Simple Tool Pegs Prognosis After Ischemic Stroke

I dislike these prediction tools because nothing in them is objective, no scans of the brain to determine penumbra size vs. dead brain size. I scored 9, so according to this mine was mild. Then why was I in critical condition the first day?
http://www.medpagetoday.com/Cardiology/Strokes/35317
A prediction tool that can be administered by nonspecialist clinicians accurately identifies patients who will have a poor outcome after admission for an acute ischemic stroke, researchers found.
The score -- called PLAN -- takes into account Preadmission comorbidities, Level of consciousness, Age, and Neurologic deficit at hospital admission, according to Martin O'Donnell, MB, PhD, of the National University of Ireland in Galway, and colleagues.
In both a derivation and a validation cohort, the score accurately predicted 30-day and 1-year mortality, as well as death or severe dependence at discharge, with C statistics ranging from 0.82 to 0.89, the researchers reported online in Archives of Internal Medicine.
They said that the tool "appears to have adequate discrimination for use in clinical practice," adding that "for individual patients, its use should complement, and not replace, clinical assessment and judgment."
They called for additional studies to validate its use in other populations.
Knowing which patients are expected to fare worse after an acute stroke could help guide management decisions, although no prediction rules have gained widespread acceptance in clinical practice because of complexity, the need for a specialist's interpretation of brain scans or stroke severity, or insufficient precision.
O'Donnell and colleagues developed and tested the PLAN score using information available to nonspecialist clinicians at the time of hospital admission.
Their analysis included data on 9,847 patients -- half for the derivation cohort and half for the validation cohort -- from the Registry of the Canadian Stroke Network. All of the patients had been hospitalized for acute ischemic stroke at one of 11 regional stroke centers in Ontario.
The mean age of the patients was 73 and the median Canadian Neurological Scale score was 9, indicating mild severity.
Overall 30-day mortality was 11.5% in the derivation cohort and 13.5% in the validation cohort.
The PLAN rule -- maximum score of 25 -- was calculated using the following variables:

• Preadmission dependence (1.5 points)
• Cancer (1.5 points)
• Congestive heart failure (1 point)
• Atrial fibrillation (1 point)
• Reduced level of consciousness (5 points)
• Age (1 point per decade with a maximum of 10 points)
• Significant or total weakness of the leg (2 points)
• Weakness of the arm (2 points)
• Aphasia or neglect (1 point)

The score accurately predicted outcomes in both the derivation and validation cohorts.
In the validation cohort, for example, the PLAN score predicted 30-day mortality with a C statistic of 0.87, death or severe dependence at discharge with a C statistic of 0.88, and 1-year mortality with a C statistic of 0.84.
It also predicted favorable outcome at discharge, defined as a modified Rankin score of 0 to 2 (C statistic 0.80).
The tool was less precise among patients with lacunar stroke and those who received thrombolytic therapy and "needs to be interpreted with caution in these patient groups," according to the authors.
In an accompanying editorial, Mitchell Katz, MD, of the Los Angeles County Department of Health Services, said that the score "while not perfect, will be of use to clinicians."
"In addition to being possible to complete by nonspecialist clinicians, the mnemonic, the interval scoring, and the fact that the same model is used to predict short-term and 1-year mortality as well as severe dependence at discharge all contribute to ease of use," he wrote.
That said, it remains unknown whether using the tool will positively affect treatment decisions.
"If use of the PLAN rule leads to deeper discussions among patients, their families, and the medical profession with regard to decisions about extraordinary interventions, such as intensive care units or feeding tubes, or if the PLAN rule results in better planning for long-term care, the rule would be invaluable," Katz wrote.
O'Donnell and colleagues acknowledged some limitations of the study, including the uncertain generalizability to patients treated in other settings and the fact that the rule does not take into account the dynamic change in neurologic deficit following stroke.