Abstract

Objective: To explore the postoperative complications and influencing factors of hip arthroplasty. Methods: We searched Chinese databases including CNKI, CBM, Wanfang, VIP, as well as English databases such as PubMed, Embase, Web of Science, and Cochrane Library. The search covered literature published in both Chinese and English up to March 1, 2023. The quality of included studies was assessed using the NOS Ottawa scale, and data analysis was conducted using Revman 5.3. Results: A total of 23 articles were included in this meta-analysis, with a combined sample size of 9099 cases. The results showed that age (RR=1.87, 95% CI 0.74, 3.00, P=0.001, P<0.05), female gender (RR=1.12, 95% CI 1.05, 1.12, P=0.001, P<0.05), surgery duration (RR=17.44, 95% CI 10.97, 23.92, P<0.00001, P<0.05), intraoperative bleeding (RR=61.30, 95% CI 16.56, 106.04, P=0.007, P<0.01) were significantly associated with postoperative complications of hip arthroplasty. BMI (RR=0.46, 95% CI -1.25, 2.18, P=0.59, P>0.05), hospital stay (RR=1.81, 95% CI -0.80, 4.42, P=0.17, P>0.05) showed no significant correlation with postoperative complications. Dislocation of prosthesis (RR=3.79, 95% CI 1.67, 8.64, P=0.001, P<0.05), prosthesis loosening (RR=3.98, 95% CI 2.76, 5.74, P<0.00001, P<0.05), periprosthetic infection (RR=2.52, 95% CI 1.09, 5.81, P=0.03, P<0.05), periprosthetic fracture (RR=5.86, 95% CI 1.85, 18.55, P=0.003, P<0.05), and postoperative deep vein thrombosis (RR=5.31, 95% CI 1.48, 19.12, P=0.01, P<0.05) were correlated with rehospitalization after hip arthroplasty. Conclusion: Female patients, older age, longer operation time, and intraoperative bleeding were related to postoperative complications in hip arthroplasty patients, while BMI and hospital stay had no obvious correlation of complications after hip arthroplasty.There was a correlation between prosthesis dislocation, prosthesis loosening, periprosthetic infection, periprosthetic fracture, and postoperative DVT and postoperative readmission. Medical staff should strengthen the care of patients with these related factors, avoid or reduce the occurrence of postoperative complications, and reduce the readmission rate of patients.

1. Introduction

Hip Replacement (HR) is a treatment method that involves crafting a prosthetic joint resembling the natural bone joint using biocompatible and mechanically sound materials. This prosthetic joint, comprising the femoral and acetabular components, is surgically implanted onto healthy bone using bone cement and screws. The procedure aims to replace the diseased joint, restoring normal function to the patient's hip joint. It is a mature and reliable therapeutic approach widely employed in clinical orthopedic surgeries, primarily addressing conditions such as femoral neck fractures, hip fractures, femoral head necrosis, developmental dysplasia of the hip, and notably, end-stage hip arthritis. However, if not managed properly, post-operative complications of HR may arise for various reasons, including prosthetic dislocation, loosening, periprosthetic infection, periprosthetic fractures, and the formation of Deep Vein Thrombosis (DVT). Research reports indicate an overall postoperative complication rate of 9.43%, with prosthetic dislocation occurring at a rate of 3% to 10% ^{1, 2}, making it a leading cause for patient rehospitalization after joint replacement surgery. Periprosthetic fractures have a reported incidence of 7.33% ³. Without prophylactic antithrombotic measures within the first 24 hours post-surgery, the occurrence of DVT after HR can reach as high as 42% to 57%, representing a significant factor in perioperative mortality ⁴. According to studies, the short-term rehospitalization rate after selective treatment ranges from approximately 2.9% to 12.6%, with 19% of readmissions being preventable ⁵.

Therefore, the management of complications and postoperative rehabilitation quality following hip replacement surgery will remain a focal point in clinical practice. This study aims to provide insights into preventing complications, alleviating patient suffering, enhancing the quality of care, and reducing rehospitalization rates through the exploration of postoperative complications and the analysis of influencing factors in HR patients.

2. Data and Methods

This study primarily selected relevant research literature on "hip joint replacement" published domestically and internationally.

Following the PICOS principle to establish inclusion criteria:

① P(Popuiations): Patients undergoing revision hip arthroplasty with complications or readmission after the procedure.

② I(Observational indicators): Age, gender, BMI (Body Mass Index), length of hospital stay, type of replacement, surgical time, reasons for replacement, surgical approach, comorbidities, etc.

③ C(Comparison): Patients without complications after the initial hip arthroplasty.

④ O(Outcome): Complications after hip arthroplasty, such as periprosthetic infection, deep vein thrombosis (DVT), prosthesis loosening, prosthesis dislocation, periprosthetic fractures, and postoperative readmission rates.

⑤ S(Study design): The included literature types are case-control, cohort studies, and cross-sectional studies, with language restricted to Chinese and English.

Exclusion Criteria:

① Animal experiments;

② Patients readmitted for reasons other than hip arthroplasty;

③ Inability to access full text, data inaccuracies, unclear or irrelevant outcome indicators, or low-quality literature;

④ Reviews, systematic reviews, meta-analyses, conference papers, etc.

Computer searches were conducted using Chinese databases, including CNKI(China national knowledge infrastructure), Wanfang Medical Network, VIP Chinese Journals, as well as English databases like PubMed, Embase, Web of Science, Cochrane Library. The search period ranged from the establishment of the database to January 01, 2023, covering publicly available Chinese and English case-control and cohort studies. References in the literature and freely available or unpublished grey literature were also searched. Chinese search terms included joint replacement, hip joint replacement, prosthesis joint replacement, joint prosthesis replacement, case-control, cohort study, complications, readmission, influencing factors, risk factors, factor analysis, etc. English search terms included Arthroplasty, Hip arthroplasty, Total Hip Arthroplasty, Hip replacement, Joint Arthroplasty, Hip and Knee Arthroplasty; complications, readmission, reoperation; factor, analysis, factor analysis, risk factors, etc. Literature searches were performed using the EndNote reference management software, organizing and classifying the retrieved literature, eliminating duplicate entries. Subsequently, based on titles and abstracts, analysis was conducted to exclude literature that did not meet inclusion criteria, including those unrelated to the "hip joint replacement" topic and those not qualifying as RCTs, case-control, or cohort studies. Full-text reading, in combination with inclusion and exclusion criteria, further refined the selection. Relevant literature was traced through references in obtained papers. Two researchers trained in evidence-based medicine jointly completed the literature screening process. In cases of disagreement, consensus was reached through discussion, or a third evaluator made the final decision.

Literature Search Process

The NOS Ottawa scoring scale was employed to assess the quality of included literature. The evaluation criteria primarily cover case selection, intergroup comparability, and the assessment of exposure or outcomes. A maximum score of 9 stars is considered full marks, with a score exceeding 6 indicating relatively good literature quality, making it eligible for inclusion in meta-analysis. Refer to Table 1 for details.

Develop standardized data extraction forms based on the content of the literature, including: first author, publication year, study country/region, study period, study type, sample size, follow-up time, as detailed in Table 2.

The information extracted from the literature was statistically analyzed using Excel sheets, and the Revman 5.3 software was employed for analysis. Mean differences (MD) and risk ratios (RR) were calculated as effect sizes based on the data type, along with the calculation of 95% confidence intervals (CI) for all effect sizes. A significance level of P<0.05 was used to indicate statistically significant differences. The I2 test was conducted to assess the heterogeneity of the statistical measures. When I2<50%, a fixed-effects model was employed; when I2>50%, a random-effects model was used. In cases of substantial heterogeneity, sensitivity analysis was performed, either by conducting a sensitivity analysis or by evaluating the source of heterogeneity through changes in effect sizes. Sensitivity analysis of the analysis results was carried out by reanalyzing the data after extracting each independent study.

3. Results

A total of 23 articles were included in this study, comprising 20 in Chinese and 3 in English. The Chinese articles originated from 20 regions, including 15 provinces, municipalities, and autonomous regions in China. Nineteen studies were case-control studies, and four were cohort studies. The follow-up duration ranged from 1 to 120 months. The major complications of hip joint replacement extracted included prosthesis dislocation, prosthesis loosening, periprosthetic infection, periprosthetic fracture, deep vein thrombosis, and readmission after surgery. The factors included in the literature can be categorized into general information such as age, gender, BMI index, and length of hospital stay; comorbidities, mainly diabetes, hypertension, osteoporosis, anemia, hyperlipidemia, and femoral head necrosis; surgical factors including surgical approach, surgical time, intraoperative bleeding, VAS score, and Harris score; and laboratory indicators such as C-reactive protein, hemoglobin, and D-dimer.

A total of 9,099 cases of hip joint replacement were included in the literature, with 7,278 cases of total hip replacement, 631 cases of hemiarthroplasty, and 1,190 cases where the type of hip joint replacement was not specified. General patient information revealed a higher prevalence of female patients (4,476/3,709) in the gender comparison, with an average age of 68.13±10.52 and BMI of 24.95±4.34 kg/m2. Refer to Table 3 for details.

This study included 11 articles in the literature, reporting a postoperative occurrence of prosthesis dislocation in patients (126/4228, 2.98%). Among them, in China, it was 80/3096 cases, 2.58%. Seven articles reported postoperative prosthesis loosening (182/2034, 8.94%), with 83/1498 cases in China, accounting for 5.54%. Eleven articles documented postoperative infections (114/3833, 2.97%), including 88/3297 cases in China, at a rate of 2.67%. Seven articles reported postoperative periprosthetic fractures (120/3096, 3.88%), with 99/2560 cases in China, at a rate of 3.87%. Eleven articles reported postoperative deep vein thrombosis (246/2360, 10.42%), with 216/2142 cases in China, at a rate of 10.08%. Four articles reported unplanned readmissions after surgery, with a 5.4% readmission rate at 30 days, 8.34% at 90 days, 3.71% at 180 days, and a 2.07% readmission rate at 2 years. See Tables 5 and 6 for details.

A total of 23 studies were included in the analysis of the relationship between age and complications after hip arthroplasty. The meta-analysis revealed heterogeneity (P＜0.00001, I2=92%). Using a random-effects model, age was identified as a factor influencing complications after hip arthroplasty (P＜0.0001), with an RR of 2.21 and a 95% CI of (1.15, 3.27). Refer to Figure 2 for details.

A total of 23 articles were included to analyze the relationship between gender and hip arthroplasty complications. The meta-analysis showed heterogeneity (P＜0.00001, I2=97%). Using a random-effects model, male patients were identified as a protective factor for hip arthroplasty (P=0.14), RR=0.86, 95% CI (0.71, 1.05). For female patients, a total of 14 articles were included to analyze the relationship between female patients and complications after hip arthroplasty. The meta-analysis showed no heterogeneity (P=0.08, I2=38%). Using a fixed-effects model, female gender was identified as a influencing factor for complications after hip arthroplasty (P=0.001), RR=1.12, 95% CI (1.05, 1.21).

A total of 9 articles were included to analyze the relationship between BMI and complications after hip arthroplasty. The meta-analysis showed heterogeneity (P<0.00001, I2=98%). Using a random-effects model, BMI was not identified as a influencing factor for complications after hip arthroplasty (P=0.59), RR=0.46, 95% CI (-1.25, 2.18).

A total of 7 articles were included to analyze the relationship between the length of hospital stay and complications after hip arthroplasty. The meta-analysis showed heterogeneity (P<0.00001, I2=99%). Using a random-effects model, the length of hospital stay was not identified as a influencing factor for complications after hip arthroplasty (P=0.17), RR=1.81, 95% CI (-0.80, 4.42).

A total of 13 articles were included to analyze the relationship between operation time and complications after hip arthroplasty. The meta-analysis showed heterogeneity (P<0.00001, I2=97%). Using a random-effects model, operation time was identified as an influencing factor for complications after hip arthroplasty (P<0.00001), RR=1.09, 95% CI (0.66, 1.52).

A total of 9 articles were included to analyze the relationship between intraoperative bleeding and complications after hip arthroplasty. The meta-analysis showed heterogeneity (P<0.00001, I2=99%). Using a random-effects model, intraoperative bleeding was identified as an influencing factor for complications after hip arthroplasty (P=0.007), RR=61.30, 95% CI (16.56, 106.04).

A total of 11 articles were included to analyze the incidence of postoperative prosthesis dislocation. The meta-analysis showed heterogeneity (P=0.005, I2=61%). Using a random-effects model, postoperative prosthesis dislocation was identified as a major factor for complications after hip arthroplasty (P=0.001), RR=3.79, 95% CI (1.67, 8.64).

A total of 7 articles were included to analyze postoperative prosthesis loosening. The meta-analysis showed heterogeneity (P=0.010, I2=43%). Using a fixed-effects model, postoperative prosthesis loosening was identified as a major factor for rehospitalization after hip arthroplasty (P＜0.00001), RR=3.98, 95% CI (2.76, 5.74).

A total of 11 articles were included to analyze the relationship between postoperative infection and complications after hip arthroplasty. The meta-analysis showed heterogeneity (P=0.0005, I2=68%). Using a random-effects model, postoperative infection was identified as an influencing factor for complications after hip arthroplasty (P=0.03), RR=2.52, 95% CI (1.09, 5.81).

A total of 7 articles were included to analyze the relationship between postoperative periprosthetic fracture. The meta-analysis showed heterogeneity (P=0.004, I2=69%). Using a random-effects model, postoperative periprosthetic fracture was identified as a major factor for complications after hip arthroplasty (P=0.003), RR=5.86, 95% CI (1.85, 18.55).

A total of 11 articles were included to analyze the relationship between postoperative DVT and complications after hip arthroplasty. The meta-analysis showed heterogeneity (P<0.00001, I2=83%). Using a random-effects model, postoperative DVT was identified as an influencing factor for complications after hip arthroplasty (P=0.01), RR=5.31, 95% CI (1.48, 19.12).

The meta-analysis indicates that female patients, older age, longer operation time, and increased intraoperative bleeding are influencing factors for postoperative complications in patients undergoing hip arthroplasty. Body mass index (BMI) and longer hospital stays are not considered factors leading to complications after hip arthroplasty. There is a correlation between postoperative prosthesis dislocation, prosthesis loosening, periprosthetic infection, periprosthetic fracture, and the occurrence of postoperative DVT with subsequent rehospitalization after hip arthroplasty.

Conduct sensitivity analysis on factors with significant heterogeneity. Initially, transform the fixed-effects model and random-effects model to assess the robustness of the results. The findings indicate noticeable differences in BMI and length of hospital stay before and after comparison, while no significant differences are observed in the results for the remaining factors. Subsequently, employ a method of systematically excluding individual studies to analyze the sources of heterogeneity, which does not evidently impact the results, suggesting relative reliability of the calculated effects. The summary of meta-analysis results and sensitivity analysis results is presented in Table 7.

A funnel plot was constructed to assess publication bias for factors included in studies exceeding 9 publications, taking patient age as an example. The results are shown in Figure 13. Asymmetric distribution of age was observed in the included literature, indicating a potential publication bias. Even after sensitivity analysis, which involved excluding relevant literature, publication bias persisted. This may be attributed to variations in sample size, confounding factors, different age criteria for inclusion across studies, or variations in study quality.

4. Discussion

The results of this study indicate that advanced age and female gender are influencing factors for complications after hip arthroplasty. With increasing age, various physiological functions gradually decline, bone density decreases, and skeletal fragility increases. Postmenopausal women, in particular, experience a reduction in estrogen secretion, leading to the occurrence of osteoporosis. The study results show no correlation between BMI and complications after hip resurfacing (HR) surgery. Although being overweight (BMI greater than 24 kg/m2) may affect the accuracy of implant positioning and exert greater force on the hip joint, posing a risk of postoperative complications such as loosening, dislocation, and infection, Duh et al.'s ²⁹ research suggests that complications are more likely to occur with a BMI greater than 35 kg/m2 or a weight exceeding 100 kg. Therefore, when assessing patients, special attention should be given to managing elderly, female, and overweight patients, with adjustments to treatment plans and drug dosages.

This study indicates that complications after hip arthroplasty are influenced by comorbidities such as diabetes, hypertension, hyperlipidemia, femoral neck fractures, and osteoporosis. The presence of multiple underlying diseases in elderly patients leads to impaired organ function, declining overall body function, reduced brain capacity, and decreased stress tolerance. This makes elderly patients more prone to fractures from falls and other injuries ³⁰. Therefore, during the perioperative period in elderly patients, adequate consideration should be given to the adverse effects of poorly controlled blood sugar, blood pressure, and lipid levels on surgical outcomes.

This study reveals a correlation between longer surgery duration, increased intraoperative blood loss, and the occurrence of complications after hip arthroplasty. Research suggests that prolonged surgery interrupts the body's autoregulation, leading to conditions such as hypercapnia, anemia, and hypothermia, reducing the body's ability to self-regulate and increasing the risk of postoperative complications ³¹. Surgery duration and intraoperative blood loss are also associated with the surgeon's proficiency, choice of surgical approach, and type of surgery. Total hip arthroplasty is superior to hip resurfacing in terms of surgery duration, intraoperative blood loss, hospital stay, and short-term efficacy, but total hip arthroplasty outperforms in postoperative Harris and VAS scores. Therefore, the selection of an appropriate surgical approach and type should be based on the patient's individual situation, optimizing the surgical and anesthesia process, minimizing patient anesthesia and surgery time, and controlling and reducing intraoperative bleeding as much as possible.

The literature included in this study indicates that common complications after hip arthroplasty include prosthesis dislocation, prosthesis loosening, lower limb deep vein thrombosis (DVT), periprosthetic fractures, and infection. The occurrence of various complications is a major reason for unplanned postoperative readmissions. Therefore, active preventive measures should be taken during the perioperative period to prevent the occurrence of postoperative complications.

This study shows a relatively high readmission rate within 90 days after HR surgery, which is attributed to the fact that it takes at least three months for clinical healing of damaged bones, tendons, and ligaments in the human body. During this period, inadequate formation of the original bone callus can easily lead to infection and dislocation after weight-bearing. There is limited research on postoperative readmission of patients after hip arthroplasty in China, while Western countries have relatively mature studies on this topic. This may be closely related to Western countries' healthcare policies, such as the inclusion of postoperative readmission rates in important quality assessment indicators for hospitals by the United States' Medicare and Medicaid Services, along with strict penalties.

5. Conclusion

Common complications after hip arthroplasty include prosthesis dislocation, periprosthetic infection, periprosthetic fractures, and postoperative DVT, all of which impact patient prognosis and postoperative quality of life. Advanced age, female gender, prolonged surgery duration, and increased intraoperative blood loss are influencing factors for complications after hip arthroplasty. Healthcare professionals should enhance care for patients with these potential risk factors to improve clinical outcomes, reduce the incidence of complications after hip arthroplasty, and decrease readmission rates.

Author Contribution

SNZ designed the study, YL professional guidance, JPZ did data analysis,DH and XTL data collection, data proofreading, SNZ wrote and revised the manuscript, and HML did data analysis and revised the manuscript. All authors contributed to and approved the final manuscript.

References