Description

Graham et al developed a logistic regression model for the diagnosis of the carpal tunnel syndrome (CTS). The authors are from the Hospital for Sick Children, Toronto Rehabilitation Institute and the University of Toronto in Canada.


 

Parameters:

(1) thenar atrophy

(2) Phalen’s test (tingling or paresthesia in medial nerve distribution after maximal flexion of both wrists with the dorsum of both hands pressed together for 1 minute)

(3) loss of 2-point discrimination

(4) Tinel’s sign (tingling or paresthesia in medial nerve distribution following repetitive tapping over the carpal tunnel at the wrist)

(5) nocturnal numbness

(6) numbness in the distribution of the medial nerve

 

Parameter

Finding

Points

thenar atrophy

absent

0

 

present

1.44

Phalen’s test

absent

0

 

present

1.44

loss of 2-point discrimination

absent

0

 

present

1.30

Tinel’s sign

absent

0

 

present

1.16

nocturnal numbness

absent

0

 

present

1.16

numbness along median nerve

absent

0

 

present

1.03

 

X =

= SUM(points for all 6 parameters) – 2.14

 

probability of CTS =

= 1 / (1 + EXP((-1) * X))

 

Performance:

• Thenar atrophy (due to muscle atrophy) would seem to be a late finding.

• The agreement between experts and the model was 0.71 which indicates disagreement in almost a third of cases.

• If all of the findings are negative the model gives a 10% chance of CTS.

 

Items used clinically but not in the model:

(1) comorbid condition (pregnancy, diabetes, hypothyroidism)

(2) improvement with splinting and/or corticosteroid injection

(3) worsening with driving, strenuous hand use or other activity

 


To read more or access our algorithms and calculators, please log in or register.