Context
Diagnosis of Parkinson disease (PD) remains challenging. An accurate
diagnosis is important because effective symptomatic treatment for PD is available.
Objective
To systematically review the literature for information on the precision
and accuracy of the clinical examination for diagnosing PD.
Data Sources
MEDLINE database was searched for all English-language articles related
to the diagnosis of PD published from January 1966 through April 2001. The
reference lists of all articles retrieved were searched for additional relevant
sources.
Study Selection
Studies in which patients presented with 1 or more typical features
of PD were included if the final diagnosis was confirmed by a suitable criterion
standard and data could be extracted to determine the accuracy of 1 or more
symptoms or signs. Variability in descriptions of symptoms and signs made
it impossible to combine data across existing studies for most findings.
Data Synthesis
We identified 6 studies that met our criteria. The positive (presence)
likelihood ratios (LRs) for tremor as a symptom of PD ranged from 1.3 to 17
(range of negative [absence] LRs, 0.24 to 0.60). Tremor as a sign of PD produced
a range of positive LRs from 1.3 to 1.5 (negative LRs, 0.47 to 0.61). Clinical
features useful in the diagnosis of PD include a history of the combination
of symptoms of rigidity and bradykinesia (positive LR, 4.5; negative LR, 0.12);
a history of loss of balance (range of positive LRs, 1.6 to 6.6; range of
negative LRs, 0.29 to 0.35), symptoms of micrographia (range of positive LRs,
2.8 to 5.9; range of negative LRs, 0.30 to 0.44), and a history of shuffling
gait (range of positive LRs, 3.3 to 15; range of negative LRs, 0.32 to 0.50).
Trouble with certain tasks such as turning in bed (positive LR, 13; negative
LR, 0.56), opening jars (positive LR, 6.1; negative LR, 0.26), and rising
from a chair (range of positive LRs, 1.9 to 5.2; range of negative LRs, 0.39
to 0.58). Useful signs include the glabella tap test (positive LR, 4.5; negative
LR, 0.13), difficulty walking heel-to-toe (positive LR, 2.9; negative LR,
0.32), and rigidity (range of positive LRs, 0.53 to 2.8; range of negative
LRs, 0.38 to 1.6). Significant selection bias was detected in all studies
included for review.
Conclusions
Symptoms of tremor, rigidity, bradykinesia, micrographia, shuffling
gait, and difficulty with the tasks of turning in bed, opening jars, and rising
from a chair should be carefully reviewed in all patients with suspected PD.
The glabella tap and heel-to-toe tests also should be assessed.