Background: Patients with cervical/upper thoracic compressive myelopathy may have autonomic dysfunction. The composite autonomic severity score (CASS) is the gold standard test to detect autonomic dysfunction, and the self-rated composite autonomic system scale (COMPASS-31) questionnaire is a screening tool to diagnose autonomic dysfunction. This study compared the COMPASS-31 and modified CASS scores for the detection of autonomic dysfunction in patients with compressive myelopathy. Methods: Patients with cervical/upper thoracic compressive myelopathy scheduled for decompressive surgery completed a COMPASS-31 questionnaire and underwent autonomic function tests to calculate the modified CASS score before surgery. Results: Forty-two patients were included in the study; 19 (45.2%) had mild autonomic dysfunction, 5 (11.9%) had moderate autonomic dysfunction, and 18 (42.9%) had severe autonomic dysfunction. Median (interquartile range) of modified CASS and COMPASS-31 scores were 19 (6.33) and 3 (2.5), respectively. There was a positive correlation between modified CASS and COMPASS-31 scores (r=0.43; P=0.004). Receiver operating characteristic curve analysis confirmed that COMPASS-31 had fair accuracy for prediction of moderate to severe autonomic dysfunction (area under the curve, 0.74; 95% confidence interval, 0.64-0.82; P=0.009). A cut-off of 30 for total COMPASS-31 score had a sensitivity of 52.2% and specificity of 89.5% to detect moderate to severe autonomic dysfunction, with positive and negative predictive values of 85.7% and 60.7%, respectively. Conclusion: Patients with cervical/upper thoracic compressive myelopathy had varying degrees of autonomic dysfunction based on the modified CASS. There was a positive correlation between the modified CASS and COMPASS-31 questionnaire. A COMPASS-31 score of >30 30 could be utilized to predict moderate to severe autonomic dysfunction in patients with compressive myelopathy.