Study population and design
From September 2019 to November 2019, we prospectively enrolled the consecutive patients who underwent simultaneous bilateral THA in this randomized self-control study. The study was approved by the local ethics committee.
Inclusion criteria: 1. the bilateral THA used the same prosthesis through the posterolateral approach; 2. the written informed consent was obtained prior to participating in this study. Exclusion criteria: 1. previous open surgery or major trauma or infection in either hip; 2. eloid, psoriasis, eczema or other skin diseases; 3. allergy to the ingredients of the tissue adhesive; 4. underlying malignant tumors; 5. regular anticoagulation therapy; 6. peripheral vascular disease; 7. bilateral surgeries performed in stages.
Sample size calculation: according to the previous study and the preliminary results of our pre-experiment, we set α=0.05, β=0.10, the mean difference of dressing change was 1.0. An estimated 24 patients would be needed to provide 90% power. Finally, we decide to enroll 30 patients, which allowed for 20% loss to follow up.
Surgical procedures of wound closure
All patients underwent the right THA firstly and left secondly. The fixed surgical team performed the surgeries and two residents performed the wound closure. Tranexamic acid (TXA) was given intravenously twice before incision and before wound closure. Complete electrocoagulation hemostasis and repeated washing of TXA were did after implanting prosthesis. The joint capsule and external rotator muscles were reconstructed with 2-0 Ethibond non-absorbable suture W4843 (Ethicon, Somerville, NJ, USA). The deep fascia and superficial fascia were closed with 2-0 absorbable knotless barbed suture (Quill, Surgical Specialties Corporation, IL, USA) and 4-0 coated Vicryl Plus antibacterial suture (Ethicon, Somerville, NJ, USA). The subcuticular tissue was closed with 4-0 absorbable knotless barbed suture (Ethicon, Somerville, NJ, USA). The allocations were done after the standard wound closure. The surgeon and assistants remained blinded before the allocations. The wound closure of right hip was randomized using the computer-generated method in the opaque envelopes. 1 represented the application of tissue adhesive as the supplement of standard wound closure; 2 represented just the standard wound closure without tissue adhesive. The left hip received the opposite method of wound closure (Figure 1).
The HISTOACRYL® tissue adhesive (B.Braun, Melsungen, German) was applied evenly on both sides of the wound and waited for air-drying for 30 seconds. When the tissue adhesive got dry, a standard wound dressing was applied to the surface of the wound. During the hospital stay, patients and caregivers were told to notify the surgeon or nurse for dressing change if blood or exudate soaked the dressing. If the dressing could keep dry and clean before discharge, the patient wound received one dressing change. The dressing change caused by discharge was not recorded. All patients received antibiotics within 24 hours and aspirin in 35 days postoperatively.
Follow-up and wound evaluation
The preoperative, intraoperative and postoperative information of each patient including age, gender, body mass index (BMI), diagnosis, postoperative LOS, the times of dressing changes and the wound-related cost, were collected prospectively.
Wound-related cost was the sum of cost of the suture, tissue adhesive, wound dressing, and other additional materials during the hospital day.
At one-month follow-up after surgery, the wound-related complications and wound-related evaluation scores were recorded. The wound-related complications were defined as redness, dehiscence, subcutaneous hematoma, prolong wound drainage (>5days), surgical site infection (SSI) and re-suture caused by any reasons. Wound-related evaluation scores included patient scar assessment score (PSAS), Hollander wound evaluation score (HWES) and Vancouver scar score (VSS). PSAS was evaluated by the patients themselves. HWES and VSS were evaluated by two independent orthopedic residents who were unknown to which hip adopted the tissue adhesive. Both two residents received professional training and maintain satisfying consistency of wound evaluation in the pre-experiment. The mean scores evaluated by the two resident were used as the final scores. In addition, all patients wound be asked which side of wound closure they prefer.
PSAS [20]: the scoring system mainly refers to the patient's own feeling and the evaluation of wound. 6 represents normal skin and 60 represents worst imaginable scar in the patient’s scale (Table 1).
Table 1. Patient scar assessment score (PSAS). |
* | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
Is the scar painful? | | | | | | | | | | |
Is the scar itching? | | | | | | | | | | |
# | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
Is the color of the scar different? | | | | | | | | | | |
Is the scar more stiff? | | | | | | | | | | |
Is the thickness of the scar different? | | | | | | | | | | |
Is the scar irregular? | | | | | | | | | | |
Total score Patient Scar Score |
* 0 means “no, no complains”,10 means “yes, more imaginable” |
# 0 means “no, as normal skin”,10 means “yes, very different” |
HWES [21]: the scoring system includes 6 items, which are step-off of borders, contour irregularities, margin separation, edge inversion, excessive distortion and overall appearance. 1 point is for each item. The lower the score, the better the wound healing (Table 2).
Table 2. Hollander wound evaluation score (HWES). |
Incision attribute | Score if absent | Score if present |
Step − off borders | 0 | 1 |
Contour Irregularities | 0 | 1 |
Margin Separation | 0 | 1 |
Edge inversion | 0 | 1 |
Excessive Distortion | 0 | 1 |
Overall appearance | 0 (satisfactory) | 1 (unsatisfactory) |
Total Hollander score | 0 (best) | 6 (worse) |
VSS [22]: the scoring system includes 4 items, which include vascularity, pliability, height and pigmentation. On this scale, lower scores represent a more normal appearance (Table 3).
Table 3
Vancouver scar score (VSS).
Score | Vascularity | Pliability | Height | Pigmentation |
0 | Normal | Normal | Flat | Normal |
1 | Pink | Supple | < 2mm | Hypopigmentation |
2 | Red | Yielding | 2 − 4 mm | Mixed |
3 | Purple | Firm | > 4mm | Hyperpigmentation |
4 | − | Banding | − | − |
Statistical Analysis
All statistical analyses were performed by SPSS version 22 (Inc., Chicago, IL, USA). Data was showed as median, mode and interquartile range (IQR) (skewed distribution) or mean±standard deviation (SD) (normal distribution). Measurement data was analyzed by student’s tests or rank-sum test. Count data was analyzed by rank-sum test or Fisher exact test. A value of α=0.05 suits all tests. The intraclass correlation coefficient (ICC) was used to assess the observers’ agreement: 0.81 to 1.00, nearly perfect reliability; 0.61 to 0.80, strong reliability; 0.41 to 0.60, moderate reliability; 0.21 to 0.40, fair reliability; and 0 to 0.20, poor reliability.