American Association for Hand Surgery

Back to 2020 Abstracts

Evaluation of the Seesaw Test as New Diagnostic Test for Thumb CMC Osteoarthritis
Denise MJ Arnold, MD, Rachel Gottlieb, BSc, Suzanne C Wilkens, MD, Rohit Garg, MD, Adam Nazzal, MD, Neal C Chen, MD and Sang-Gil Lee, MD, Massachusetts General Hospital/Harvard Medical School, Boston, MA

Introduction: Radiographic presence of thumb carpometacarpal (CMC) osteoarthritis (OA) doesn't always correlate with symptoms. Clinical testing specific to thumb CMC OA can be helpful in elucidating symptomatic joints and differentiating from other common sources of pain around the thumb. Sensitivity values of the most commonly used grind test are low and vary greatly. We aim to evaluate a new diagnostic test named the seesaw test and compare this test to the grind test.

Methods: We prospectively enrolled 80 subjects from March 2017 to March 2018 at a single institution, excluding patients who had previous thumb arthroscopy, surgery or pathology. Each subject underwent both the seesaw and grind test by the hand surgeon and hand fellow. Both tests were considered positive if eliciting pain. The seesaw test consists of hyperextending the thumb metacarpal from a neutral resting posture, followed by adduction with continued pressure of the metacarpal base. We included 24 patients with a mean age of 73 years (σ 8.9) in the CMC OA group (Eaton stages 2 to 4) and 44 patients with a mean age of 66 years (σ 10.8) in the control group (Eaton stage 0 or 1). We calculated sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV), and we used the multirater kappa measure as interpreted by Landis and Koch to estimate agreement among the care providers.

Results: The seesaw test had a substantially higher sensitivity (71% and 42%, hand surgeon and fellow respectively) than the grind test (13% and 17%). However, the specificity of the seesaw test was slightly lower (82% and 86%) when compared to the grind test (91% and 98%). The PPV is more consistent between examiners for the seesaw test (68% and 63%) than for the grind test (42% and 80%). The NPV is higher for the seesaw test (84% and 73%) than for the grind test (66% and 68%). There is a slight agreement between the surgeon and fellow when performing the grind test (Kappa = 0.08) and a moderate, borderline substantial agreement when performing the seesaw test (Kappa = 0.59).

Conclusions: The seesaw test is superior to the grind test, with a much higher sensitivity, slightly lower specificity, more consistent PPV, slightly higher NPV, and more consistent results among healthcare providers. Therefore, the seesaw test could be a valuable addition or even replacement of the grind test, and we recommend considering it in daily practice.

Back to 2020 Abstracts