This randomised clinical trial was conducted in two tertiary hospitals. A random sample of patients were recruited between 2020 and 2021. They were provided with and signed written informed consent forms before enrolment. Ethical approval for this study was obtained from the related Ethics Committee (approval number: IR.MUMS.REC.1397.697). This study is registered at Iranian Registry of Clinical Trials (IRCT) with approval code of IRCT20200313046759N1 at 25/10/2021. The conduction of this research accords with the Declaration of Helsinki and adheres to the CONSORT guidelines. Our unit’s most common standard care for unstable distal radius fractures is percutaneous pinning (PCP) and external fixation. The other option is internal fixation using plates. In our consent form we described both methods and their advantages and disadvantages, so patients could choose between them. Patients who agreed with PCP and external fixation then enrolled in present study.
In this parallel-designed randomised clinical trial, patients who were attending the hospital emergency department with acute unilateral distal radius fractures were randomly recruited by our hand surgeon (A. M). A priori sample size compute was conducted using G*power 220.127.116.11, with the effect size of 0.84 based on similarly designed study’s  results, alpha error of 0.05, power of 95%, and 1:1 ratio allocation; which resulted in 38 patients in each group. Thus, 77 patients with unstable fracture were considered based on having one or more of the following criteria: intra-articular radiocarpal fracture, over 20° of dorsal angulation, dorsal comminution, and more than 5 mm shortening. In case they had the indication of treatment with external fixator, they were contacted and informed about the study. Of those, 68 patients agreed to participate, and were provided with written informed consent forms. The exclusion criteria were patients with prior history of wrist fracture, inflammatory diseases in affected wrist, open fracture, and concomitant carpal bones fracture. Accordingly, 8 patients were further excluded, leaving 60 patients as study sample for final evaluation.
Before operation, imaging study was performed for measuring the radiological parameters of joint displacement (palmar tilt, radial high, radial inclination, articular step and wrist alignment). These measurements were conducted before allocation of patients to either group in order to ensure blindness of data. The letters A and B were used for conventional treatment; the letters C and D were used considered for our proposed method of treatment, before randomization to conceal the group assignment Patients were then randomly allocated (1:1 ratio) to one of the two groups using block randomization method (block size: 4, block number: 15, permutation number: 24) by a biostatistician, and one group was set to undergo external fixation operation applying fixator parallel to radius shaft (group A = 30) and the other group with external fixator perpendicular to the distal radius articular surface (group B = 30).
All patients were hospitalized and were initially managed by long forearm splint. The operation procedure was performed with the patient in the supine position following induction of general anaesthesia. Then, the arm was prepped and draped. Using the appropriate manoeuvre of traction, flexion, and ulnar deviation, closed reduction was performed. Before fixation, re-imaging was done to ensure appropriate reduction by fluoroscopy. Then, with one radial side and one ulnar side pins, closed fixation was performed. Closed reduction and correct fixation (placement of the pins) were again verified with fluoroscopy and proximal schanz pins were inserted in radius shaft. First, we determined the entrance points for pins in lateral side, and then through a 5-mm skin incision and soft tissue dissection to the bone (using 11 bistoury). Two separated 2.5 mm schanz pins were placed perpendicular to the radius shaft and proximal to the fracture line. Via a limited incision on second metacarpal bone base, a 2.5 mm schanz pin was placed in metaphysis-diaphysis junction and extended through third metacarpal lateral cortex. Another 2.5 mm schanz pin was placed in second metacarpal shaft. For fixation and distraction, we used a pre-fabricated external fixator with two adjustable joints for palmar tilt and radial deviation respectively. External fixator was mounted to the schanz pin using the two techniques below:
- All external fixator indices were set on zero and, longitudinal distraction force was applied parallel to distal radius axis (group A) [Figure 2-A].
- The external fixator indices were set in accordance with mean population wrist radial inclination (24 degrees) and palmar tilt (10 degrees). Then, distraction force was applied perpendicular to the wrist joint (group B) [Figure 2-B]. In order to make sure the distraction force is exerted perpendicular to the articular surface, we estimated the direction of force based on Mashhad population normal distal radius indices previously determined in Vaezi et al. Accordingly, normal radial inclination and palmar tilt were considered 24 and 10 degrees, respectively.
After external fixator insertion, over distraction under guide of fluoroscopy was performed in a controlled way until 2 mm distraction occurred in radio-lunate joint. We then gradually decreased the distraction to 1 mm. The day after operation and after performing true PA and lateral distal radius radiographs, we recorded the radial inclination and palmar tilt as well as joint displacement (radial height, radial inclination, articular step, and wrist alignment) using IC Measure software (version 18.104.22.1686, the imaging source, Bremen, Germany) to compare the radiological outcome between two groups.
To gather data, we further recorded radiological data parameters at 6 weeks follow up. The fixator was also removed at 6 weeks, after we clinically ensured that the union achieved. Patients’ clinical data using grip strength dynamometry and wrist range of motion as well as VAS, Quick DASH, Mayo wrist score and PRWE questionnaires were gathered at 6 and 12 weeks post-operatively. Follow-up data was gathered by a medical intern who is trained in orthopaedic research fields (KE). To ensure blindness, we asked her to fill measurement tools after extraction of external fixators.
Tools (Data Sources)
We used the biplanar adjustable joint bridging external fixator (Avisa Co., Mashhad, Iran). It includes a radius fixing plate with two clamps, a metacarpus fixing plate with two clamps, and a coupling treaded bar. The radius plate includes two joints. The proximal joint has one degree of freedom along coronal plan (for radial deviation adjustment) and the distal joint on coronal plan (for palmar tilt adjustment). A single nut between two plates distracts the clamps [Figure 3].
Below are tools for measuring clinical variables. These data were recorded at 6 and then 12 weeks post-operatively:
- Goniometry: Wrist range of motion in six positions of flexion, extension, pronation, supination, radial deviation and ulnar deviation were measured three times, the mean of which was recorded.
- Grip strength: With the patients in sitting position, elbow in 90 degrees flexion and neutral forearm and wrist position, grip strength was measured three times with Jamar grip dynamometer (J. A. Preston Corporation, Clifton, NJ), the mean of which was recorded.
- Patient-Rated Wrist Evaluation (PRWE) questionnaire: This questionnaire evaluates three factors of wrist pain, disability in activities of daily living and disability with doing specific activities. It consists of 15 items and each item has ten scores; based on item scores, scale scores are calculated ranging from 0 (no pain or disability) to 100 (most severe pain and disability). We used the translated and validated version of the questionnaire in Persian.
- Quick Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire: The Quick DASH questionnaire includes 11 items from the original 30-item DASH evaluating upper limb symptoms and disabilities. Questions are about the patient’s ability to perform different activities, sleep quality, social and regular daily activities, pain severity, and tingling. Each item has five response options; based on item scores, scale scores are calculated ranging from 0 (no disability) to 100 (most severe disability). We used the translated and validated version of the questionnaire for Persian speakers.
- Mayo wrist score questionnaire: It assesses four domains of pain, satisfaction, wrist range of motion and grip strength. Each domain is scored from 0 to 25 points to produce a total score out of 100 points. Higher scores mean better function: Scores of 90-100 are interpreted as “excellent” function, 80-89 as “good”, 65-79 as “intermediate” and a score of less than 65 is considered “poor”. The physician completes this questionnaire for patient, and thus the translated version is not required.
- Pain Visual Analogue Score (VAS): In order to quantify the severity of pain, we used the VAS scale, scaled continuously from 0 (no pain) to 10 (worst pain) on a 10 centimetre scale.  We then measured the distance from 0 to the point where patients marked their pain level in cm.
- Radiography: Radiological parameters were measured three times by taking PA and lateral distal radius radiographs once before surgery and then immediately and 6 weeks post-operatively. Radiographs were performed under supervision of one of our researchers (KM) to reassure the radiology beam is perpendicular to sagittal plane of radius shaft. Radiological parameters were measured by a radiologist’s technician, who was blind to group allocation of patients. The measured parameters are:
- Radius palmar tilt: Taking lateral view, this parameter is the angle made by the line vertical to radius shaft and the line tangent to the volar to dorsal aspect of the distal radius [Figure 4-A].
- Wrist malalignment: The angle between the lines drawn along the long axes of Capitate and radius from a lateral view [Figure 4-B].
- Radial inclination: The angle between the line vertical to the radius shaft axis and the line that connects the distal radio-ulnar joint (the midpoint of volar and dorsal lips) with styloid process in PA view [Figure 4-C].
- Radius height: Taking PA view, this parameter is the distance in millimeter between two parallel lines which are vertical to radius shaft. One line is drawn from level of the ulnar aspect of the articular surface (the midpoint of volar and dorsal lips) and the other from apex of radius styloid [Figure 4-D].
- Articular step: Measurement of depression or protuberance in joint surface using AP view. In fractures with multiple articular steps, we calculated this value by considering the most depressed and bulged steps among them.
Data were analysed using SPSS (version 22). Quantitative data are reported as the mean ± SD, and qualitative data as number in percentage. After determination of variable data distribution with Kolmogrov-Smirnov test, comparison between two groups were performed using independent T-test when the data were normally distributed; otherwise, we used Mann Whitney test. In order to compare the data before and after surgery within each group, paired T-test was performed for variables which had normal distribution and Wilcoxon test for variables with non-normal distribution. P values less than 0.05 were considered statistically significant. Finally, power analysis was conducted using G*power (version 22.214.171.124) for each comparison between two groups. For normally distributed variables initially analyzed by independent T-test, we used difference between two independent means (t-test) Post hoc, considering alpha error = 0.05. For variables with non-normal distribution initially analyzed by Mann Whitney test, we used two independent groups’ (non-parametric) test Post hoc, considering parent distribution as Laplace and alpha error=0.05.