Study design:
This prospective, randomized, controlled and blinded monocenter study based on a non-inferiority study design is enrolled in a Level 1 Trauma Centre in Central Switzerland. A total of 234 pin sites (+/- 70 patients) will be included. Ethical approvement of this study was obtained from the Swiss Ethics board with the project-ID 2018–01316.
All patients treated with a temporary external fixator are screened for eligibility. The in- and exclusion criteria are shown in figure 1. All pin sites except the pin sites at the calcaneus, due to the low mobilizability of the skin and thus a lack of tension-free wound closure, will be included. After both, written and oral informed consent, patients are included.
Inclusion and Exclusion criteria:
The population includes all patients who completed the 18th year of life and were treated with an external fixator in our hospital. Patients with immunodeficiency or patients who cannot follow up on structural reasons are excluded. Patients with a lack of knowledge of German or a lack of consent to study participation are also excluded. The in- and exclusion criteria are listed in figure 1.
Randomization process:
To obtain an equal patient population or groups regarding to preexisting conditions, health status and trauma condition, all pin sites of the patients are treated at the time of removal of the external fixator alternately with a primary wound closure and by secondary wound healing. If a patient has consented to the study, the patient will be allocated to group A or group B by using a computerized randomization. Patients assigned in group A, the proximal pin site will be closed by a single-button suturing according to the Allgoewer-Technique. In patients of group B, the proximal pin site is treated by secondary open wound healing.
The pre- and postoperative protocol is standardized for all pin sites, including preoperative antibiotic prophylaxis with a single preoperative dose of cefazolin 2g intravenously 30–60 minutes prior to surgery. The pin sites assigned to the intervention group will be treated according to our current standard protocol, which implies that they will be closed by single button technique. The control group is simultaneously subjected to secondary wound healing without wound closure.
No routine postoperative antibiotics are given. Patients with an open fracture will be treated according to the local protocol. Patients with a Gustillo grade 1 or 2 open fracture are treated with Cefazoline 2g i.v. TID for 24 hours. Patients with open fractures classified Gustillo 3 receive Amoxicillin/Clavulanic-Acid 2.2g i.v. TID for 72 hours [17].
The postoperative pin site care includes the daily inspection of pin entry sites, disinfection with Betadine ©, followed by a dry gauze dressing by the nursing staff during inpatient stay. In the further course this is done either by an outpatient wound care, the family doctor, or in case of good compliance the patient.
A photo documentation of the pin sites is made 2 and 52 weeks postoperatively. All patients are clinically controlled at 2, 6, 12, 26, and 52 weeks during regular follow-up visits. The chosen therapy for the examining doctor is blinded at the 6 weeks and 52 weeks controls. A flowchart is shown in figure 2 and a study schedule in figure 3
Statistical Methods:
For the primary study objective, it should be shown in a non-inferiority approach that the rate of postoperative wound infection (within 12 weeks of removal of the external fixator) is not significantly greater following simple wound closure of the pin entry sites than in open secondary wound healing. The non-inferiority limit for this proof is 10%. The primary study objective to be confirmed is achieved when the upper limit of the 95% confidence interval (2-sided) for the difference in infection rates (simple wound closure - open wound healing) does not exceed the non-inferiority limit of 10%. Similarly, wound infection rates will also be evaluated at the other assessment time points. Furthermore, also for the secondary parameters, the rates of wound healing and the rates of revision surgeries and antibiotic therapies, the proportions per treatment group and time point will be calculated in the same manner as for the primary parameter, and 95% confidence intervals will be presented for the difference in proportions between treatment groups. For the secondary parameters, a comparison to a non-inferiority limit is no longer in the main focus. The wound healing rate, for example, is more about being able to possibly deduce from the pattern of proportions an earlier onset of the healing process after simple wound closure. All other parameters will be evaluated purely descriptively.
Sample size and Determination:
70 patients, resulting in up to 234 pin entry points, are included in the study. For each patient, up to four individual wounds are treated alternately with simple wound closure or by open wound healing; the treatment of the proximal pin entry site per patient is determined by a randomization scheme (1: 1). For secondary wound healing, an infection rate of 5% is assumed (within 12 weeks postoperatively), as well as an infection rate of 5% for primary wound healing. Assuming that the infection probabilities of the individual wounds (even within the same patient) are independent, at least 156 evaluable wounds are needed to maintain a non-inferiority limit of 10% for the difference in infection rates with a power of 80%. Inclusion of 70 patients is expected to achieve the required number of wounds (even if individual patients contribute less than four evaluable wounds). With a possible dropout rate of approximately 25% we will generously include 234 pin sites, which equates to a number of around 70 patients.
Based on retrospective analyses it is expected that about 50 patients are being treated with an external fixator annually at our hospital. Therefore, the inclusion period will be from Januar 2019 to the middle of 2020, with an estimated end of this trial one year later after the follow up will be completed.
The sponsor, the investigator and swiss ethics have approved the trial’s protocol version 2, dated 30.10.2018. This trial will be conducted according to the ethic - protocol and the current version of the World Medical Association declaration of Helsinki, the ICH GCP guidelines and the ISO 14155, performing the standard, applied to the local legal requirements.
Methods of minimizing bias:
To avoid an initial bias, all patients are treated identical. Therefore, and to obtain an equal patient population or groups regarding preexisting conditions, health status and trauma condition, all pin sites of the patients are treated at the time of removal of the external fixator alternately by means of primary wound closure and secondary open wound healing. Randomization only applies to the treatment of the proximal pin site: closed by Allgoewer single button technique or left open for a secondary open wound healing, while the rest of the pin sites are treated alternately. This minimizes bias. Included are all pin sites, except those that affect on the calcaneus, due to the low mobilizability of the skin and thus a lack of tension-free wound closure. Regarding the back of the hand and foot, a medially located pin site is considered to be the proximal pin site.