functional status. Emerging fields such as geriatric trauma in the critically ill are a testament to this. Some would say that the concept of a frail patient is one that applies only to older (geriatric) patients – but this is not necessarily true. The reason for this is that the frail individual can be defined as “any patient with diminished physiologic and cognitive reserve”. This is a well recognized concept in Geriatrics and therefore work has been done to identify and quantify the degree of frailty in that population of patients.
Broadly speaking, frail patients have a higher risk of falls, illness, unplanned hospital admissions, complications after surgery, disability and death. In the older frail patient, a major illness can be a catalyst for decline into a “new baseline” of function.
Therefore, this raises the notion that we should consider frail patients, and frailty as an entity, as part of our comprehensive approach to defining co-morbid risk factors in ICU patients. At present, frailty is not well captured by common critical care admission scores…and maybe it’s time that we consider this. How should we be quantifying this?
To this effect a recent Canadian study by Bagshaw et al. titled “Association between frailty and short- and long-term outcomes among critically ill patients: a multicentre prospective cohort study” characterized the prevalence of frailty and used a well-established frailty score (developed originally in the outpatient setting) to look at outcomes in critically ill patients.
Study Highlights
Prevalence of frailty in ICU patients is ~ 33% (national average in Canada is ~ 7%)
Multi-centre, prospective, provincial study, 6 hospitals (2 tertiary, 4 community), with hospital and post-hospital follow-up
Included all patients with age greater or equal to 50, and frailty score of > 4.
Excluded moribund patients, previously enrolled in the study (those with more than once admission)
Primary outcome – all cause in-hospital mortality
Secondary outcomes: death, ICU or at 6 months, Health related quality @ 6, 12 months, Intensity of ICU Tx, ICU length of stay, re-admission, and serious events within the ICU (such as self extubations, re-intubations, CRBSI)
Mean age 67.1, ~40% women, mean APACHE II 19.5 (SD 7), mean ICU stay (~5 days, median IQR 2-10) – it is important to note that the patients who were categorized as frail had similar ICU illness severity scores (e.g. APACHE II scores)
Frail patients tended to be older, female, having experienced prior hospital admissions, and have more co-morbid conditions.
Exciting results
higher incidence of in-hospital mortality (~32% frail vs 16% non-frail)
longer ICU stays 7 (4-13) vs 6 (3-10) days, and also overall hospital stay
Much fewer patients made it back home to live independently 22% frail vs. 44% non-frail patients discharged with “new” dependent status ~71% frail vs. ~52% non-frail frail patients tended to have greater hospital readmissions
Key take-away messages
Frailty is prevalent, at least ~30% or more of patients admitted to ICUs
Frailty is not just a geriatric condition Patients with similar ICU severity of illness but worse frailty status tend to do worse and we should actively consider such factors.
As the authors state eloquently “This suggests that an episode of critical illness in a frail patient may herald a momentous transition toward greater homeostatic instability, disability, and risk of death”
Future directions
Do we need a combined score or one that considers frailty? Can this be used for prognostication? Can it help us speak with families when discussing their loved one’s trajectory?
Do we need a frailty score developed within the ICU setting for the ICU setting? (most current scores were developed in the out-patient setting)