Correlation Between Serum Creatinine-to-Cystatin C Ratio and Prognosis of Patients with Hip Fracture

Introduction

Hip fractures represent a major public health concern in aging populations, with profound implications on morbidity, mortality, and global healthcare resource utilization.^1,2 Previous studies have reported a 1-year mortality rate of 10–30% following hip fracture surgery, underscoring its life-threatening nature.^3–5 As demographic shifts towards older age groups accelerate, projections indicate an increase in both the incidence of hip fractures and the number of all-cause deaths after hip fracture surgery,^6,7 further straining healthcare systems. Therefore, early identification of risk factors associated with mortality is crucial to enable risk stratification, guide targeted interventions, and improve prognosis.

Previous study has shown that postoperative neutrophil-to-lymphocyte ratio and aggregate inflammatory systemic index markers were highly associated with the surgical invasiveness during humeral shaft fracture osteosynthesis in a younger population.⁸ In addition, a retrospective cohort study reported the postoperative systemic immune-inflammation index was correlated with surgical trauma sustained by an older population with hip fracture.⁹ Therefore, the inflammatory markers in peripheral blood are associated with the prognosis of hip fractures.

Emerging evidence suggests that the creatinine-to-cystatin C ratio (CCR), a biomarker that integrates renal function and muscle metabolism, may offer prognostic insight beyond conventional renal indices.¹⁰ CCR correlates with skeletal muscle mass, systemic inflammation, and oxidative stress,^11–13 both of which are closely associated with the prognosis of patients with hip fractures. Observational studies have further associated low serum CCR with adverse outcomes in critical illnesses, cardiometabolic diseases, and malignancies.^14–16 Nevertheless, its utility in predicting the 1-year mortality after hip fracture surgery remains unclear.

Therefore, we conducted a retrospective cohort study to explore the association between serum CCR levels and prognosis of patients undergoing hip fracture surgery. This study provides a novel method of using CCR as a tool for the early identification of patients at elevated mortality risk following hip fracture surgery.

Materials and Methods

Study Design and Participants

In this retrospective cohort study, we recorded the data of patients with hip fracture including femoral neck and intertrochanteric fractures, at Changhai hospital from January 2018 to December 2023. This retrospective study was approved by the ethics committee of Changhai hospital (CHEC2024-257). The ethics committee waived the requirement for informed consent due to the retrospective collection of de-identified clinical data from patients. This study was conducted in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology guidelines and conformed to the principles of the Declaration of Helsinki.

To screen participants for this study, patients undergoing hip surgery who met the following criteria were included: patients with a diagnosis of femoral neck fractures and intertrochanteric fractures by X-ray examination at admission; aged≥18 years. Patients with renal function impairment according to an estimated glomerular filtration rate < 60 mL / (min × 1.73 m²) using the Chinese-based equation,¹⁷ with multiple fractures, dementia, and missing data on creatinine or cystatin C were excluded.

Variables

Based on existing literature and clinical practice, we collected the following variables. The demographic characteristics included age, gender, American Society of Anesthesiologists (ASA) physical status, body mass index (BMI), alcohol history, and smoking history. The medical history evaluation included hypertension, diabetes, cardiovascular disease (coronary heart disease, arrhythmia, and chronic heart failure), cerebrovascular disease (cerebral infarction and cerebral hemorrhage), lung disease (chronic bronchitis, bronchial asthma, and chronic obstructive pulmonary disease), and cancer.

We also calculated Charlson comorbidity index to perform overall assessment of comorbidities.¹⁸ Laboratory results included serum creatinine, cystatin C, albumin, urea nitrogen, glucose, D-dimer, activated partial thromboplastin time (APTT), hemoglobin, lymphocyte count, monocyte count, neutrophil count, leucocyte count, and platelet count. We also recorded the type of anesthesia, surgical procedure, time to admission (time from injury to admission) and time to surgery (time from admission to surgery). Time to admission and time to surgery were derived from medical records.

The most recent preoperative laboratory results were used in this study. The CCR was calculated as creatinine (mg/L) / cystatin C (mg/dL). We used multiple imputations to deal with missing data in continuous variables whose missing values were less than 10%.

Outcomes

The primary endpoint was all-cause mortality within one year of hip fracture surgery. The secondary endpoint was 6-month mortality after surgery. We collected the data on mortality by telephone follow-up.

Statistical Analysis

Participants were divided into two groups based on their CCR values. Continuous variables with normally distributed data were described as mean ± SD and compared using Student’s t-test. The Mann–Whitney U-test was used to compare continuous variables with non-normally distributed data, described as medians (IQR). Categorical variables are presented as frequencies (%) and analyzed using the chi-square test.

CCR levels were categorized as low (<7.3) or high (≥7.3), based on the median. The association between CCR and 6-month, 1-year mortality was constructed using multivariate Cox proportional hazard models when CCR levels were both continuous and binary variables. Variates with a p-value of <0.1 in the comparison of baseline characteristics between the high and low CCR groups, were selected for model adjustment. In Model I, no covariates were adjusted. Model II was adjusted for age, gender, BMI, ASA physical status, time to admission, alcohol history, smoking history, hypertension, diabetes, cardiovascular disease, and lung disease. In model III, covariates including albumin, glucose, D-dimer, APTT, hemoglobin, neutrophil count, and leukocyte count were further adjusted.

We used a smoothed curve fit to assess the association between the CCR and 1-year mortality. Kaplan-Meier curves were generated to compare the probability of survival for the different CCR groups. Stratified and interaction analyses were performed according to age (< 80 or ≥ 80 years), gender (male or female), and ASA physical status (< III or ≥ III), hypertension (yes or no), diabetes (yes or no), and cardiovascular disease (yes or no). Each stratification was adjusted for age, gender, BMI, ASA physical status, time to admission, alcohol history, smoking history, hypertension, diabetes, cardiovascular disease, lung disease, albumin, glucose, D-dimer, APTT, hemoglobin, neutrophil count, and leukocyte count, with exception of the stratification factor itself.

We also performed the propensity score matching and multivariate Cox analyses as sensitivity analyses after excluding patients with younger than 60 years and pathologic fractures due to specific mechanism of hip fracture. This multivariate Cox analysis adjusted for age, gender, BMI, ASA physical status, time to admission, alcohol history, smoking history, hypertension, diabetes, cardiovascular disease, lung disease, albumin, glucose, d-dimer, APTT, hemoglobin, neutrophil count, and leucocyte count. Which were also included in multivariable logistic regression model to conduct propensity score matching analysis. A 1:1 nearest-neighbor matching algorithm with a caliper width of 0.2 was applied and a standardized mean difference (SMD) <0.1 was considered indicative of variable balance between the groups before and after matching.

Free Statistics software version 1.7.1 (Beijing FreeClinical Medical Technology Co., Ltd, Beijing, China), and R 4.2.2 software (http://www.Rproject.org; The R Foundation, Vienna, Austria) were used to conduct all statistical analyses. Statistical significance was set at P < 0.05.

Results

A total of 2,659 patients with hip fracture were identified. After excluding 745 patients with multiple fracture, renal function impairment, dementia, and missing values on creatinine or cystatin C, 1,914 patients with hip fracture were included in the final data analysis as shown in the flow chart (Figure 1).

Figure 1 Flowchart of study population.

Baseline Characteristics

The baseline information of all the participants is presented in Table 1. The mean age of the patients was 73.4 ± 14.0 years, and females were prevalent (61.9%). The median baseline CCR was 7.3 (6.3–8.5). The 1-year mortality rate of hip fractures was 7.3%. Participants were divided into two groups based on their CCR values. Patients with low CCR had significantly higher 6-month mortality and 1-year mortality (all P < 0.05) than those with high CCR. Comparisons of baseline information between the two groups are shown in Table 1.

Table 1 Demographics and Clinical Characteristics Stratified by CCR

Association of CCR with Prognosis in Hip Fracture Patients

The linear correlation between the CCR and 1-year mortality after adjusting for confounding variables is shown in Figure 2. Kaplan-Meier curves showed that patients in the high CCR group had a significantly higher 1-year survival rate. A similar result was observed for the 6-month survival curve (Figure 3).

Figure 2 Curve fitting of CCR and 1-year mortality in patients with Hip fracture surgery. Data were fit by a Cox proportional hazard regression model based on restricted cubic splines. P-value for non-linearity derived from Likelihood Ratio Test. CCR was entered as continuous variable. Data were adjusted for the factors (age, gender, BMI, ASA physical status, time to admission, alcohol history, smoking history, hypertension, diabetes, cardiovascular disease, lung disease, albumin, glucose, d-dimer, APTT, hemoglobin, neutrophil count, and leucocyte count). The curves line and shaded areas around depict the estimated values and their corresponding 95% confidence intervals. Curves below depict the distribution of the CCR. Abbreviations: CCR, creatinine-to-cystatin C ratio; BMI, body mass index; ASA, American Society of Anesthesiologists; APTT, activated partial thromboplastin time.

Figure 3 Kaplan–Meier curve of mortality for patients with Hip fracture surgery. (A) Kaplan–Meier curve of 6-month mortality after hip fracture surgery, (B) Kaplan–Meier curve of 1-year mortality after hip fracture surgery. P-value derived from Log Rank Test. Abbreviation: CCR, creatinine-to-cystatin C ratio.

Multivariate Cox regression models indicated that a high CCR was independently related to reduced 6-month mortality regarding continuous variable (hazard ratio (HR) = 0.82, 95% confidence interval (CI) = 0.71–0.95) or categorical variable (HR = 0.51, 95% CI = 0.31–0.84) in the adjusted model. Moreover, a high CCR was independently related to reduced 1-year mortality with respect to the continuous variable (HR = 0.88, 95% CI = 0.79–0.99) or categorical variable (HR = 0.61, 95% CI = 0.42–0.90) in the adjusted model. The associations between the unadjusted and adjusted models were robust in all three models (all P < 0.05), as shown in Table 2.

Table 2 Univariate and Multivariate Cox Regression Analysis to Assess the Association Between CCR and Prognosis of the Patients with Hip Fracture Surgery

Additionally, we also performed sensitivity analyses to provide additional robustness to our findings after excluding patients with younger than 60 years and pathologic fractures. The multivariate Cox regression analysis showed that high CCR was associated with low 1-year mortality (HR = 0.58, 95% CI = 0.40–0.85) (Supplementary Table 2). After matching, all covariates were well-balanced (SMD <0.1) (Supplementary Table 1). Moreover, the propensity score matching also indicated that increased CCR was independently related to decreased 1-year mortality (HR = 0.53, 95% CI = 0.35–0.80) (Supplementary Table 2).

Subgroup Analysis

Subgroup analyses showed that a high CCR was independently associated with reduced 1-year mortality after hip surgery in patients with diabetes (HR = 0.39; 95% CI = 0.17–0.89), males (HR = 0.48; 95% CI = 0.26–0.86) and those without hypertension (HR = 0.55; 95% CI = 0.32–0.97) or cardiovascular disease (HR = 0.54; 95% CI = 0.34–0.84) (Figure 4).

Figure 4 Association between CCR and 1-year mortality in subgroups for patients with Hip fracture surgery. Low CCR was the reference for high CCR. Each stratification adjusted for the factors (age, gender, BMI, ASA physical status, time to admission, alcohol history, smoking history, hypertension, diabetes, cardiovascular disease, lung disease, albumin, glucose, d-dimer, APTT, hemoglobin, neutrophil count, and leucocyte count) in the multivariable Cox regression, except for the stratification factor itself. Abbreviations: CCR, creatinine-to-cystatin C ratio; BMI, body mass index; ASA, American Society of Anesthesiologists; APTT, activated partial thromboplastin time; HR, hazard ratio; CI, confidence interval.

Discussion

This study included 1,914 patients with hip fracture, and the 1-year mortality rate was 7.3%. We found a negative linear relationship between the CCR and 1-year mortality in patients with hip fracture. Meanwhile, we revealed that a high CCR is associated with reduced 6-month and 1-year mortality in patients with hip fractures after adjusting for covariates. In addition, subgroup analysis showed that a high CCR was independently associated with reduced 1-year mortality in patients with diabetes, males, and those without hypertension or cardiovascular disease. Moreover, the sensitivity analyses provided additional robustness to our findings.

Recent studies have shown that 1-year mortality following hip fractures can vary widely across different populations, with estimates in some cohorts ranging from 10% to 30%,^3,19–21 higher than the 7.3% observed in our study. This discrepancy may be due to our inclusion criteria (patients aged ≥18 years) and the short median time from injury to hospital admission (only 19 hours). Previous studies have identified factors such as advanced age, comorbidities, and prolonged time from injury to admission as significant risk factors for increased mortality, which is consistent with the results of our study.^22–24 In addition, sarcopenia has also been associated with higher mortality rates after hip fracture surgery.²⁵ Moreover, our study found that CCR, as a biomarker of skeletal muscle mass and strength, was related to 6-month and 1-year mortality following hip fracture surgery.

Studies have shown that low CCR is associated with poor renal function assessment and may reflect systemic immune-inflammation status,^26–28 which can influence patient outcomes. Additionally, CCR has been increasingly recognized for its potential prognostic value in various clinical contexts including chronic kidney, cardiovascular, and chronic obstructive pulmonary diseases.^29–31 Furthermore, an elevated CCR has been linked to improved outcomes after major surgery.³² In line with previous studies, we found that high CCR levels were associated with low mortality rates after hip fracture surgery.

The protective effect of higher CCR against mortality may stem from its dual representation of renal health and muscle integrity. Cystatin C, a sensitive marker of glomerular filtration, reflects renal clearance capacity,³³ while creatinine level indirectly indicates muscle mass.³⁴ Therefore, a higher CCR indicates both adequate renal function and preserved muscle mass, which are critical factors for recovery after hip fracture surgery.

However, the definite causes and pathophysiological mechanisms underlying the relationship between CCR and a poor prognosis remain unclear. Patients with low CCR may experience decreased renal function reserve,³⁵ reduced muscle mass,³⁶ and impaired muscle anabolism,³⁷ resulting in decreased postoperative mobility and an increased risk of complications, such as deep vein thrombosis and pneumonia. This exacerbates frailty and susceptibility to adverse outcomes.

This study has several clinical implications and strengths. First, we analyzed data from a large cohort of 1,914 patients with hip fracture to provide an accurate and reliable association between CCR and the prognosis of patients after hip fracture surgery. Second, we performed strict statistical adjustments to investigate the association between CCR and prognosis in patients to minimize potential confounders. And we conducted sensitivity analyses to verify the robustness of our results. Moreover, the CCR is simple, easy to calculate, convenient for clinical use, and is not affected by individual subjectivity. To our knowledge, this is the first study to explore the relationship between CCR and mortality in patients undergoing hip fracture surgery.

However, our study has some limitations. First, the nature of the retrospective study limits its ability to establish causation. Second, this study had limited generalizability to diverse populations because it included only patients with hip fracture. Additionally, although strict statistical adjustments were made to minimize the effects of confounding factors, some potential confounding factors were not determined. Lastly, the mechanisms behind the relationship of CCR with mortality were unclear. Therefore, prospective studies with larger cohorts are needed to validate our findings and to explore the potential mechanisms underlying the relationship between CCR and mortality.

Conclusion

Our study showed a negative linear association between CCR and 1-year mortality in patients who underwent hip fracture surgery. Moreover, our study highlights the prognostic significance of preoperative CCR in patients with hip fracture, with low CCR independently predicting increased 6-month and 1-year mortality.