This was a secondary analysis of a previously published RCT. The study sample consisted of adult patients (≥18 years) with AO type C fractures treated at Örebro University Hospital, a tertiary referral hospital, between June 2017 and July 2019. Inclusion and exclusion criteria are listed in Table 1. The study was approved by the regional ethical committee (ref: 2016 − 455), and the study was registered in the Swedish research database FoU in Sweden (registration number: 274674). All patients gave written informed consent before participation according to the Helsinki Declaration.(19)
Table 1
Inclusion and exclusion criteria.
Inclusion criteria | Exclusion criteria |
---|
Age 18–80 years | Previous fracture of the same wrist |
Operation within 12 days of injury | Bilateral fractures |
AO type C with one or more of the following: | Other concomitant fractures |
> 20° dorsal angulation of the distal radial articular surface | Open fracture |
Fracture extending to the diaphysis |
> 2 mm ulnar plus | |
| Ongoing chemotherapy or radiotherapy |
| Metabolic diseases that affect bone |
| Dementia |
| Mental illness |
| Alcohol misuse disorder or opiate addiction |
| Difficulty understanding Swedish |
| Severe neurological disease |
| Severe cardiopulmonary disease |
| Not in workforce at time of injury |
AO = Arbeitsgemeinschaft für Osteosynthesefragen. |
A total of 150 patients (n = 75 in each group) were randomized to open reduction and fixation using a volar locking plate (TriMed) or combined plating with a volar T-plate and a low-profile dorsal plate with variable angle locking screws (TriMed). All surgeries were performed by specialists in hand surgery. The patients were randomized using sealed opaque envelopes containing the study number and surgical technique, which were opened prior to surgery by the operation room nurse.
The sample size for the original study was calculated on the basis of the QuickDASH score. A sample size of 63 patients in each group available at the 1-year follow-up was considered the minimum to detect a 10-point difference in the QuickDASH score at a significance level of p = .05 and a common standard deviation of 20 points with 80% power. With an estimated dropout of approximately 20%, the sample size was set to 150 patients. A post-hoc power analysis for this secondary analysis demonstrated a 73% power for the detection of a 10% increase in risk of delayed return to work.
Surgical technique
All surgeries were performed under combined brachial plexus block and general anaesthesia. A tourniquet was used. A volar central incision was made to expose the volar ulnar portion of the distal radius, and the carpal tunnel was opened.(20) The volar portion was visualized between the finger flexors ulnarly, the median nerve, and the thumb flexor radially. The pronator quadratus was divided with a central split, and the volar cortex was reduced. The volar plate was placed, and the pronator quadratus was repaired using resorbable sutures when possible. This surgical approach was clinical routine for AO type C DRFs at the time of the study.
In the combined plating group, an additional dorsal longitudinal incision was made. This was placed centrally over Lister’s tubercle. The fracture was dorsally exposed through the fourth extensor compartment, following the division of the extensor retinaculum and subperiosteal dissection. Impacted fracture fragments were reduced and the dorsal plate was fixated starting proximately and going distally.
If an associated ulnar styloid fracture was present, the stability of the distal radioulnar joint (DRUJ) was assessed after plate fixation of the DRF. DRUJ instability was treated with reduction and fixation of the styloid with either a 3.0-mm cannulated screw (DePuy Synthes) or a 2.0-mm locking ulna hook plate (DePuy Synthes). (Fig. 1)
Antibiotic prophylaxis (intravenous cloxacillin 2 g, or in case of intolerance to β-lactam antibiotics, clindamycin 600 mg) was used if the duration of surgery exceeded 2 hours or if wounds were present in or near the surgical field.
Postoperative care
All patients were given instructions on active finger ROM exercises and oedema control by an experienced hand therapist on the first postoperative day. A cast was applied for 2 weeks. At 2 weeks postoperatively, the cast was replaced with an orthosis which was removed for active wrist and finger ROM exercises with a maximum load of 1 kg. The load was gradually increased, and the orthosis was removed 4 weeks postoperatively.
Clinical evaluation
At the 1-year follow-up, a hand therapist performed clinical measurements including ROM, hand grip strength, visual analogue scale (VAS) pain scores, and patient-reported outcome measurements. Both the Patient-Rated Wrist Evaluation (PRWE) score and the short version of the QuickDASH questionnaire were used in validated Swedish translations.(21, 22) The PRWE is a 15-item questionnaire used to calculate a score of 0–100, where 0 represents no pain or disability in activities of daily living. The QuickDASH questionnaire evaluates a patient’s upper-extremity disability during the past week. A score of 0–100 is calculated from an 11-item questionnaire, where 100 represents the most severe disability and symptoms.
Wrist ROM, including flexion, extension, radial deviation, ulnar deviation, and pronation and supination (degrees), was evaluated using a goniometer according to the Swedish National Quality Registry for Hand Surgery guidelines.(23) Hand grip strength (in kg) was measured with a Jamar Hand Dynamometer (Biometrics Ltd) in handle position 2. The mean value of 3 measurements was calculated.(23) For right-handed patients, correction of grip strength was calculated as a percentage of the uninjured side, assuming that the right upper extremity was 10% stronger. No correction was made for left-handed patients.(24) Radiographic evaluation of the results was performed on the 12-month follow-up radiographs. Batra score was calculated and graded into 4 categories of radiographic results: excellent (90–100), good (80–89), fair (70–79), or poor (< 70).(25)
Pain was evaluated both at rest and during activity using the VAS pain score (0 = no pain, 10 = worst imaginable pain).
Data on the duration of time of work and manual labour were collected from patient data records and patient interviews at the 12-month follow-up. Having spent more than 3 months off work was considered as delayed return to work. The type of work was defined as manual or non-manual, based on the Swedish Standard Classification of Occupations (SSYK 2012), which in turn is based on the International Standard Classification of Occupations (ISCO-08).(26, 27)
Statistical analysis
Statistical analyses were performed using version 29.0 of IBM SPSS Statistics for Windows. Demographic data are presented as the number of cases with median and interquartile range (IQR). The Shapiro-Wilk test was used to assess normality of distribution (data not shown). Because the data were not normally distributed and ordinal data were used, the results are presented as median and IQR. Comparisons between groups were made using the Mann-Whitney test, χ2 test, and Fisher’s exact test as appropriate.
A univariate logarithmic regression analysis was performed, and parameters with α < 0.1 were considered for inclusion in the multivariate model. Collinearity between parameters was checked before inclusion in the final analysis, using a variance inflation factor of < 5 as a cut-off. Potential parameters were also tested for interaction.
P-values < 0.05 were considered statistically significant.