Assessing sensory complaints in the outpatient clinic: predicting neurologic disease versus idiopathic sensory phenomena
Leah Zuroff1, Maya Alexis2, David Do3
1Hospital of the University of Pennsylvania, 2Perelman School of Medicine at the University of Pennsylvania, 3University of Pennsylvania
Objective:
To determine which features of sensory complaints can predict the presence of neurologic disease in the outpatient setting.
Background:
Sensory symptoms are common chief complaints in the neurology outpatient clinic. Not all cases are due to neurologic disease. Clinicians should risk stratify patients to determine whether additional testing is warranted. We hypothesized that predictors of neurologic disease (ND) rather than idiopathic sensory phenomena (ISP) would include numbness rather than paresthesias, persistent rather than intermittent symptoms, and focal rather than generalized symptoms.
Design/Methods:
Outpatient visits between 2015-2018 were queried for mentions of the keywords “tingling”, “numbness”, and “paresthesia”. Adult patients with at least one year of follow-up were included. Pre-existing neurologic conditions or prior neuroimaging/electrophysiologic studies were exclusionary. We extracted demographic data and the following symptom features: symptom quality (numbness, tingling), timing (intermittent, persistent), location (generalized, focal, bilateral limbs), and exam findings (normal, abnormal). Patients were categorized as having either ISP or ND based on ICD 10 codes. Relationships between presenting features and odds of ND were evaluated by logistic regression, controlling for demographic variables.
Results:
A total of 1819 cases were reviewed, of which 72 cases were included. Fifty-five patients had ND and seventeen had ISP. Categories of ND included: neuromuscular (53%), demyelinating (19%), headache (12%), and other (15%). The presence of focal symptoms, at least one concerning historical feature, or abnormal exam increased the odds of ND (OR 5.27, p=0.012; OR 21.0, p=0.006; OR 14.9, p=0.005; respectively). Symptom quality and timing were not independently associated with ND. When considering all symptom features and exam together, only abnormal exam was associated with ND (OR 32.13, p=0.011).
Conclusions:
This study quantifies the relative importance of historical features and exam findings that physicians typically consider when evaluating sensory complaints. Focal symptoms and abnormal exam most strongly predict presence of ND and should prompt further workup.