A study on the early clinical outcomes and learning curve of ARTHROBOT-assisted total hip arthroplasty under the ERAS concept Authors: Mingyou Wang, Yan Wen, Mingli Wang, Yuping Lan, Zhuodong Tang, Xunzhou Song, Zaihai Yang, Tao Dai, Hongping Wang Published: 2025 ABSTRACT AbstractAs robotic-assisted systems gain prominence in total hip arthroplasty (THA), their integration within Enhanced Recovery After Surgery (ERAS) pathways presents new opportunities for precision and early recovery. This retrospective study compared 104 patients undergoing THA performed either conventionally or assisted by the domestic ARTHROBOT robotic system under an ERAS protocol. While the robotic group experienced longer operative time and slightly longer incisions, it demonstrated significantly superior accuracy in component positioning. Acetabular component positioning was more accurate with ARTHROBOT assistance, achieving a 92.3% rate within the Lewinnek safe zone versus 75.0% conventionally, along with better control of leg length discrepancy and improved restoration of femoral and combined offsets. Although patients in the robotic group reported higher pain scores on postoperative day 3, this difference resolved by one month. Importantly, both groups achieved comparable outcomes in terms of early full weight-bearing time, hospital stay, functional recovery (Harris Hip Score, range of motion), complication rates, and high patient satisfaction (over 94%). The learning curve for the ARTHROBOT system was approximately 20 cases. These findings support the clinical integration of the domestic ARTHROBOT system in ERAS-based THA, enabling enhanced implant precision without compromising early safety or recovery.Keywords: Total hip arthroplasty, Robot-assisted surgery, ARTHROBOT, Enhanced recovery after surgery, Learning curve, Implant positioning accuracy Introduction Total Hip Arthroplasty (THA) is the gold standard surgical procedure for treating end-stage hip disorders [1]. With the accelerating ageing of the population, the demand for THA is increasing annually, placing greater demands on the precision of the procedure and the quality of post-operative recovery. Conventional THA relies heavily on the surgeon’s experience and tactile feedback, introducing a degree of subjectivity and variability in aspects such as prosthesis positioning and limb length control [2]. To enhance surgical precision and reproducibility, robot-assisted surgical systems have emerged. These systems facilitate the individualised and precise placement of prostheses through pre-operative three-dimensional planning and intra-operative accurate execution, thereby theoretically reducing the risk of complications arising from prosthesis malpositioning [3, 4]. In recent years, domestically produced surgical robot platforms in China have developed rapidly. Among them, the ARTHROBOT (Jianjia) robotic system, as one of the representative technologies, has been progressively introduced into clinical practice [5]. Concurrently, with the widespread adoption of the Enhanced Recovery After Surgery (ERAS) concept, higher demands have been placed on the perioperative management of THA. While the core tenet of ERAS is to minimize surgical stress and promote swift recovery via optimized perioperative management [6, 7], the application of the ARTHROBOT robotic system within this framework, particularly concerning early results and the learning curve, remains underexplored. Therefore, this study aimed to evaluate the early clinical efficacy of ARTHROBOT-assisted THA within an ERAS protocol, and to analyse its learning curve, by conducting a retrospective cohort study in comparison with conventional THA. It is anticipated that this investigation will provide real-world evidence for the clinical value of this technology and furnish an empirical basis for optimising its training and implementation strategies. The findings are presented herein. Methods General information Ethical approval for this investigation was granted by the Ethics Committee of Panzhihua Central Hospital. Informed consent was secured from all participating patients or their legal representatives. This study enrolled patients receiving THA at our center from January 2024 to March 2025. Based on the outcomes of doctor-patient communication and the different surgical techniques, they were divided into two groups: the Conventional Total Hip Arthroplasty (CTHA) group and the Robot-Assisted Total Hip Arthroplasty (RATHA) group. Inclusion Criteria: ① Patients with a primary diagnosis of primary hip osteoarthritis (OA), osteonecrosis of the femoral head (ONFH, Ficat III-IV), or developmental dysplasia of the hip (DDH, Crowe I-II), scheduled for their first unilateral THA. ② Aged between 18 and 80 years. ③ Availability of complete clinical and radiological data, including comprehensive pre-operative planning data, intra-operative records, and scheduled post-operative follow-up data (encompassing at least the 3-month and 6-month post-operative intervals). ④ All procedures were performed by the same team of senior consultant surgeons who were proficient in conventional THA but were first-time users of the ARTHROBOT robotic system. Exclusion Criteria: ① A history of previous ipsilateral hip surgery (e.g., osteotomy, internal fixation). ② THA performed for inflammatory arthropathies such as rheumatoid arthritis or ankylosing spondylitis, or for post-traumatic arthritis. ③ Severe hip deformity (e.g., Crowe type III or IV DDH) or severe deformities requiring femoral osteotomy. ④ With coexisting neurological or muscular diseases (e.g., sequelae of stroke, Parkinson’s disease, myasthenia). ⑤ Intra-operative or post-operative occurrence of major complications, as these would directly confound the assessment of early recovery indicators and the learning curve. ⑥ Loss to follow-up or incomplete follow-up data rendering the case unsuitable for the study analysis. This study was reviewed and approved by the Scientific Research Ethics Committee of Panzhihua Central Hospital (Approval No.: Pan Ke Lun Shen Zi No. 2024-008). Surgical technique All surgeons completed a manufacturer-provided standardized training program encompassing theoretical, plastic model, sessions prior to performing the first clinical RATHA procedure. All procedures all procedures were conducted by experienced surgeons. General anaesthesia (intravenous-inhalation combined) was administered to all patients, who were positioned laterally. A posterolateral approach was utilised for the total hip arthroplasty, following standard skin preparation and draping. An intravenous dose of 1 g tranexamic acid was given 15 minutes prior to skin incision. Following closure of the joint capsule, an intra-articular injection of 3 g tranexamic acid was administered, accompanied by a peri-articular ‘cocktail’ injection with a standardized composition (Ropivacaine 200 mg, Dexamethasone 1 g, Epinephrine 1 g, diluted with normal saline to a total volume of 60 ml) [8]. Cefuroxime sodium (1.5 g) was administered routinely 30 minutes pre-operatively. An intravenous dose of 1 g tranexamic acid was given 15 minutes prior to skin incision. Following closure of the joint capsule, an intra-articular injection of 3 g tranexamic acid was administered, accompanied by a peri-articular ‘cocktail’ injection [8]. RATHA group The ARTHROBOT robotic system was employed. Pre-operatively, surgical planning was conducted on the ARTHROBOT platform using CT scans of the hip and femur. Intra-operatively, registration of the robotic arm and the signal probe was carried out. For pelvic tracking, two Kirschner wires were placed into the anterosuperior iliac spine on the involved side to affix the reference array. A marker was placed on the lateral border of the patella at the distal femur. A standard posterolateral approach was utilized, with a layered incision made through the skin and subcutaneous tissue. The gluteus maximus was split, and dissection was carried down to the joint capsule. Following exposure of the femoral neck, an osteotomy was carried out to transect it 0.5 to 1.0 cm above the lesser trochanter. Guided by prompts on the computer screen, 25 anatomical landmarks of the acetabulum were collected using the signal probe for further verification. The robotic arm was then utilised to ream the acetabulum to the target position with the pre-operatively planned reamer size. Subsequently, the pre-operatively planned acetabular component was implanted to the target angle via the robotic arm. The proximal femur was exposed by flexing and internally rotating the lower limb. Following broaching, the pre-operatively selected femoral stem and head components were implanted. The hip was then reduced to complete the procedure. The positioning pins were removed, and the joint capsule, subcutaneous tissue, and skin were closed in layers. The wound was dressed with a sterile gauze, concluding the operation. CTHA group In the CTHA group, pre-operative planning, including the preliminary selection of prosthesis type and size, was based on anteroposterior pelvic radiographs, anteroposterior and oblique radiographs of the affected hip, and hip CT scans. The surgeon performed the procedure using conventional instruments. After adequate exposure of the femoral neck, an osteotomy was performed proximal to the lesser trochanter, guided by surgical experience. The acetabulum was fully exposed and reamed manually. Acetabular cup positioning was determined by the surgeon’s experience, with reference to the patient’s position and acetabular reamer angle guide. The depth of reaming was judged by the bleeding of the bone bed. The acetabular component and liner were then implanted manually. Subsequently, the femoral canal was broached, and the femoral component was implanted. The final sizes of the femoral stem and head were determined intra-operatively using trial reductions, assessing bilateral patellar and heel alignment, as well as joint capsule tension. Postoperative management All patients received a standardised postoperative analgesic regimen comprising nerve blockade, patient-controlled analgesia (PCA) pumps, and intravenous flurbiprofen axetil. Antimicrobial prophylaxis with cefuroxime sodium (1.5 g, administered every 8 h) was continued for 2 days. For thromboprophylaxis, low-molecular-weight heparin was administered for one week, followed by oral rivaroxaban; this was supplemented with intermittent pneumatic compression devices applied to the lower limbs. Rehabilitation exercises and weight-bearing ambulation were commenced on postoperative day 1. Outcome measures The collected data encompassed baseline demographics and operative details, which included age, sex, height, and weight, operative side, incision length, and operative time (with sub-categories for acetabular reference array placement time, registration time, and osteotomy time). Laboratory parameters, including haemoglobin (HB), haematocrit (HCT), white blood cell count (WBC), erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP) levels, were collected to evaluate perioperative blood loss and inflammatory response. Postoperative anteroposterior pelvic radiographs and CT scans were obtained to assess implant positioning. The following parameters were measured: leg length discrepancy (LLD), acetabular abduction angle (AAA), acetabular anteversion (AA), femoral offset deviation (FOD), acetabular offset deviation (AOD), and combined offset deviation (COD). The measurement methodologies are illustrated in Fig. 1. Clinical outcomes were assessed by collecting the time to full weight-bearing ambulation, the visual analogue scale (VAS) for pain, the Harris Hip Score, hip range of motion (ROM) in flexion-extension and internal-external rotation, the accuracy rate of femoral stem size prediction, complication rates, and in-hospital patient satisfaction scores. Patients were followed up at the following perioperative time points: preoperatively, and postoperatively on day 1, day 3, day 5, at 1 month, 3 months, and 6 months. Measurement methodology for femoral offset, acetabular offset, combined offset, and the change in the centre of rotation before and after THA. A: Pre-operative status. B: Post-operative status. L1: The line connecting the superior aspects of the bilateral lesser trochanters. L2: The line connecting the bilateral teardrops. L3: The perpendicular line from the symphysis pubis to the teardrop line. L4: The perpendicular line from the affected-side teardrop to the bilateral teardrop line. L5: The central axis of the femoral medullary canal. The red solid line represents the femoral offset. The blue dashed line represents the acetabular offset. The white dashed line indicates the distance from the centre of rotation to the midline. The green solid line indicates the distance from the centre of rotation to the teardrop line. The point of intersection marked by the arrows denotes the centre of rotation Statistical analysis Statistical analyses were performed using SPSS 29.0. Continuous data, expressed as mean ± standard deviation, were compared between groups with the independent samples t-test (parametric) or non-parametric equivalents. Intra-group comparisons employed one-way ANOVA for normally distributed data. Categorical variables were analyzed using the Chi-square or Fisher’s exact test. A two-sided P < 0.05 defined statistical significance for all tests. Results General results This study comprised a cohort of 104 patients. The two groups demonstrated no significant differences (P > 0.05) across a range of baseline demographics and clinical characteristics, such as age, sex, BMI, operative side, underlying hip pathology, symptom duration, functional and pain scores, and hip range of motion (Table 1). Every surgical procedure was deemed successful, accompanied by uncomplicated primary healing of all incisions. The incision length in the RATHA group was approximately 1 cm longer than that in the CTHA group. Furthermore, the operative time in the RATHA group was significantly longer than in the CTHA group by approximately 14 min (P < 0.05). Comparison of the cohorts revealed no significant differences regarding intra-operative blood loss, duration of hospitalization, or overall hospital costs (P > 0.05) (Table 2). Table 1Demographic and clinical profiles by study groupIndicatorCTHA Group (n = 52)RATHA Group (n = 52)χ2/t/Z P Age (year)67.4 ± 9.565.5 ± 8.91.0520.295Sex (F/M, female%)23/2925/270.1550.649Body mass index (kg/m2)25.8 ± 2.425.2 ± 2.21.3290.187Operated side (R/L)25/2727/250.1540.695 Hip pathologies OA16170.3810.827ONFH2930DDH75Disease Duration (months, ± s)89.6 ± 34.495.34 ± 40.31−0.7810.437 Preoperative hip function Harris score68.6 ± 9.365.2 ± 10.41.7570.082VAS pain score5.1 ± 0.94.9 ± 0.91.1330.260Hip flexion-extension ROM (°)84.1 ± 9.986.5 ± 10.2−1.2180.226Hip internal-external rotation ROM (°)29.2 ± 6.930.1 ± 7.2−0.6510.517 Demographic and clinical profiles by study group Table 2Comparison of perioperative data by study groupIndicatorCTHA Group (n = 52)RATHA Group (n = 52)χ2/t P Surgical dataTotal incision length (cm, ± s)14.9 ± 1.315.9 ± 1.2−4.076< 0.001Operative time (min, ± s)55.7 ± 7.669.6 ± 6.3−10.154< 0.001Intra-operative blood loss (ml, ± s)150.5 ± 26.2157.7 ± 23.4−1.4780.142Wound healing (n, Grade A/B/C)52/0/052/0/01.000Number of postoperative hospital stay (d)7.2 ± 0.67.0 ± 0.51.8470.068Total hospitalisation costs (in ten-thousands RMB, ± s)3.0 ± 0.53.1 ± 0.8−0.7640.446 Comparison of perioperative data by study group Comparison of inflammatory and radiographic outcomes at postoperative 3 days Levels of key inflammatory markers (WBC, ESR, CRP) were comparable between the two groups at postoperative day 3, with no statistically significant differences. A significant intergroup difference was observed in the magnitude of LLD, which was markedly smaller in the RATHA cohort than CTHA controls (P < 0.001). At the 3-month follow-up, all patients in both groups exhibited well-positioned hip prostheses without signs of loosening. There was no significant inter-group difference in the AAA. However, the AA in the RATHA group was closer to the target value of 15° (P < 0.05). A significantly larger percentage of patients in the RATHA cohort (92.3%) had the acetabular simultaneously placed within the target Lewinnek safe zone, a result that was statistically superior to the CTHA group (P < 0.05). Four cases (7.7%) in the RATHA group fell outside the Lewinnek zone. Analysis attributed two to early learning-phase registration errors, one to challenging anatomy with obscured landmarks, and one to a deliberate intraoperative decision to prioritize bone coverage over the planned angle. Additionally, the deviations in FOD, AOD and COD from the contralateral side were significantly smaller in the RATHA group (P < 0.05), indicating a closer restoration of the native hip biomechanics. The pre-operative planning accuracy for the femoral stem size was 98.1% in the RATHA group. In only one case was the stem size intra-operatively upsized due to concerns over soft tissue laxity and potential instability (Table 3). Table 3A comparative analysis of early postoperative inflammatory markers and radiographic outcomesIndicatorCTHA Group (n = 52)RATHA Group (n = 52)χ2/t P WBC (*109/L)8.4 ± 1.98.6 ± 2.5−0.4590.647ESR (mm/L)40.3 ± 18.638.6 ± 15.20.5100.611CRP (mg/L)51.0 ± 15.349.5 ± 14.50.5130.609LLD (mm, ± s)0.6 ± 0.20.4 ± 0.25.099< 0.001AAA (°, ± s)42.3 ± 7.240.5 ± 6.81.3110.193AA (°, ± s)23.2 ± 6.019.6 ± 5.73.1370.002Deviation of femoral offset (FOD) (mm, ± s)0.8 ± 0.40.3 ± 0.28.062< 0.001Deviation of acetabular offset (AOD) (mm, ± s)0.6 ± 0.30.3 ± 0.16.841< 0.001Deviation of combined offset (COD) (mm, ± s)0.9 ± 0.60.4 ± 0.35.375< 0.001Position within Lewinnek safe zone (n, %)39 (75.0)48 (92.3)5.6960.017Accuracy of femoral stem size prediction (n, %)–51 (98.1)–– A comparative analysis of early postoperative inflammatory markers and radiographic outcomes Clinical outcomes All patients in both groups completed the 6-month follow-up. The time to full weight-bearing was comparable between the study groups, with no statistical significance (P > 0.05), with a mean ranging from 3.2 to 3.4 days. Longitudinally, both cohorts exhibited marked improvements across all assessed metrics, including reductions in VAS pain scores and increases in Harris Hip Scores and hip ROM (flexion-extension and rotation), all with high statistical significance (P < 0.001). A comparative assessment of early postoperative pain revealed that the VAS score in the RATHA group was elevated relative to the CTHA group at the 3-day mark. At all other corresponding time points, the VAS scores, hip ROM (flexion-extension and rotation) remained comparable between the two cohorts, with no statistical significance (P > 0.05) (Table 4). Table 4Follow-up outcomes across the cohorts ( ± s)IndicatorCTHA Group (n = 52)RATHA Group (n = 52)χ2/t/Z P Time to Full Weight-Bearing (d)3.2 ± 0.53.4 ± 0.7−1.6770.097 Harris score Postoperative Day 357.6 ± 5.259.4 ± 7.8−1.3850.1691 Month Postoperatively73.6 ± 7.372.2 ± 9.40.8480.3983 Months Postoperatively79.6 ± 9.980.3 ± 7.7−0.4020.6886 Months Postoperatively86.3 ± 8.787.3 ± 9.9−0.5470.585P< 0.001< 0.001 VAS pain score Postoperative Day 34.5 ± 1.04.9 ± 0.7−2.3630.0201 Month Postoperatively1.7 ± 0.61.6 ± 0.41.0000.3203 Months Postoperatively1.0 ± 0.60.9 ± 0.50.9230.3586 Months Postoperatively0.5 ± 0.60.6 ± 0.5−0.9230.358P< 0.001< 0.001 Hip flexion-extension ROM (°) Postoperative Day 360.3 ± 4.661.9 ± 5.1−1.6800.0961 Month Postoperatively85.1 ± 4.984.5 ± 5.20.6060.5463 Months Postoperatively97.5 ± 6.496.4 ± 6.80.8490.3986 Months Postoperatively109.5 ± 9.4110.9 ± 8.9−0.7800.437P< 0.001< 0.001 Hip internal-external rotation ROM (°) Postoperative Day 318.6 ± 4.919.3 ± 4.50.7590.4501 Month Postoperatively31.2 ± 4.930.4 ± 5.20.8070.4213 Months Postoperatively40.5 ± 4.641.4 ± 4.3−1.0310.3056 Months Postoperatively47.2 ± 6.345.4 ± 5.61.5400.127P< 0.001< 0.001 Follow-up outcomes across the cohorts ( ± s) Complication rates and patient satisfaction Postoperative muscular calf vein thrombosis occurred in patients from both groups (4 cases in the CTHA group and 5 in the RATHA group). One patient in the CTHA group experienced a hip dislocation one month postoperatively following a twisting injury while squatting. This was managed with closed reduction under anaesthesia and spica casting. The cast was removed at 1.5 months post-injury, at which point the hip was confirmed to be stable. In the RATHA group, one patient developed wound discharge in the absence of infection, which resolved with routine wound care. No significant disparity was observed in the overall complication rates when comparing the two cohorts (P > 0.05). There were no instances of serious complications, such as robot-related adverse events, neurovascular injury, or acute periprosthetic joint infection. Patient satisfaction during hospitalization was high in both cohorts, with rates exceeding 94.23%, and no significant difference was found between them (Table 5). Table 5Comparative analysis of perioperative complications and satisfaction outcomesIndicatorCTHA Group (n = 52)RATHA Group (n = 52)χ2/t/Z P Complication rate (n, %)Intramuscular venous thrombosis (n)45DVT (n)00PE (n)00Hip dislocation10Periprosthetic fracture000.1020.750Periprosthetic joint infection00Wound discharge01Other (Robot-related adverse events)00In-hospital patient satisfaction (n, %)50 (96.15)49 (94.23)1 Comparative analysis of perioperative complications and satisfaction outcomes Learning curve The learning curve was assessed using CUSUM analysis, with the peak of the curve marking the transition to proficiency. In the RATROBOT group, a cubic model provided the optimal fit (P < 0.05, R² = 0.985). The resulting equation, CUSUM(n) = 0.009n³ − 1.128n² + 34.490n − 2.623, identified the learning phase inflection point at 20 cases, representing 33.33% of the series (Fig. 2). Longitudinal tracking of intraoperative phases revealed that the reduction in total operative time was primarily driven by a significant decrease in the robotic execution phase, which plateaued after the initial learning period. Learning curves of RATHA Group Typical cases See Figs. 3 and 4. Representative Case 1. A 50-year-old male patient with osteonecrosis of the left femoral head (Ficat stage IV) underwent conventional total hip arthroplasty. (A) Pre-operative anteroposterior pelvic radiograph showing collapse of the left femoral head, subluxation of the hip joint, and acetabular degenerative changes. (B) Post-operative anteroposterior pelvic radiograph demonstrating satisfactory prosthesis positioning, with the acetabular component abduction angle measuring 35°. (C) Post-operative CT scan (or radiograph) revealing an acetabular component anteversion angle of 11.1° Representative Case 2. A 67-year-old male patient with osteonecrosis of the left femoral head (Ficat stage III) underwent ARTHROBOT-assisted total hip arthroplasty. (A) Pre-operative anteroposterior pelvic radiograph showing collapse of the left femoral head. (B) Intra-operative photograph demonstrating the sequential registration of the acetabulum as guided by the system prompts. (C) Intra-operative photograph showing acetabular reaming being performed according to the pre-operatively planned angle via the robotic arm. (D) Intra-operative screenshot from the robotic system interface displaying the measured acetabular component angles after implantation (abduction 38.5°, anteversion 21.6°). (E) Post-operative anteroposterior pelvic radiograph demonstrating satisfactory prosthesis positioning, with the acetabular abduction angle measuring 39.0°. (F) Post-operative CT scan (or radiograph) confirming an acetabular component anteversion angle of 21.6° Discussion Total hip arthroplasty stands as one of the most successful surgical interventions for end-stage hip disorders. With an accelerating global ageing population and the widespread adoption of surgical techniques, the volume of THA procedures is exhibiting a sustained upward trend worldwide. Against this backdrop, the demand for surgical precision and individualisation has reached an unprecedented level. The precise placement of the acetabular component is a cornerstone of successful THA, directly influencing joint stability, functional longevity, and long-term patient satisfaction. The concept of the ‘safe zone’ for acetabular cup positioning, as proposed by Lewinnek et al. (anteversion 5°–25°, abduction 30°–50°), has become a classic benchmark [9]. Positioning the prosthesis within this zone is associated with a dislocation rate of 1.5%, whereas deviation beyond these boundaries can lead to a sharp increase in dislocation risk to 6.1% [10]. The restoration of hip offset in THA is crucial for reinstating abductor muscle strength, maintaining hip joint stability, and reducing polyethylene liner wear [11–15]. A discrepancy in femoral offset restoration exceeding 5 mm has been shown to increase linear wear by 33% and volumetric wear by 32% [16]. Similarly, a combined offset discrepancy of more than 5 mm adversely affects abductor muscle lever arm and strength, demonstrating a negative correlation [15, 17]. However, achieving these targets poses a significant challenge in conventional surgery. Studies indicate that even when performed by experienced surgeons, the accuracy rate of acetabular cup placement within the desired target zone in conventional THA ranges only from approximately 55% to 78% [18–20]. This rate is likely to be even lower for early-career surgeons who are navigating their learning curve. This inherent subjectivity and variability, contingent upon the surgeon’s experience and tactile feedback, represent a critical bottleneck impeding the overall enhancement of THA quality. To overcome this bottleneck, robot-assisted surgical systems have emerged. Internationally, robotic-assisted THA platforms, exemplified by the MAKO system, have been in development for over a decade. Their efficacy in enhancing the precision of prosthesis placement has been substantiated by extensive clinical research, indicating a relatively mature level of technological refinement [21]. In contrast, the clinical application of orthopaedic surgical robots in China commenced later, and the technology was initially concentrated in large tertiary medical centres. In recent years, with the rise of ‘Intelligent Manufacturing in China’, domestically developed orthopaedic robot platforms such as ARTHROBOT and HURWA have advanced rapidly [22]. This progress has broken the long-standing monopoly of imported technology, creating new opportunities for its wider dissemination and adoption. Consequently, a pivotal challenge we now face is determining how to deploy these advanced domestic technologies efficiently and standardisedly to serve the goals of clinical precision medicine, while simultaneously fostering their ongoing innovation and development. This investigation focused on the ARTHROBOT robotic system for THA within an ERAS pathway, providing a concurrent assessment of its early clinical outcomes and associated learning curve. Our findings demonstrate that the RATHA group achieved superior radiographic precision postoperatively. Specifically, the acetabular anteversion was closer to the target angle, and the restoration of LLD, FOD, AOD, and COD were all significantly better than in the conventional THA group. More importantly, a significantly higher proportion of prostheses (92.3%) in the RATHA group were positioned within the Lewinnek safe zone, compared to 75.0% in the conventional group. These results are highly consistent with numerous domestic and international studies. For instance, Liao et al. [23] reported that Mako robotic-assisted THA achieved a combined offset deviation within 5 mm in 71.4% of cases, compared to only 45.2% with conventional technique. Similarly, Jia et al. [18] confirmed that ARTHROBOT-assisted THA enables precise restoration of hip offset and LLD. Precise prosthesis positioning and biomechanical reconstruction are paramount for restoring the abductor muscle lever arm, maintaining joint stability, reducing edge loading and polyethylene wear, and minimising the risk of fracture in ceramic components. Effective control of LLD is directly linked to postoperative gait, patient satisfaction, and long-term function. In support of this, Li et al. [24] found that robotic assistance enabled 97.4% of patients to achieve an LLD within the clinically acceptable threshold of 5 mm. CUSUM analysis demonstrated that operative proficiency with the ARTHROBOT system was achieved after approximately 20 cases. Prior to surpassing this threshold, the operative time was significantly longer than that of the conventional group, primarily attributable to the time required for system setup, registration, and verification. As surgical experience accumulated, the total operative time progressively decreased, eventually approaching that of conventional THA. This observation aligns with the learning curve reported by Dai et al. [25] for the Mako robotic system. Once the surgeon crosses the learning curve, surgical efficiency improves markedly, while the precision advantages conferred by the robotic system are consistently maintained. Within the ERAS context, although the initial modest increase in operative time warrants consideration, our study confirmed that it did not translate into prolonged hospital stays or an increased complication rate. This indicates that the technology can be effectively integrated into the ERAS pathway. Based on our learning curve analysis, we recommend a structured implementation program for new centres, including manufacturer training, proctoring for the initial 5–10 cases, and a phased commencement with standard anatomical cases to ensure safe and efficient adoption. International studies suggest that robotic-assisted THA can reduce soft tissue trauma through its precise execution, thereby alleviating postoperative pain, dampening the short-term inflammatory response, and decreasing the incidence of complications such as deep vein thrombosis and surgical site infection. In the long term, this may potentially lead to a reduced risk of revision surgery [26–28]. However, some domestic studies have shown that robotic-assisted THA does not demonstrate a significant advantage over conventional methods in terms of hospital stay duration, short-term hip function, or overall complication rates [18]. A potential explanation for this observation involves the later market entry and subsequent evolution of domestic robotic platforms, the surgeon learning curve (encompassing factors such as operational unfamiliarity and prolonged registration time), as well as lower patient awareness and heightened anxiety regarding the new technology [29]. It is particularly noteworthy that robotic-assisted THA requires additional steps such as pelvic reference array placement and bony registration, which can lead to prolonged operative time and marginally longer incisions. This presents a specific challenge for patient management within the ERAS framework. Against this backdrop, the integration of robotic-assisted THA into a well-established ERAS pathway is paramount. Our department, designated as a National Health Commission pilot centre for ERAS in orthopaedics, has developed a comprehensive perioperative management protocol for THA since the introduction of the ERAS concept in 2015, with demonstrated efficacy. To address the specificities of robotic-assisted THA, our management strategy focuses on: (1) intensifying preoperative education and psychological support to alleviate patient apprehension and anxiety towards the new technology; (2) optimising wound and pain management protocols in response to the potentially extended operative duration; and (3) formulating and rigorously implementing an individualised, goal-oriented postoperative rehabilitation plan. Despite a longer operative duration, this study confirms that RATHA within an ERAS protocol did not significantly differ from conventional THA in key recovery metrics, including the time to full weight-bearing, length of hospital stay, and the restoration of early function and pain relief. This indicates that the technical complexity of the robotic system did not impede the achievement of early accelerated recovery for patients. The combination of ERAS and RATHA exemplifies a profound integration of precision surgery and the ERAS philosophy. On one hand, robotic technology, through pre-operative 3D planning and intra-operative precise execution, facilitates individualised prosthesis placement and soft tissue protection, thereby creating favourable biological conditions for rapid recovery from the outset. On the other hand, the ERAS concept, by optimising perioperative management, effectively mitigates the surgical stress response, thereby safeguarding the quality and safety of recovery for patients undergoing robot-assisted surgery. The decision to integrate ARTHROBOT within our ERAS protocol was driven by both a clinical hypothesis—that surgical precision enhances the biological potential for rapid recovery—and an institutional alignment with national priorities in adopting domestic medical technology within standardized care pathways. Although the VAS pain score was marginally higher in the RATHA group at postoperative day 3—potentially attributable to the longer operative time, the additional incision for pelvic marker placement, intra-operative soft tissue retraction, or the registration process—this difference had resolved by the one-month follow-up and did not impact the overall recovery trajectory. The combination of comparable complication rates and the absence of serious robotic system-related adverse events provides evidence for the clinical safety of the ARTHROBOT system. As a representative domestic platform, the ARTHROBOT system demonstrates clinical efficacy comparable to that of imported products, such as MAKO. The breakthrough offered by domestic platforms extends beyond technological parity; it lies in superior cost controllability, enhanced technical adaptability, and the potential for deeper integration within the domestic healthcare ecosystem. With ongoing technological iterations and accumulated clinical experience, domestically produced robots are poised to play a pivotal role in elevating the overall standard of joint replacement surgery in China and facilitating the dissemination of advanced surgical technology to a broader range of healthcare institutions. The promotion of robotic-assisted THA must consider the varying levels of acceptance among surgeons of different seniority [30]. This study confirms that the prolonged operative time resulting from steps such as pelvic registration and referencing may limit its adoption among senior surgeons who prioritise operational efficiency. For early-career surgeons, however, the technology demonstrates significant value in surgical education and quality assurance. The pre-operative quantitative planning and intra-operative real-time navigation it provides create a form of ‘digital certainty’. This effectively compensates for the relative lack of experiential knowledge in less experienced surgeons, reduces intra-operative decision-making time, and enhances surgical precision and safety, thereby significantly bolstering their surgical confidence. Therefore, robotic-assisted technology serves not only as an instrument for achieving utmost precision but also as a vital platform that supports early-career surgeons in navigating their learning curve safely and standardisedly. The initial trade-off between increased operative time and superior precision must be contextualized within the learning curve. As proficiency is achieved, the time penalty diminishes while the accuracy advantage remains. In current practice, this supports the selective use of robotic assistance for patients where long-term implant longevity is paramount or where anatomy increases the risk of conventional mispositioning. It also highlights its role as an educational tool for surgical training. This study has several limitations. Firstly, its design as a single-centre, retrospective analysis with a relatively limited sample size may introduce selection bias. Secondly, the follow-up duration was short (a maximum of 6 months), which precludes the assessment of the long-term survivorship and functional outcomes of robotic-assisted THA. Thirdly, as all procedures were conducted by experienced surgeons, the generalisability of these findings to practitioners early in their learning curve requires further validation. Furthermore, this study excluded patients with severe hip deformities. While robotic systems hold promise for managing complex anatomy through enhanced 3D planning, their performance and registration reliability in such scenarios (e.g., Crowe III-IV DDH) require dedicated investigation, as landmark identification and software planning algorithms may face significant challenges. Conclusion Within an ERAS protocol, domestically developed ARTHROBOT-assisted THA demonstrates superior accuracy in prosthesis positioning, limb length control, and biomechanical reconstruction compared to conventional surgery, alongside a favourable early-term safety profile and clinical efficacy. The associated learning curve, approximately 20 cases, indicates that this technology can be efficiently integrated into the clinical workflow following systematic training and initial procedural experience. This study provides robust empirical evidence to support the wider clinical adoption of the ARTHROBOT system, signifying a substantive advancement for domestic orthopaedic robotic technology on its trajectory towards precision and intelligent surgery. Abbreviations Total Hip Arthroplasty Conventional Total Hip Arthroplasty Robot-Assisted Total Hip Arthroplasty Enhanced Recovery After Surgery Developmental Dysplasia of the Hip Osteoarthritis Osteonecrosis of the Femoral Head Acetabular Abduction Angle Acetabular Anteversion Acetabular Offset Deviation Combined Offset Deviation Femoral Offset Deviation A domestic robotic-assisted surgical system (Jianjia) Body Mass Index C-Reactive Protein Erythrocyte Sedimentation Rate White Blood Cell Count Haemoglobin Haematocrit Cumulative Summation Leg Length Discrepancy Patient-Controlled Analgesia Range of Motion Visual Analogue Scale Author contributions MYW and YW participated in the study design, data collection and analysis, and wrote the main manuscript; YPL and WHP performed the surgical procedures and participated in the perioperative management; MLW, XZS, ZHY and TD contributed to data curation and statistical analysis; HPW supervised the entire study, including conceptualization, methodology, and final review. All authors read and approved the final manuscript. Funding This research was funded by Sichuan Provincial Primary Health Development Research Center in 2024, North Sichuan Medical College (No. SWFZ24-Q-86) and Chengdu High-tech Medical Association “2023 annual Flurbiprofen gel paste treatment of osteoarthritis special research fund” project (2024013), Sichuan Provincial Rehabilitation Medical Association’s 2025 Annual Scientific Research Projects (No.SCKFKY20250210), 2025 Annual Science and Technology Bureau of Panzhihua City project (2025ZD-S-2), 2024 Annual Science and Technology Bureau of Panzhihua City project (2024ZD-S-87), 2023 Annual Science and Technology Bureau of Panzhihua City project (2023ZD-S-5) who is not involved in study design, data collection, analysis and interpretation or manuscript preparation. Data availability No datasets were generated or analysed during the current study. Declarations Competing interests The authors declare no competing interests. Ethics approval and consent to participate Ethical approval for this investigation was granted by The Institutional Review Board of Panzhihua Central Hospital (Pankelun Trial No. [2024-008]). Written informed consent was secured from all individual participants included in the study. Consent for publication Written informed consent was obtained from the patient for publication of this paper. . Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Mingyou Wang and Yan Wen contributed equally to this work. References Associated Data Data Availability Statement No datasets were generated or analysed during the current study.