INTRODUCTION
Hip arthroplasty is a well-established procedure that restores mobility and quality of life for patients with degenerative or traumatic hip conditions. However, postoperative outcomes vary substantially across age groups. Older adults, especially those classified as very old (80 years old or greater),1) often face health problems that may impact surgical outcomes, including increased comorbidities, higher perioperative risk, and prolonged recovery times.2) In contrast, younger patients tend to exhibit better functional recovery but may face higher revision rates due to longer life expectancy and increased activity levels, which place greater mechanical stress on the prosthesis.3) Understanding these age-related differences is crucial for optimizing treatment strategies and improving overall patient outcomes.
The impact of advanced age on postoperative outcomes following hip arthroplasty has been extensively studied in developed countries. Several reports indicate that older patients have a heightened risk of complications such as venous thromboembolism, periprosthetic fractures, infections, and cardiovascular events.4-7) Moreover, prolonged hospital stays, increased dependency on rehabilitation services, and lower functional recovery rates have been documented among this population.7,8) It is still unclear whether chronological age itself independently increases postoperative risk, or if this association is predominantly mediated by age-related comorbidities.9) This uncertainty is particularly pronounced in developing nations, where challenges such as limited access to specialized care and economic barriers can lead to more advanced disease at presentation, which in turn may further influence surgical outcomes.6,10)
Globally, population aging is driving a growing demand for hip arthroplasty, a trend that is increasingly evident in low- and middle-income countries. Vietnam is undergoing a rapid demographic transition, with a steadily expanding older population accompanied by a rising burden of hip fractures and degenerative hip disease.11) Consequently, both fracture-related and elective hip arthroplasty procedures are becoming more common within the Vietnamese healthcare system. In this developing-country context, patients often present with substantial comorbidity burden, reduced physiological reserve, and uneven access to perioperative optimization and postoperative rehabilitation.10) These factors may alter the relationship between age, comorbidities, and surgical outcomes, thereby limiting the direct applicability of evidence derived from high-income settings.
Therefore, this study was designed to investigate the independent and synergistic effects of age and comorbidities on surgical outcomes following hip arthroplasty within a developing country context. To achieve this overarching goal, our investigation was structured around three key objectives: (1) compare comorbidity profiles across different age groups; (2) evaluate the independent association between age and major postoperative complications; and (3) explore factors that mediate the relationship between age and adverse outcomes.
MATERIALS AND METHODS
Study Design
This cohort analysis was performed at the University Medical Center Ho Chi Minh City (UMC-HCMC) focusing on patients undergoing hip arthroplasty. We performed a comprehensive review of electronic medical records for all patients who underwent hip arthroplasty between January 2021 and December 2024. The reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.12)
Ethics
The study protocol was approved by the Institutional Review Board of University Medical Center Ho Chi Minh City (Approval No. 07/GCN-HĐĐĐ, dated January 24th, 2025). The requirement for individual informed consent was waived due to the retrospective design. To ensure patient confidentiality, all patient data were anonymized, and confidentiality was strictly maintained. The study was conducted in accordance with the principles of the Declaration of Helsinki.
Patients
Patients aged 18 years or older who underwent primary or revision hip arthroplasty were included. We excluded those with American Society of Anesthesiologists (ASA) physical status class IV or higher, emergency or multi-trauma operations, bilateral procedures, severe hip deformity or instability, Crowe type IV hip dysplasia, prior spinal surgery, and cases with incomplete or missing medical data. Patients were categorized into younger adults (18–64 years), older adults (65–79 years), and oldest old (≥80 years). These cut-offs follow commonly used geriatric and clinical health-system thresholds: 65 years is widely used to define the “older adults” population. In comparison, ≥80 years corresponds to the oldest old recognized in commonly adopted outcomes research. These thresholds distinguish younger adults, older adults (elderly), and the oldest old (octogenarians), the latter being recognized as a population with substantially reduced physiological reserve and increased perioperative vulnerability.13)
Protocol of Care
All patients received a standardized preoperative evaluation followed by coordinated, multidisciplinary perioperative management. Perioperative management adhered to institution-specific multimodal strategies for analgesia, thromboprophylaxis, and infection prevention. The preoperative assessment consisted of a comprehensive clinical examination, laboratory investigations, and anesthetic risk evaluation. Intraoperative management, conducted by experienced anesthesiologists and orthopedic surgeons, adhered to standardized anesthetic techniques with continuous hemodynamic and fluid balance monitoring. Tranexamic acid was administered selectively to minimize surgical blood loss when deemed appropriate. Postoperatively, patients were managed with a multimodal pain control regimen that combined non-opioid agents, opioids, and regional anesthesia techniques. Venous thromboembolism prophylaxis—through anticoagulants or antiplatelet therapy—was tailored to individual risk factors. Nearly all patients received prophylactic antibiotics, with treatment provided when necessary. Early mobilization and structured rehabilitation were encouraged as soon as clinically feasible to accelerate recovery.
Outcomes
All postoperative complications were identified through clinical records, laboratory data, and imaging results documented during hospitalization. Events were validated independently by two investigators to ensure consistency and accuracy. When multiple complications occurred in a single patient, only one instance was counted toward the composite outcome to prevent duplication.
The primary outcome was defined as a composite of major postoperative complications. In a resource-limited, single-center setting, utilizing a composite outcome provides a more comprehensive assessment of the perioperative physiological challenge. This approach also increases statistical power, allowing for a more robust identification of shared risk factors that might be underrepresented if individual, lower-frequency complications were analyzed in isolation.
Postoperative complications were categorized by organ system (e.g., cardiovascular, respiratory, infectious, renal, gastrointestinal, neurologic) using standard diagnostic criteria. Infectious complications comprised urinary tract infections, wound infections, systemic sepsis, and septic shock. Gastrointestinal complications included ileus, acute bowel obstruction, gastrointestinal bleeding, abdominal compartment syndrome, hepatic dysfunction, pancreatitis, and other acute digestive disorders. Cardiovascular complications encompassed deep venous thrombosis, pulmonary embolism, myocardial ischemia or infarction, arrhythmia, cardiac arrest, cardiogenic shock, cerebrovascular accident or stroke, and acute heart failure. Respiratory complications involved pneumonia, respiratory failure, and the need for postoperative mechanical ventilation. Neurological complications included delirium, postoperative cognitive dysfunction, and psychosis, while renal complications referred primarily to acute kidney injury.
Statistical Analysis
Data were analyzed using R (version 4.3.2; R Foundation for Statistical Computing, Vienna, Austria), with descriptive statistics for both categorical and continuous variables, and appropriate group-comparison tests. Categorical variables were summarized as frequencies and percentages, while continuous variables were reported as the mean with standard deviation or as the median with interquartile range, depending on the data distribution. Group comparisons used the chi-square or Fisher exact test for categorical data, one-way ANOVA for normally distributed continuous data, and the Kruskal–Wallis test for non-normally distributed continuous data.
Independent predictors of postoperative outcomes were identified using a multivariable logistic regression model. To isolate these predictors, the model was adjusted for a comprehensive set of potential confounders, including key demographic variables, preoperative clinical assessments, surgical-related assessments, and a panel of clinically significant comorbidities. Adjusted associations were presented as odds ratios (ORs) with corresponding 95% confidence intervals (CIs).
To further explore the potential causal mechanisms underlying the observed associations, a causal mediation analysis was performed within a structural equation modeling (SEM) framework using a counterfactual-based approach. Separate mediation models were specified to evaluate whether heart failure and preoperative anemia mediated the association between age (modeled as a continuous variable) and postoperative complications. The proportion of the total effect mediated by each intermediate variable was calculated. Statistical inference was based on simulation-based confidence intervals with 1,000 Monte Carlo draws and robust standard errors. All analyses were two-sided, and a p-value <0.05 was considered statistically significant.
Sensitivity and Subgroup Analyses
To address potential confounding by surgical indication, prespecified sensitivity analyses were conducted by stratifying patients into two groups: fracture-related arthroplasty and non-fracture-related arthroplasty. Within each stratum, both multivariable logistic regression models and SEM analyses were refitted using the same covariate structure and model specifications as in the primary analysis. The consistency of effect estimates and mediating pathways across strata was evaluated. Effect modification by surgical indication was formally assessed in regression analyses by including interaction terms. All statistical tests were two-sided, and a p-value <0.05 was considered statistically significant.
RESULTS
Demographics
A total of 769 patients underwent hip arthroplasty during the study period. Patients were classified into three groups: younger adults (n=363, 47.2%), older adults (n=241, 31.3%), and oldest old (n=165, 21.5%) (Table 1, Fig. 1). The median age progressed from 54 years in the younger adults to 72 years in the older adults and 86 years in the oldest old, accompanied by a shift from a male predominance (65.6%) in the youngest group to a clear female majority in the older cohorts (p<0.001).
Surgical indications varied markedly by age: avascular necrosis was the primary diagnosis in the younger adults (76.0%), whereas fracture-related arthroplasty predominated in the older adults and oldest old. The proportion of patients with an ASA class III status increased sharply from 14.3% in the younger adults to 46.1% in the oldest old, accompanied by a corresponding rise in the median Nutritional Risk Screening score (p<0.001).
Perioperative Interventions
Total hip arthroplasty was more frequently performed in the younger adults (90.9%) compared to the older cohorts (p<0.001). Primary hip arthroplasty rates were comparable across groups (p=0.207). While intraoperative vasopressor use increased with age, median surgical time and surgical blood loss decreased progressively in the older groups (p<0.001). Postoperatively, NSAIDs and morphine use was significantly lower in the older adults and oldest old, whereas prescriptions for antiplatelet agents and therapeutic antibiotics were more common (all p<0.01) (Table 2).
Aging and Comorbidities
The burden of comorbidities increased substantially with advancing age group, a trend reflected in both composite scores and individual diagnoses (Fig. 2). The median Charlson Comorbidity Index score, a global measure of comorbidity, rose sharply from 1.0 in the younger adults to 4.0 in the older adults and 5.0 in the oldest old (p<0.001).
The prevalence of numerous specific conditions drove this pattern. Preoperative anemia demonstrated a steep increase across the three respective age groups (24.2%, 48.5%, and 67.3%), as did hypertension, chronic kidney disease, cardiovascular disease, and dementia, underscoring the systemic physiological decline associated with aging.
Aging and Outcomes
The composite complication rate was significantly higher in the oldest old (60.0%) compared to the older adults (36.9%) and younger adults (18.7%) (p<0.001) (Fig. 3). Older patients also experienced significantly higher rates of specific complications, including urinary tract infections, sepsis, septic shock, and acute kidney injury. Sepsis and septic shock rates were significantly higher in the oldest old (19.4% and 8.5%, respectively; p<0.001) (Fig. 3).
Aging, Comorbidities and Outcomes
Although age group was strongly associated with complications in unadjusted analyses, it was no longer an independent predictor after multivariable adjustment. Instead, pre-existing heart failure (aOR=5.49; 95% CI 2.19–13.74; p<0.001) and preoperative anemia (aOR=1.77; 95% CI 1.21–2.59; p=0.003) were identified as the independent predictors of postoperative complications (Fig. 4).
Our findings identify pre-existing heart failure as a powerful, age-independent determinant of adverse outcomes. The high odds ratio (aOR=5.10) suggests that limited cardiovascular reserve is a far more decisive factor in surgical recovery than chronological age. Clinically, this underscores the necessity for targeted perioperative strategies for these high-risk patients to mitigate the risk of acute decompensation.
In fracture-related arthroplasty, both anemia and heart failure remained strong independent predictors of complications across all age groups (Table 3). In non-fracture-related arthroplasty, anemia lost its independent association after adjustment, whereas heart failure remained consistently associated with adverse outcomes.
In the overall population, advanced age was not associated with heart failure, highlighting that heart failure acted as a powerful age-independent determinant of complications (Fig. 5A). Mediation analysis revealed that advanced age was associated with preoperative anemia (β=0.023, p=0.006), and anemia was associated with postoperative complications (β=0.548, p=0.005), with a mediated proportion of 32.3% in the overall population (Fig. 5B). In fracture-related arthroplasty, anemia was associated with postoperative complications whereas advanced age was not associated with anemia (Fig. 5B1). In non-fracture-related arthroplasty, advanced age was associated with anemia, but anemia was not associated with postoperative complications (Fig. 5B2).
DISCUSSION
This study examined whether chronological age independently predicts postoperative complications following hip arthroplasty in a resource-limited setting. Although complication rates increased markedly with age, reaching 60.0% in the oldest old, our findings demonstrate that age itself is not an independent risk factor once comorbidities are considered. Instead, the excess surgical risk in older patients is primarily driven by the higher burden of age-related conditions, particularly pre-existing heart failure and preoperative anemia. While our multivariable analysis demonstrated that chronological age was not an independent predictor of postoperative complications, it remains a highly relevant clinical factor. In this study, advancing age was inextricably linked to a substantial increase in systemic physiological decline and comorbidity burden. Therefore, while age itself may not drive the risk, it serves as a critical surrogate for the cumulative multi-organ vulnerability often encountered in the old population.9)
While the association between advanced age and increased complication rates aligns with existing literature, the overall magnitude of complications in our cohort was more pronounced than typically reported in developed countries.14) This disparity likely reflects differences in healthcare infrastructure and perioperative management. In developed nations, the widespread implementation of enhanced recovery after surgery (ERAS) protocols, specialized geriatric co-management, and advanced prehabilitation have proven effective in mitigating age-related surgical risks.15) Studies demonstrated that with optimized perioperative care, older adults (elderly patients) can achieve outcomes comparable to those of younger patients.15,16) In contrast, patients in developing countries often present with more advanced disease and face uneven access to such multidisciplinary care, exacerbating their baseline physiological vulnerability. These findings emphasize the importance of comprehensive preoperative assessment and optimization of comorbidities to mitigate the risk of complications in older patients undergoing hip arthroplasty.
Our mediation analysis provides further insight, demonstrating that preoperative anemia mediates nearly one-third (32.3%) of the effect of age on postoperative complications. This finding is consistent with prior studies demonstrating that anemia is highly prevalent in older surgical patients and independently associated with higher morbidity and mortality after total hip arthroplasty.17,18) The divergent findings observed between surgical indications provide further context: in fracture-related procedures, anemia reflects acute physiological stress predisposing patients to complications irrespective of age, whereas in elective arthroplasty it acts an age-associated condition that can be mitigated by patient selection. Clinically, this highlights that the excess risk in the oldest old is largely attributable to modifiable factors, underscoring preoperative evaluation and optimization of hemoglobin levels as a mandatory component of perioperative management.17,18) Conversely, heart failure acted as a powerful, age-independent determinant of adverse outcomes. Patients with pre-existing heart failure experienced markedly higher complication rates across all age groups and surgical indications, identifying this condition as a critical marker of limited cardiovascular reserve rather than a mere consequence of chronological aging.9) Clinically, these results underscore the importance of meticulous preoperative cardiovascular assessment and optimization. Strategies such as volume status optimization, careful hemodynamic monitoring, and early postoperative surveillance may be particularly critical to minimize morbidity.15)
This study has several limitations. The retrospective, single-center design may introduce selection and information biases, and despite rigorous multivariable adjustments, residual unmeasured confounding remains possible. Furthermore, reliance on electronic medical records restricted our ability to assess long-term functional recovery, and findings from a single tertiary hospital may lack generalizability to differently resourced healthcare settings. Future prospective, multicenter studies with extended follow-up are required to validate these findings and formally evaluate the impact of targeted comorbidity optimization on long-term surgical outcomes in older adults.
In conclusion, advanced age was associated with a higher prevalence of comorbidities and increased rates of postoperative complications in patients undergoing hip arthroplasty. However, age itself was not identified as an independent predictor of adverse outcomes. The increased risk is driven mainly by comorbidities, particularly pre-existing heart failure and preoperative anemia. Anemia partially mediated the effect of age in the overall cohort, with context-dependent effects across surgical indications. These findings support a shift from age-based risk stratification to the targeted identification and optimization of modifiable comorbidities, aiming to improve outcomes in older patients.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
