Diagnostic Testing for Patients With Peripheral Neuropathy

Abstract

This cross-sectional study compares the diagnostic testing for patients with peripheral neuropathy of patients from 1998 to 2007 with patients in 2019.

American Academy of Neurology (AAN) guidelines recommend diagnostic testing for fasting glucose, vitamin B₁₂, and serum protein electrophoresis (SPEP) in patients with peripheral neuropathy (PN) and a glucose tolerance test (GTT) in those without diabetes on routine testing.¹ Magnetic resonance imaging (MRIs) of the brain and spinal cord and electrodiagnostic testing (EMG) are frequently ordered and drive most of the diagnostic costs but rarely change diagnosis or management.² Current guidelines^3,4 recommend against using MRIs for evaluation of PN and performing routine EMG for diabetic neuropathy. Using Medicare data from 1998 to 2007, we found that AAN guideline–recommended tests, particularly SPEP and GTT, were likely underused and MRIs and EMGs were likely overused.⁵ However, whether use has changed after publication of current recommendations^3,4 remains unknown.

Methods

This cross-sectional study used the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline and was approved by the institutional review board at the University of Michigan. Informed consent was waived because patient data were deidentified.

We used a 20% randomly selected national sample of 2019 Medicare fee-for-service claims and determined use of AAN-recommended tests, MRIs of the neuroaxis, and EMG in patients with PN. We used the same methods from our previous study⁵ to determine incident PN and diagnostic testing. Included patients were at least aged 65 years, had no PN diagnosis in the preceding 30 months, and had continuous Medicare enrollment for 30 months prior to and 6 months after incident PN diagnosis. Analyses were descriptive and performed using SAS version 9.4 (SAS Institute) from March to April 2024.

Results

We identified 83 693 patients with incident PN in 2019 (mean [SD] age, 77 [7.1] years, 44 880 female [53%], 7162 Black patients [9%], 4655 Hispanic patients [6%], 68 982 White patients [82.4%], and 46 476 had diabetes [59%]). Among AAN-recommended tests, B₁₂ was completed most frequently (23 685 [28.3%]) followed by SPEP (11 215 [13.4%]) and fasting glucose (8871 [10.6%]) (Figure 1). Although GTT was completed in 167 patients (1.0%), hemoglobin A_1c was completed in 42 098 patients (50.3%). EMGs were completed in 18 915 patients (22.6%) and MRIs of at least 1 neuroaxis segment in 21 593 patients (25.7%) (Figure 2). In those without diabetes, B₁₂ (11 746 [33.2%]), SPEP (6793 [19.2%]), EMGs (11 711 [33.1%]), and MRIs (11 357 [32.1%]) were more frequently completed. Use of fasting glucose decreased from 1998 to 2007 (−12.8%; 95% CI, −15.4% to −10.2%) whereas hemoglobin A_1c increased over time (7.1%; 95% CI, 4.1% to 10.2%). Most testing was stable over time, but small increases in EMG (2.8%; 95% CI, 0.4% to 5.3%) and MRIs (2.6%; 95% CI, 0% to 5.2%) were observed.

Figure 1. Use of American Academy of Neurology (AAN)-Recommended Tests in Patients With Peripheral Neuropathy.

B₁₂ indicates vitamin B₁₂; GTT, glucose tolerance test; HbA_1c, hemoglobin A_1c; SPEP, serum protein electrophoresis.

Figure 2. Use of Electrodiagnostic Studies (EMG) and Magnetic Resonance Imaging (MRI) of the Neuroaxis in Patients With Peripheral Neuropathy.

EMG and MRI of the neuroaxis, including brain and spinal cord.

Discussion

Despite recommendations,^3,4 diagnostic test use for PN remains largely unchanged from 2007 to 2019. EMGs and MRIs were still likely overused. Small increases in EMG and MRI over time suggest slightly worsening overuse. Furthermore, SPEP and GTT were still likely underused despite AAN guidelines recommending their use. Current recommendations and guidelines were based on robust data that simple blood tests were the most cost-effective way to identify intervenable causes of PN whereas EMGs and MRIs were expensive and rarely changed management.

Interventions designed to optimize PN testing might improve care, reduce inappropriate diagnostic cascades, and lower costs. These interventions are likely needed at the clinician and/or system level because patients are unlikely to refuse laboratory testing and unlikely to request EMG, which can be uncomfortable. Potential clinician-level interventions include accountable justification (free-text in electronic health record [EHR]) and peer comparison when ordering EMG and MRIs given past success in antibiotic ordering.⁶ Potential system-level interventions include EHR peripheral neuropathy order sets that contain SPEP and GTT. Guidelines and the passage of time will not suffice. Limitations include lack of generalizability to non-Medicare populations, lack of information on appropriateness, and potential misclassification bias.

Supplement.

Data Sharing Statement