Effect of a wound protector on soft-tissue injury and early recovery after endoscopic direct anterior approach total hip arthroplasty: a randomized controlled trial

Abstract

Background

The endoscopic direct anterior approach (Endo-DAA) for total hip arthroplasty (THA) aims to minimize soft-tissue trauma and accelerate recovery. Whether adding a disposable wound protector can further reduce early muscle injury and enhance recovery remains unclear. This randomized controlled trial evaluated the impact of wound protector use on biological, clinical, and functional outcomes after Endo-DAA THA.

Methods

Seventy-six patients undergoing primary Endo-DAA THA were prospectively randomized (1:1) to either a wound protector or control group. The only intraoperative difference between groups was the use of a wound protector. The primary outcome was serum creatine kinase (CK) at 24 h postoperatively. Secondary outcomes included serum myoglobin at 6 h, C-reactive protein at 24 h, pain visual analog scale (VAS) during activity, independent ambulation within 12 h, lateral femoral cutaneous nerve (LFCN) symptoms, wound complications and satisfaction, and functional scores using the Harris Hip Score (HHS) and Forgotten Joint Score-12 (FJS-12). Statistical significance was defined as P < 0.05.

Results

Baseline characteristics were comparable between groups. The wound protector group had significantly lower postoperative CK (422.5 vs. 665.5 U/L; median difference: -228.0; 95% CI: -354.0 to -109.0; P < 0.001), myoglobin (299.9 vs. 481.2 ng/mL; P < 0.001), and C-reactive protein levels (30.3 vs. 45.1 mg/L; P = 0.024). Pain scores were lower at 12 and 24 h, and more patients achieved independent ambulation within 12 h (86.8% vs. 55.3%; risk ratio [RR]: 1.57; 95% CI: 1.14 to 2.16; P = 0.002). Fewer wound complications (7.8% vs. 34.2%; RR: 0.23; 95% CI: 0.07 to 0.75; P = 0.005) and LFCN symptoms (36.8% vs. 65.7%; P = 0.012) were observed. Early functional scores were higher in the protector group, while midterm outcomes were similar.

Conclusions

In Endo-DAA THA, the use of a disposable wound protector significantly reduced early muscle injury and systemic inflammation, alleviated immediate postoperative pain, facilitated earlier mobilization, and enhanced early wound healing and functional recovery without increasing operative time or hospital stay. These benefits were primarily observed during the early postoperative phase, which aligns with the principles of enhanced recovery after surgery.

Trial registration

Chinese Clinical Trial Registry, ChiCTR2300076225. Registered on September 27, 2023.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13018-026-06710-w.

Introduction

Total hip arthroplasty (THA) reliably alleviates pain and restores function in patients with end-stage hip disease [1, 2]. The direct anterior approach (DAA), which utilizes an intermuscular and internervous plane, has gained popularity because it is associated with reduced early pain, faster mobilization, and shorter hospital stay compared with other approaches [3, 4]. More recently, various technological advancements have been introduced to improve THA outcomes, ranging from robotic-assisted systems that enhance component placement accuracy [5, 6] to endoscopic techniques that improve visualization. The endoscopic direct anterior approach (Endo-DAA) has been developed to further enhance intraoperative visualization through a limited incision while maintaining the same soft-tissue–sparing philosophy [7–9]. Although early clinical reports are encouraging, most are observational and combine several technical adjuncts—such as mini-incision exposure, endoscopic assistance, and the use of disposable wound protectors—making it difficult to isolate the contribution of any single factor. Nevertheless, concerns remain regarding retractor-related soft-tissue trauma, wound complications, and very early postoperative pain, which are particularly relevant within enhanced recovery pathways.

Sustained pressure and friction from conventional metal retractors within a small incision can damage the skin edges, subcutaneous tissue, and underlying muscles, and may also irritate the lateral femoral cutaneous nerve (LFCN), leading to numbness or paresthesia over the anterolateral thigh. Excessive retraction within a confined surgical corridor can compress and shear muscle fibers, resulting in myotrauma that contributes to very early postoperative pain and delays mobilization [10, 11]. LFCN symptoms after DAA are relatively common, although most cases resolve spontaneously over time [12–14]. Such soft-tissue and neural sequelae may blunt the expected early advantages of tissue-sparing approaches by impeding very early mobilization. Therefore, pragmatic strategies that minimize mechanical trauma at the wound edge and within the retraction corridor are of clinical relevance.

Disposable wound protectors (WPs) provide circumferential retraction and continuous protection of the cutaneous and subcutaneous margins while allowing surgical instruments to pass through the incision. Randomized trials and meta-analyses in general and biliary surgery suggest that wound edge protectors reduce superficial surgical site infection (SSI) [15, 16]; however, high-quality evidence regarding their efficacy in orthopedic surgery—and in THA in particular—remains limited. Consequently, it remains uncertain whether incorporating a WP into the Endo-DAA can yield measurable improvements in early biological markers of soft-tissue injury and recovery.

To address this lack of evidence, we conducted a prospective, randomized controlled trial of Endo-DAA THA in which the use of a disposable wound protector was the only intraoperative variable between groups. We hypothesized that incorporating a WP would reduce early soft-tissue injury and facilitate early recovery.

Methods

Study design and ethical approval

This study was designed as a prospective, randomized controlled trial conducted in accordance with the principles of the Declaration of Helsinki. The protocol was approved by the Institutional Ethics Committee (approval No. K2023-06-008), and written informed consent was obtained from all participants before enrollment. The trial was registered in the Chinese Clinical Trial Registry (ChiCTR2300076225) on September 27, 2023, prior to participant recruitment.

Patient selection

Patients scheduled to undergo primary endoscopic direct anterior approach total hip arthroplasty (Endo-DAA THA) at our institution between October 2023 and October 2024 were screened for eligibility (Fig. 1).

Fig. 1.

Flow diagram of the study design. Seventy-six patients were randomized to the wound protector (WP) group or control group. All participants completed follow-up and were included in the final analysis

The inclusion criteria were: (i) age between 18 and 85 years; (ii) diagnosis of end-stage hip disease requiring total hip arthroplasty, including osteonecrosis of the femoral head, developmental dysplasia of the hip with secondary osteoarthritis, femoral neck fracture, primary osteoarthritis, or rheumatoid arthritis; (iii) undergoing first-time unilateral total hip arthroplasty using the Endo-DAA technique; (iv) body mass index (BMI) between 18.5 and 35 kg/m²; (v) ability to understand the study protocol and comply with postoperative rehabilitation and follow-up assessments for at least six months; and (vi) provision of written informed consent before enrollment.

The exclusion criteria were: (i) revision surgery or a history of prior ipsilateral hip procedures (such as osteotomy, internal fixation, or previous arthroplasty); (ii) severe developmental dysplasia of the hip (Crowe type III–IV); (iii) presence of severe systemic comorbidities that could preclude safe anesthesia or surgery; (iv) active local or systemic infection or inflammatory condition; (v) neurological, psychiatric, or cognitive disorders that could interfere with accurate pain assessment, rehabilitation participation, or scheduled follow-up; and (vi) refusal or inability to provide informed consent.

Randomization and blinding

Patients were randomly assigned in a 1:1 ratio to the WP or control group using a computer-generated randomization sequence. Allocation concealment was ensured with sequentially numbered, opaque, sealed envelopes, which were opened in the operating room immediately before surgery. Patients and outcome assessors were blinded to group allocation, whereas the operating surgeons were necessarily unblinded. In the WP group, the wound protector was kept in place during the procedure and was removed immediately before wound closure to maintain blinding. Since no drains were placed in either group and the incision location and size were identical, the postoperative outward appearance was indistinguishable. Identical surgical dressings were applied to all patients immediately after skin closure. Consequently, postoperative care providers, physiotherapists responsible for assessing mobilization, and all outcome assessors were strictly blinded to group allocation. The randomization process was conducted by an independent coordinator not involved in surgery or data analysis.

Surgical technique

All procedures were performed under general anesthesia with patients in the supine position, without use of a traction table or urinary catheter. A standardized Endo-DAA THA was performed by a single senior surgeon (J.X.) as previously described in our earlier reports [8, 9, 17]. The surgeon had performed more than 200 Endo-DAA procedures prior to this trial. According to our previous study [17], the learning curve for this technique stabilizes after approximately 68 cases; thus, the surgeon had well surpassed the learning phase. A 5–6 cm transverse incision was made along the groin crease, predominantly lateral to the anterior superior iliac spine to minimize risk to the lateral femoral cutaneous nerve. Dissection proceeded through the Hueter interval with careful protection of the nerve branches. The anterior capsule was incised to expose the femoral neck, and the femoral head was resected according to the preoperative plan.

Acetabular preparation and cup implantation were performed under endoscopic assistance, using a 1-cm auxiliary portal when necessary to optimize working angles. On the femoral side, exposure was achieved with a proximal femoral lift retractor system to permit canal broaching and stem insertion. After trialing, component stability, offset, and limb length were confirmed; the capsule was repaired and no drains were placed. Perioperative management was standardized across patients, including prophylactic antibiotics, tranexamic acid, multimodal analgesia, and same-day mobilization.

The only standardized intraoperative difference between groups was the use of a wound protector (Fig. 2). In the WP group, a single-use wound protector (VPA60; Changzhou Weik Medical Co., Ltd., China) was placed immediately after the skin incision and deployed by compressing the inner ring to facilitate insertion, then adjusting the outer ring to ensure proper orientation and close apposition to the wound edges. The protector remained in place throughout the procedure and was removed just before wound closure and allowed adequate exposure without the use of conventional metal incision-edge retractors. In the control group, no wound protector was used, and exposure was achieved entirely with conventional metal incision-edge retractors. All other surgical steps, implants, and anesthesia protocols were standardized and identical between groups.

Fig. 2.

Application of a disposable wound protector in Endo-DAA THA. a Overview of the single-use wound protector showing the inner and outer rings with a transparent protective sleeve. b Intraoperative application during endoscopic direct anterior approach total hip arthroplasty (Endo-DAA THA). Adequate exposure was achieved without conventional metal incision-edge retractors, and surgical instruments were introduced through the protected incision

Postoperative rehabilitation and pain management

A standardized multimodal analgesia and rehabilitation protocol was strictly followed for all patients. Intraoperatively, a periarticular cocktail injection (150 mg ropivacaine and epinephrine diluted in 100 mL normal saline) was administered for local infiltration. Postoperatively, patients received a patient-controlled analgesia (PCA) pump for the first 24 h. Subsequently, oral celecoxib (400 mg/day) and gabapentin (0.3 g at bedtime) were administered for 14 days.

Mobilization was supervised by physiotherapists blinded to group allocation. Isometric quadriceps exercises and ankle pump movements were initiated immediately upon regaining consciousness. Standing and weight-bearing exercises began as soon as the patient recovered motor control and hemodynamic stability. “First no-assistive-device walking” was strictly defined as the ability to perform a sit-to-stand maneuver and walk at least 5 m without the use of assistive devices (e.g., crutches or walkers) or human assistance.

Outcome measures and sample size calculation

The primary outcome was serum creatine kinase (CK) at 24 h postoperatively, as a quantitative indicator of early muscle injury. Based on pilot data, the expected mean ± standard deviation (SD) values were approximately 650 ± 350 U/L in the control group and 450 ± 200 U/L in the WP group. Using a two-sided, two-sample t-test with α = 0.05 and 80% power for a 1:1 allocation, the required sample size was 33 patients per group. Allowing for potential loss to follow-up, at least 38 patients per group were planned for enrollment.

Secondary outcomes included perioperative and clinical measures collected at prespecified time points. Baseline demographic and preoperative variables (age, sex, BMI, operative side, hypertension, diabetes, preoperative Harris Hip Score [HHS] [18], preoperative VAS during activity, and baseline laboratory markers) were recorded for comparability. Perioperative variables included operative time, incision length, hemoglobin reduction, length of stay, and the proportion of patients achieving independent ambulation within 12 h after surgery (≤ 12 h vs. > 12 h). Radiographic assessment was performed on standard anteroposterior pelvic radiographs taken before discharge. Acetabular cup inclination and anteversion were measured to evaluate implant positioning. Laboratory markers of soft-tissue injury and inflammation comprised CK and C-reactive protein (CRP) measured preoperatively and at 24 h postoperatively, and myoglobin (MYO) measured preoperatively and at 6 h postoperatively. Pain was assessed using VAS during activity at baseline, 12 h, 24 h, 6 weeks, and 3 months postoperatively. LFCN dysesthesia was evaluated at 6 weeks, 3 months, and 6 months postoperatively. It was assessed as a binary outcome (present/absent) based on patient-reported subjective numbness or paresthesia in the anterolateral thigh distribution. This was objectively confirmed using Semmes-Weinstein monofilaments to detect sensory deficits compared with the contralateral side, performed by a trained and blinded assessor. Wound-related events were assessed within the first 6 weeks postoperatively. Superficial SSI was defined according to the CDC/NHSN criteria [19]. Wound ooze was defined as any visible serosanguinous discharge from the incision within 14 days postoperatively. Delayed wound healing was defined as persistent scabbing or incomplete skin closure at the 14-day follow-up that resolved with conservative management without the need for secondary surgical intervention. Patient-reported wound satisfaction (“very satisfied,” “satisfied,” or “unsatisfied”) was recorded at 3 months postoperatively. Functional outcomes included HHS (baseline, 6 weeks, 3 months, and 6 months) and Forgotten Joint Score-12 [20] (FJS-12; 6 weeks, 3 months, and 6 months).

Statistical analysis

Statistical analyses were performed using SPSS version 29.0.1.0 (IBM Corp., Armonk, NY, USA). The normality of continuous variables was assessed before analysis. Normally distributed data are presented as mean ± SD and compared between groups using the independent-samples t-test. Non-normally distributed data are expressed as median [interquartile range, IQR] and compared using the Mann–Whitney U test. Categorical variables were analyzed with the chi-square or Fisher’s exact test, as appropriate. Figures were generated using GraphPad Prism version 10 (GraphPad Software, San Diego, CA, USA). Between-group effect sizes were calculated to assess clinical significance. For continuous variables, effect estimates are reported as mean differences (MD) or median differences (Hodges-Lehmann estimate) with 95% confidence intervals (CIs). For binary outcomes, risk ratios (RR) with 95% CIs were calculated. No adjustments for multiplicity were applied to secondary outcomes. Consequently, findings regarding secondary endpoints should be interpreted as exploratory and hypothesis-generating. A two-tailed P < 0.05 was considered statistically significant.

Results

Patient enrollment

A total of 84 patients were assessed for eligibility. Of these, 5 did not meet the inclusion criteria and 3 declined to participate. Consequently, 76 patients were randomized to the WP group or the control group. All randomized patients completed follow-up and were included in the final analysis (Fig. 1).

Demographic and preoperative characteristics

There were no significant differences in baseline demographic and preoperative characteristics between the two groups (Table 1). Age, gender distribution, body mass index (BMI), operative side, and the prevalence of hypertension or diabetes were comparable (all P > 0.05).

Table 1.

Demographic and preoperative characteristics of the participants

	WP group (n = 38)	Control group (n = 38)	P value
Age (years)	54.3 ± 15.0	56.0 ± 15.2	0.628
Gender (female/male)	17/21	23/15	0.168
BMI (kg/m2)	23.2 ± 2.7	23.9 ± 3.1	0.306
Side (right/left)	20/18	22/16	0.645
Hypertension (n)	11 (28.9%)	9 (23.6%)	0.602
Diabetes (n)	4 (10.5%)	5 (13.1%)	0.723

WP, wound protector; BMI, body mass index; SD, standard deviation. Values are expressed as mean ± SD or number (percentage), as appropriate. P < 0.05 was considered statistically significant

Perioperative and surgical variables

The perioperative and surgical outcomes are summarized in Table 2. The median operation time, incision length, hemoglobin reduction, and length of hospital stay did not differ significantly between groups (all P > 0.05). However, a significantly higher proportion of patients in the WP group achieved walking without an assistive device within 12 h after surgery compared with the control group (86.8% vs. 55.3%, RR: 1.57; 95% CI: 1.14 to 2.16; P = 0.002). Postoperative radiographic assessment confirmed satisfactory implant positioning in all patients, with no significant differences observed in acetabular cup inclination or anteversion between the two groups.

Table 2.

Perioperative and surgical variables between the two groups

	WP group (n = 38)	Control group (n = 38)	Effect estimate (95% CI)	P value
Operation time (min)	60.0[58.8,66.3]	60.0[60.0,65.0]	0.0 (-5.0, 5.0)	0.931
Length of incision (cm)	5[5,5.3]	5[5,6]	0.0 (0.0, 0.0)	0.442
Reduction of hemoglobin (g/L)	24.2 ± 10.5	22.1 ± 12.6	2.1 (-3.2, 7.4)	0.430
Time to first no-assistive-device walking			1.57 (1.14, 2.16)	0.002
≤12 h	33 (86.8%)	21 (55.3%)
>12 h	5 (13.2%)	17 (44.7%)
Length of stay (days)	3[2,3]	3[2,4]	0.0 (-1.0, 0.0)	0.062

h, hours; WP, wound protector; CI, confidence interval; SD, standard deviation; IQR, interquartile range. Values are expressed as mean ± SD, median (IQR) or number (percentage), as appropriate. P < 0.05 was considered statistically significant

Effect Estimate: Reported as mean difference, median difference, or risk ratio, as applicable

Laboratory test results

Perioperative changes in laboratory markers are presented in Table 3. Preoperative CK, MYO, and CRP levels were comparable between the two groups (all P > 0.05). At 24 h postoperatively, the WP group had significantly lower CK (422.5 [286.8–612.8] vs. 665.5 [483.0–865.3]; median difference: -228.0; 95% CI: -354.0 to -109.0; P < 0.001; Fig. 3a) and CRP levels (30.3 [20.7–43.1] vs. 45.1 [21.7–75.4]; median difference: -14.2; 95% CI: -26.8 to -2.0; P = 0.024; Fig. 3c). Similarly, MYO levels at 6 h postoperatively were significantly lower in the WP group compared with the control group (299.9 [212.7–367.7] vs. 481.2 [369.4–597.2]; median difference: -178.5; 95% CI: -249.7 to -109.5; P < 0.001; Fig. 3b). In an exploratory analysis, patients who achieved independent ambulation within 12 h exhibited lower median CK levels compared with those who did not (493.0 vs. 582.5 U/L), although this difference did not reach statistical significance (P = 0.180).

Table 3.

Comparison of laboratory test results between groups

	Time point	WP group (n = 38)	Control group (n = 38)	Effect estimate (95% CI)	P value
CK (U/L)	Preoperative 24 h postoperative	62.5[44.3,76.5] 422.5[286.8,612.8]	64.5[48.0,73.3] 665.5[483.0,865.3]	-1.0 (-10.0, 8.0) -228.0 (-354.0, -109.0)	0.835 < 0.001
MYO (ng/mL)	Preoperative	46.6[36.0,55.3]	44.0[31.6,55.2]	1.6 (-4.6, 8.3)	0.644
	6 h postoperative	299.9[212.7,367.7]	481.2[369.4,597.2]	-178.5 (-249.7, -109.5)	< 0.001
CRP (mg/L)	Preoperative	6.0[2.9,10.1]	7.1[1.1,10.7]	-0.1 (-2.3, 2.5)	0.893
	24 h postoperative	30.3[20.7,43.1]	45.1[21.7,75.4]	-14.2 (-26.8, -2.0)	0.024

h, hours; WP, wound protector; CK, creatine kinase; MYO, myoglobin; CRP, C-reactive protein; CI, confidence interval; IQR, interquartile range. Values are expressed as median (IQR). P < 0.05 was considered statistically significant

Effect Estimate: Reported as median difference

Fig. 3.

Postoperative biomarkers, pain, and functional outcomes between groups. a Serum creatine kinase (CK) at 24 h; b serum myoglobin (MYO) at 6 h; c C-reactive protein (CRP) at 24 h; d pain visual analog scale (VAS) during activity; e Harris Hip Score (HHS); and f Forgotten Joint Score-12 (FJS-12). Preoperative (baseline) values did not differ between groups; therefore, only postoperative comparisons are shown. Data are expressed as mean ± SD or median (IQR), as appropriate. P < 0.05 (*), P < 0.01 (**), P < 0.001 (***), two-tailed. Abbreviations: h, hours; wk, weeks; mo, months

VAS pain scores across time points

VAS scores during activity at different time points are shown in Table 4. There was no significant difference in preoperative pain levels between the two groups (P = 0.730). At 12 and 24 h postoperatively, the WP group reported significantly lower VAS scores compared with the control group (12 h: 3.0 ± 0.9 vs. 4.0 ± 1.4; mean difference [MD]: -1.0; 95% CI: -1.5 to -0.4; P < 0.001; 24 h: 3.7 ± 1.1 vs. 4.8 ± 1.3; MD: -1.1; 95% CI: -1.7 to -0.5; P < 0.001; Fig. 3d). At 6 weeks and 3 months postoperatively, pain scores declined markedly in both groups, and no significant between-group differences were observed (both P > 0.05).

Table 4.

VAS scores during activity in the two groups at different time points

Time point	WP group (n = 38)	Control group (n = 38)	Effect estimate (95% CI)	P value
Preoperative	6.7 ± 1.5	6.5 ± 1.4	0.2 (-0.5, 0.9)	0.730
12 h postoperative	3.0 ± 0.9	4.0 ± 1.4	-1.0 (-1.5, -0.4)	< 0.001
24 h postoperative	3.7 ± 1.1	4.8 ± 1.3	-1.1 (-1.7, -0.5)	< 0.001
6 wk postoperative	1.3 ± 0.5	1.4 ± 0.5	-0.1 (-0.3, 0.1)	0.285
3 mo postoperative	0.6 ± 0.4	0.7 ± 0.4	-0.1 (-0.3, 0.1)	0.261

h, hours; wk, weeks; mo, months; VAS, Visual Analog Scale; WP, wound protector; CI, confidence interval; SD, standard deviation. Values are expressed as mean ± SD. P < 0.05 was considered statistically significant

Effect Estimate: Reported as mean difference

Complications and wound satisfaction

Table 5 summarizes the perioperative complications and adverse events observed in both groups. Regarding major systemic complications, no instances of dislocation, periprosthetic fracture, pulmonary embolism, deep infection, or unplanned readmission were observed in either cohort. One patient in the WP group (2.6%) developed an asymptomatic deep vein thrombosis (DVT) detected by routine ultrasound, compared with none in the control group (P = 1.000).

Table 5.

Comparison of complications between the two groups

Complications	WP group (n = 38)	Control group (n = 38)	Effect estimate (95% CI)	P value
Major complications
Dislocation	0 (0.0%)	0 (0.0%)	NA	NA
Periprosthetic fracture	0 (0.0%)	0 (0.0%)	NA	NA
Deep vein thrombosis	1 (2.6%)	0 (0.0%)	NA*	1.000
Pulmonary embolism	0 (0.0%)	0 (0.0%)	NA	NA
Periprosthetic joint infection	0 (0.0%)	0 (0.0%)	NA	NA
Blood transfusion	0 (0.0%)	0 (0.0%)	NA	NA
Unplanned readmission	0 (0.0%)	0 (0.0%)	NA	NA
Unplanned reoperation	0 (0.0%)	0 (0.0%)	NA	NA
LFCN dysesthesia
6 wk postoperative	14 (36.8%)	25 (65.7%)	0.56 (0.34, 0.91)	0.012
3 mo postoperative	6 (15.7%)	11 (28.9%)	0.55 (0.22, 1.33)	0.169
6 mo postoperative	1 (2.6%)	3 (7.8%)	0.33 (0.04, 3.11)	0.607
Wound complications	3 (7.8%)	13 (34.2%)	0.23 (0.07, 0.75)	0.005
Superficial SSI	0	0	NA	NA
Wound ooze	2 (5.2%)	8 (21.0%)	0.25 (0.06, 1.11)	0.042
Delayed wound healing	1 (2.6%)	5 (13.1%)	0.20 (0.02, 1.63)	0.202

wk, weeks; mo, months; LFCN, lateral femoral cutaneous nerve; WP, wound protector; SSI, surgical site infection; CI, confidence interval; NA, not applicable. Values are expressed as number (percentage). P < 0.05 was considered statistically significant

Effect Estimate: Reported as risk ratio

* Risk Ratio is not calculable because there were zero events in the control group

In terms of local soft-tissue complications, the incidence of LFCN dysesthesia at 6 weeks was significantly lower in the WP group than in the control group (36.8% vs. 65.7%; RR: 0.56; 95% CI: 0.34 to 0.91; P = 0.012), whereas no between-group differences were detected at 3 or 6 months (both P > 0.05). Wound complications occurred less frequently in the WP group (7.8% vs. 34.2%; RR: 0.23; 95% CI: 0.07 to 0.75; P = 0.005), mainly due to a lower incidence of wound ooze (5.2% vs. 21.0%; RR: 0.25; 95% CI: 0.06 to 1.11; P = 0.042). Delayed wound healing was also less common in the WP group, although the difference did not reach statistical significance (2.6% vs. 13.1%; RR: 0.20; 95% CI: 0.02 to 1.63; P = 0.202). No superficial surgical site infections were observed in either group.

Patient-reported wound satisfaction was significantly better in the WP group (P = 0.030). Specifically, 81.5% of patients in the WP group reported being “very satisfied” (RR: 1.55; 95% CI: 1.11 to 2.16), 13.1% “satisfied”, and 5.2% “unsatisfied”, compared with 52.6%, 28.9%, and 18.4% in the control group, respectively (Table 6).

Table 6.

Functional outcomes and patient-reported wound satisfaction between groups at different follow-up time points

HHS	WP group (n = 38)	Control group (n = 38)	Effect estimate (95% CI)	P value
Preoperative	34.7 ± 8.6	36.3 ± 7.2	-1.6 (-5.2, 2.0)	0.387
6 wk postoperative	74.1 ± 10.7	62.6 ± 12.0	11.5 (6.3, 16.7)	< 0.001
3 mo postoperative	82.7 ± 7.3	80.8 ± 8.4	2.0 (-1.6, 5.6)	0.277
6 mo postoperative	90.5 ± 5.2	89.4 ± 6.1	1.1 (-1.5, 3.7)	0.409
FJS-12
6 wk postoperative	59.7 ± 15.1	51.7 ± 16.1	8.0 (0.9, 15.1)	0.029
3 mo postoperative	69.2 ± 13.6	64.8 ± 14.5	4.4 (-2.0, 10.8)	0.177
6 mo postoperative	77.2 ± 10.7	75.0 ± 12.8	2.2 (-3.2, 7.5)	0.428
Wound satisfaction				0.030
Very satisfied	31 (81.5%)	20 (52.6%)	1.55 (1.11, 2.16)
Satisfied	5 (13.1%)	11 (28.9%)	-
Unsatisfied	2 (5.2%)	7 (18.4%)	-

wk, weeks; mo, months; HHS, Harris Hip Score; FJS-12, Forgotten Joint Score-12; WP, wound protector; CI, confidence interval; SD, standard deviation. Values are expressed as mean ± SD or number (percentage). P < 0.05 was considered statistically significant

Effect Estimate: Reported as mean difference or risk ratio, as applicable

Functional outcomes

Functional scores are summarized in Table 6. Preoperative HHS did not differ significantly between groups (P = 0.387). At 6 weeks postoperatively, the WP group had significantly higher HHS scores than the control group (74.1 ± 10.7 vs. 62.6 ± 12.0; MD: 11.5; 95% CI: 6.3 to 16.7; P < 0.001; Fig. 3e). However, no significant differences in HHS were observed between groups at 3 or 6 months after surgery (both P > 0.05). Similarly, FJS-12 scores at 6 weeks postoperatively were higher in the WP group (59.7 ± 15.1 vs. 51.7 ± 16.1; MD: 8.0; 95% CI: 0.9 to 15.1; P = 0.029; Fig. 3f), whereas differences at 3 and 6 months after surgery were not significant (both P > 0.05).

Discussion

Despite the growing adoption of anterior-based techniques for primary THA [3, 4], debate persists regarding soft-tissue stress, wound-related events, and the consistency of early recovery advantages associated with the direct anterior approach (DAA) [21–23]. In this randomized trial isolating a single intraoperative variable, the addition of a disposable wound protector (WP) was associated with lower early muscle injury and systemic inflammation, less early postoperative pain, a greater proportion of patients ambulating independently within 12 h, and fewer early lateral femoral cutaneous nerve (LFCN) symptoms and wound-related complications. Patients in the WP group also had higher functional scores at six weeks and greater incision satisfaction at three months. These benefits occurred without increasing the risk of major perioperative complications (such as dislocation or fracture) or prolonging operative time or hospital stay, and the functional outcomes converged by three months postoperatively. These findings suggest that a simple mechanical adjunct can enhance early recovery in Endo-DAA THA without altering the minimally invasive philosophy of the approach.

Just as specific markers like metal ions are employed to diagnose local tissue reactions in specific implant types [24], serum biomarkers such as creatine kinase (CK), myoglobin, and C-reactive protein (CRP) are commonly used to assess perioperative muscle injury and systemic inflammation, and have been employed in previous studies [25–27] to objectively evaluate the surgical invasiveness of total hip arthroplasty. In the present study, compared with the control group, the WP group showed significantly lower postoperative levels of creatine kinase (CK), myoglobin, and C-reactive protein (CRP), suggesting that the device may reduce soft-tissue injury and the associated inflammatory response during Endo-DAA THA. A previous high-quality randomized controlled trial [26] demonstrated no significant difference in muscle injury biomarkers such as CK and CRP between the direct anterior and posterolateral approaches, suggesting that the DAA itself is already relatively muscle-sparing. Our results suggest that incorporating a wound protector may provide additional soft-tissue protection even within an already minimally invasive approach. By mitigating early muscle injury, the WP is associated with a pattern of clinical benefits characterized by less pain, earlier ambulation, and better early function.

The concurrent reduction in postoperative CK, myoglobin, and CRP levels, together with less pain at 12–24 h and earlier mobilization, suggests a plausible biomechanical explanation. Circumferential wound-edge protection acts as a physical buffer to prevent direct instrument contact, improves endoscopic visualization by isolating subcutaneous fat from the surgical field, and likely distributes retraction forces more evenly, and limits shear, pressure, and desiccation at the skin–subcutis–fascia interface. Beyond tissue protection, such devices may improve visualization through uniform, atraumatic retraction. In the WP group, satisfactory exposure was achieved without conventional metal incision-edge retractors, which may have reduced focal edge pressure, thereby contributing to lower biomarker levels and fewer early LFCN symptoms. Consistent evidence from open hernia surgery has shown that disposable wound protectors provide superior exposure for a given incision length compared with traditional techniques, supporting this mechanism of action [28]. Within the DAA corridor, excessive static or dynamic retraction near the tensor fasciae latae–rectus interval can induce transient muscle damage or neurapraxia; reducing edge pressure plausibly mitigates these effects. Lower early pain, in turn, facilitates earlier and more confident ambulation within the first 12 postoperative hours, consistent with the higher rate of early independent mobilization observed in the WP group. A previous study by Zhao G et al. [25] demonstrated that intraoperative retraction technique and soft-tissue handling were associated with biochemical indicators of muscle injury and early postoperative recovery, further supporting the biological plausibility of our findings. We also explored the relationship between muscle injury and functional recovery. Patients who achieved independent ambulation within 12 h had lower median CK levels compared with those who did not (493.0 vs. 582.5 U/L). Although this difference did not reach statistical significance, the trend supports the hypothesis that minimizing myotrauma contributes to the physiological readiness for early mobilization, alongside other factors such as pain control and hemodynamic stability.

The reduced incidence of LFCN symptoms at six weeks further supports this mechanism. Although LFCN disturbance after DAA usually resolves [14], its frequency depends on local compression or traction near the anterior superior iliac spine. By minimizing edge friction and focal pressure, a WP likely reduces transient neurapraxia in the early postoperative period, consistent with the convergence of LFCN symptoms between groups by three to six months. Collectively, these observations support a unified mechanism in which WP use lessens muscular and neural trauma, attenuates pain and inflammation, promotes early mobilization, and yields better early functional recovery.

Evidence for wound protectors is strongest in open abdominal and hepatobiliary surgery, where multiple randomized trials and meta-analyses have demonstrated reduced superficial surgical site infection (SSI) rates with WPs [15, 29, 30]. In contrast, their role in orthopedic surgery remains largely unexplored, and existing data are limited and often confounded by bundled perioperative changes. By isolating WP use as the only intraoperative variable, the present trial addresses this gap and demonstrates consistent early-phase benefits across biochemical, pain, mobilization, and patient-reported wound outcomes. Regarding overall safety, the incidence of major systemic adverse events was negligible in both cohorts. Only one case of asymptomatic distal DVT was observed in the WP group, which resolved following standard anticoagulation, and there were no instances of dislocation or periprosthetic fracture. Unlike the general-surgery literature, no superficial SSIs were observed in either group in this study. This apparent discrepancy likely reflects the inherently low baseline risk of surgical site infection in primary THA [31, 32]. Modern arthroplasty is performed within standardized, aseptic, and Enhanced Recovery After Surgery (ERAS)-based pathways—which optimize strategies such as low-dose aspirin to mitigate PJI risks [33]—that further minimize infection risk and make it difficult for single-center studies to detect differences in such rare events. Therefore, the benefits of WPs in THA are more likely to manifest in proximal wound outcomes—such as reduced drainage (a factor also sensitive to low molecular weight heparin dosing [34]), fewer minor wound issues, and greater incision satisfaction—rather than in infection rates, unless evaluated in larger multicenter trials. It is important to emphasize that the relatively high wound complication rate observed in the control group (34.2%) was driven predominantly by minor wound ooze. In this study, we employed a highly sensitive recording standard where any visible exudation within 14 days postoperatively was captured as an event. While this rigorous criterion led to a higher numerical incidence, these occurrences were exclusively minor and self-limiting; they by no means indicate a risk of clinically significant complications, such as infection or surgical failure.

From a clinical perspective, the combination of lower early postoperative pain and a higher rate of independent ambulation within 12 h is directly actionable within the ERAS framework. This pattern enables earlier and safer mobilization, accelerates attainment of functional milestones, and may improve discharge readiness without increasing operative burden. These improvements occur without longer operative time, greater blood loss, or extended hospital stay and align with ERAS Society recommendations for hip and knee arthroplasty and with the operational needs of modern arthroplasty services [35–37]. The higher patient satisfaction with wound appearance and the reduced drainage also suggest fewer dressing changes and lower nursing workload. Given the low cost of the disposable device (approximately $14 USD), these short-term benefits—particularly the enhanced patient satisfaction and reduced early pain—present a favorable cost-benefit profile within an ERAS framework. Furthermore, regarding hospital stay, although patients in the WP group mobilized earlier, the length of stay (LOS) showed only a trend toward reduction (P = 0.062). This likely reflects a “floor effect” where discharge timing is influenced by standardized institutional observation protocols or reimbursement policies, which may mask the efficiency gains derived from earlier functional independence.

Functional outcomes, reflected by the HHS and FJS-12, were significantly higher in the WP group at six weeks but became comparable between groups at three and six months. This temporal pattern suggests that the benefit of WP use is confined to the early recovery phase, when soft-tissue irritation and pain most strongly influence mobility and perceived joint function. The improved short-term scores likely result from reduced peri-incisional trauma, less early pain, and faster restoration of confidence in ambulation. As healing progresses and rehabilitation advances, muscle strength, proprioception, and adaptation to the prosthesis become the dominant determinants of functional performance, diminishing the relative contribution of initial soft-tissue differences. This explains why functional convergence occurred by three months. Importantly, an early functional advantage remains clinically meaningful, as it aligns with ERAS principles by accelerating recovery milestones, enhancing patient satisfaction, and potentially reducing the risk of delayed discharge [37].

Strengths of this study include randomization with patient and assessor blinding, isolation of a single intraoperative variable, a standardized Endo-DAA technique and perioperative pathway, and the incorporation of objective biochemical markers alongside clinical and patient-reported outcomes. This design provides a coherent chain of evidence that strengthens causal inference. Limitations include the single-center setting and a sample size powered for biochemical endpoints rather than rare infections. Because no superficial SSIs occurred, the study lacks power to determine whether WP use might affect SSI or PJI (Periprosthetic joint infection) rates. Additionally, given the number of secondary endpoints evaluated, no statistical adjustment for multiplicity was performed; therefore, positive findings in secondary outcomes should be interpreted with caution and considered exploratory. The six-month follow-up is too short to evaluate long-term sensory abnormalities or scar appearance, and dressing changes and nursing time were not recorded. While the analgesia protocol was standardized, total opioid consumption (e.g., morphine milligram equivalents) was not quantified. Furthermore, objective functional mobility metrics, such as the Timed Up and Go (TUG) test, were not employed, limiting our assessment to binary ambulation milestones. Finally, because the cohort reflects Endo-DAA with small incisions and a standardized pathway, generalizability to other approaches, larger incisions, or higher-risk populations requires confirmation.

Future research should include multicenter, adequately powered randomized trials or registry-based studies focusing on wound-related resource utilization, such as dressing changes, unplanned clinic visits, patient-reported wound satisfaction, and infrequent infection endpoints. Mechanistic studies should quantify retraction forces and wound-edge pressures with and without WP use and assess tissue oxygenation or edema in relation to early biomarkers and mobilization milestones. Stratified analyses by BMI, soft-tissue thickness, and diabetes will help determine whether patient-related factors modify the effect of WP use on wound and recovery outcomes.

Conclusions

In Endo-DAA THA, adding a disposable WP was associated with consistent early advantages, including less tissue injury and inflammation, lower pain, faster independent ambulation, fewer early LFCN symptoms, and better wound-related outcomes, without additional operative or hospital-stay burden. Mid-term functional results converged, placing the value of WP use squarely in the early recovery phase emphasized by ERAS. Given the inherently low baseline SSI risk in primary THA, the absence of an SSI signal here is expected and underscores the need for larger infection-powered evaluations. Pending such data, these findings support considering a WP as a simple, reproducible adjunct within ERAS-oriented anterior THA pathways.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

BMI

body mass index

CI

confidence interval

CK

creatine kinase

CRP

C-reactive protein

DAA

direct anterior approach

Endo-DAA

endoscopic direct anterior approach

ERAS

enhanced recovery after surgery

FJS-12

forgotten joint score-12

HHS

harris hip score

IQR

interquartile range

LFCN

lateral femoral cutaneous nerve

MD

mean difference

MYO

myoglobin

PJI

periprosthetic joint infection

RCT

randomized controlled trial

RR

risk ratio

SD

standard deviation

SSI

surgical site infection

THA

total hip arthroplasty

VAS

visual analog scale

WP

wound protector

Author contributions

MX performed the data analysis and was a major contributor to writing the manuscript. ZC and JL collected the data used in this study. FL and JX supervised the study and critically reviewed the manuscript. FL and JX are the corresponding authors of this article. All authors read and approved the final manuscript.

Funding

This work was supported by the National Natural Science Foundation of China (Grant No. 82472473) and Fujian Provincial Natural Science Foundation Projects (Grant No.2022J011016, 2023J011200, 2025J01062).

Data availability

The datasets generated and/or analyzed during the current study are not publicly available due to institutional regulations protecting patient privacy. However, de-identified data are available from the corresponding author on reasonable request, subject to approval by the institutional ethics committee.

Declarations

Ethics approval and consent to participate

This study was approved by the Institutional Ethics Committee (approval No. K2023-06-008) and conducted in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained from all participants prior to enrollment. The trial was prospectively registered with the Chinese Clinical Trial Registry (ChiCTR2300076225; registered on September 27, 2023).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.