|Abstract / Summary|
Trapeziometacarpal osteoarthritis, carpal tunnel syndrome, Dupuytren’s disease,
trigger finger, and wrist ganglia are all common reason for patients to be referred to a
hand surgeon. The majority of the surgical treatments of these diseases lead to a
positive outcome, but there are still some patients who are without improvement,
unsatisfied, or even end up with an outcome that is worse than before the surgery. The
Boston Carpal Tunnel Questionnaire (BCTQ) was designed to measure function and
symptoms in patients with carpal tunnel syndrome. The measurement properties of
this questionnaire have been examined in several countries, but not in Denmark.
There has been an increasing attention on psychological factors as predictors of
surgical outcome. Catastrophic thinking about pain, characterized by an exaggerated
negative response in relation to anticipated or actual pain experiences, can be
measured using the Pain Catastrophizing Scale (PCS).
The overall aim of this thesis was to identify preoperative risk factors for
unsatisfactory outcome in patients treated for trapeziometacarpal osteoarthritis with
total joint arthroplasty, and examine the predictive effect of PCS score in patients
treated surgically for carpal tunnel syndrome, Dupuytren’s disease, trigger finger, or
wrist ganglia. Further, the aim was to evaluate the measurement properties of the
Study I is a prospective cohort study in 287 patients undergoing total
trapeziometacarpal joint arthroplasty. Age; gender; the Disabilities of the Arm,
Shoulder, and Hand (DASH) score; pain; and grip strength were used as predictors
and outcomes. Lower preoperative DASH score and lower preoperative grip strength
increased the risk of a low improvement in pain at activity (VAS<3). Women were at
increased risk of low improvement in pain at rest (VAS<3) compared to men.
Study II is a prospective cohort study in 714 patients treated with decompression surgery for
carpal tunnel syndrome. Preoperative age, gender, DASH score, EQ-5D (EuroQol-5d)
score, distal motor latency, operation technique, and the Pain Catastrophizing Scale
were used as predictors of postoperative satisfaction. Patients improved in both DASH
score (mean=12.29) and EQ-5D (mean=0.14) after median nerve decompression
surgery. Preoperative PCS score was the only predictor of 12-month postoperative
patient satisfaction, where a higher PCS score increased the risk of low postoperative
patient reported satisfaction.
Study III examined the measurement properties of the
Danish BCTQ in 188 patients treated for carpal tunnel syndrome with median nerve
decompression surgery. The Danish BCTQ showed high responsiveness, internal
consistency, and reliability. Further, the Danish BCTQ was moderate to strongly
correlated to the Danish QuickDASH.
Study IV used postoperative patient satisfaction
as outcome, and used preoperative age, gender, DASH score, EQ-5D, dominant hand,
civil status, and PCS score as preoperative predictors in 645 patients with Dupuytren’s
disease, trigger finger, or wrist ganglia. Patients improved in both DASH score
(median=10.9) and EQ-5D (median=0.18). The most important preoperative cut-points
on the PCS for postoperative patient satisfaction were 27.5 and 2.9. Only 2.9 remained
statistically significant after adjustment for demographics and preoperative disability.
Conclusion: It was not possible to identify one preoperative risk factor for all
outcomes after total trapeziometacarpal joint arthroplasty. Preoperative score on the
PCS was a contributing risk factor for low postoperative patient satisfaction after
surgical treatment of carpal tunnel syndrome and Dupuytren’s disease, trigger finger,
and wrist ganglia. We found satisfactory measurement properties of the Danish BCTQ
with special regard to reliability and responsiveness.
Perspectives: In future studies, it would be interesting to examine the predictive value
of preoperative PCS score in patients surgically treated for trapeziometacarpal
osteoarthritis. To establish a useful risk estimation tool for clinical use, studies should
try to build a prediction model with variables known to have predictive abilities,
including pain catastrophizing measured with the PCS. Preferably, the guidelines for
building prediction models like the “Transparent Reporting of a multivariable
prediction model for Individual Prognosis or Diagnosis” should be followed.