Split-hand index for amyotrophic lateral sclerosis diagnosis: A frequentist and Bayesian meta-analysis

Wei Zhen Lu, Hui An Lin, Sen Kuang Hou, Cheng Fan Lee, Chyi Huey Bai, Sheng Feng Lin

Research output: Contribution to journalArticlepeer-review

3 Citations (Scopus)


Objective: Preferential wasting of the thenar muscles, the split-hand sign, may be used for early diagnosis of amyotrophic lateral sclerosis (ALS). Methods: Electronic databases were searched for studies assessing the split-hand index (SHI) and the compound muscle action potential (CMAP) amplitudes of abductor pollicis brevis (APB), first dorsal interosseous (FDI), and abductor digiti minimi (ADM). The SHI was obtained by multiplying CMAP amplitudes of APB and FDI and dividing the product by the CMAP amplitude of ADM. The Bayesian analysis was used for validation. Results: In total, 17 studies and 1635 patients were included. Our meta-analysis revealed that ALS patients had significantly decreased SHI (standardized mean difference [SMD], −1.60, P < 0.001), CMAP of the APB (SMD, −1.67, P < 0.001), FDI (SMD, −1.12, P < 0.001), and ADM (SMD, −1.09, P < 0.001). The binormal receiver operating characteristic curve analysis showed a threshold of < 7.4 for SHI, and cutoff values of < 6.4 mV for APB and < 8.4 mV for FDI, respectively. The Bayesian analysis validated decreased SHI in ALS patients (posterior mean difference of − 5.91). Conclusions: An SHI of < 7.4 can be used facilitating earlier diagnosis of ALS. Significance: SHI can be used as a standard neurophysiological biomarker for early diagnosis.

Original languageEnglish
Pages (from-to)56-66
Number of pages11
JournalClinical Neurophysiology
Publication statusPublished - Nov 2022


  • Amyotrophic lateral sclerosis (ALS)
  • Split-hand index (SHI)
  • Split-hand sign

ASJC Scopus subject areas

  • Sensory Systems
  • Neurology
  • Clinical Neurology
  • Physiology (medical)


Dive into the research topics of 'Split-hand index for amyotrophic lateral sclerosis diagnosis: A frequentist and Bayesian meta-analysis'. Together they form a unique fingerprint.

Cite this