Ranking of tests for pain hypersensitivity according to their discriminative ability in chronic neck pain

• Published in: Regional anesthesia and pain medicine Vol. 38; no. 4; pp. 308 - 320
• Main Authors: Neziri, Alban Y, Limacher, Andreas, Jüni, Peter, Radanov, Bogdan P, Andersen, Ole K, Arendt-Nielsen, Lars, Curatolo, Michele
• Format: Journal Article
• Language: English
• Published: England BMJ Publishing Group LTD 01.07.2013
• Subjects: Adult; Aged; Area Under Curve; Case-Control Studies; Chronic Pain - diagnosis; Chronic Pain - physiopathology; Chronic Pain - psychology; Cold Temperature; Discriminant Analysis; Electric Stimulation; Female; Hot Temperature; Humans; Hyperalgesia - diagnosis; Hyperalgesia - physiopathology; Hyperalgesia - psychology; Likelihood Functions; Logistic Models; Male; Middle Aged; Neck pain; Neck Pain - diagnosis; Neck Pain - physiopathology; Neck Pain - psychology; Odds Ratio; Pain Measurement - methods; Pain Perception; Pain Threshold; Predictive Value of Tests; Pressure; Prospective Studies; Regional anesthesia; ROC Curve
• ISSN: 1098-7339; 1532-8651
• DOI: 10.1097/AAP.0b013e318295a3ea

Quantitative sensory testing (QST) is widely used to investigate peripheral and central sensitization. However, the comparative performance of different QST for diagnostic or prognostic purposes is unclear. We explored the discriminative ability of different quantitative sensory tests in distinguishing between patients with chronic neck pain and pain-free control subjects and ranked these tests according to the extent of their association with pain hypersensitivity. We performed a case-control study in 40 patients and 300 control subjects. Twenty-six tests, including different modalities of pressure, heat, cold, and electrical stimulation, were used. As measures of discrimination, we estimated receiver operating characteristic curves and likelihood ratios. The following quantitative sensory tests displayed the best discriminative value: (1) pressure pain threshold at the site of the most severe neck pain (fitted area under the receiver operating characteristic curve, 0.92), (2) reflex threshold to single electrical stimulation (0.90), (3) pain threshold to single electrical stimulation (0.89), (4) pain threshold to repeated electrical stimulation (0.87), and (5) pressure pain tolerance threshold at the site of the most severe neck pain (0.86). Only the first 3 could be used for both ruling in and out pain hypersensitivity. Pressure stimulation at the site of the most severe pain and parameters of electrical stimulation were the most appropriate QST to distinguish between patients with chronic neck pain and asymptomatic control subjects. These findings may be used to select the tests in future diagnostic and longitudinal prognostic studies on patients with neck pain and to optimize the assessment of localized and spreading sensitization in chronic pain patients.