Efficacy of surgical intervention over conservative management in intertrochanteric fractures among nonagenarians and centenarians: a prospective cohort study

Abstract

Background: Optimal treatment strategy for nonagenarians and centenarians with hip fractures (NCHF) remained unknown. The authors aimed to compare the outcomes of surgical and conservative management in NCHF.

Methods: A prospective cohort study was conducted based on CPMHF database with NCHF patients hospitalized during 2014-2020. Comorbidities were evaluated by mECM score and restricted cubic spline was utilized to visually assess the dose-effect relationship between the mECM and outcomes. Propensity score matching was performed to balance baseline characteristics between nonsurgical and surgical groups. Multivariate logistic regression, Cox proportional hazard analysis, and survival analysis were employed for unfavorable outcomes (UFO) evaluation. Competing risk of death were analyzed based on Fine and Gray's hazard model and then constructed nomogram models for predicting survival rates. Subgroup analyses were used to determine potential population heterogeneity and sensitivity analyses were performed to test robustness of the results.

Results: The authors found increasing trends for UFO with the increase in the mECM score, and that high mECM score (HMS, ≥3) was independently associated with a 2.42-fold (95% CI: 2.07-3.54; P =0.024) increased risk of UFO, which remained significant after considering the competing role of death and were more pronounced in nonsurgical treatment, women, no insurance, and patients with spouse (all P for interaction <0.05). Surgical intervention was identified to be significant protective factors for UFO (RR, 0.59; 95% CI: 0.46-0.75; P <0.001) and severe complications (RR, 0.63; 95% CI: 0.41-0.96; P =0.033) after propensity score matching, as well as survival (HR, 0.40, 95% CI: 0.28-0.58; P <0.001), which remained significant after considering the competing role of death and in all sensitivity analyses and were more pronounced in HMS participants ( P for interaction=0.006). Subgroup analyses revealed surgical patients with HMS had a significantly higher UFO rate (excluding death, P <0.001) while nonsurgical patients with HMS had higher mortality rate as compared to the others ( P =0.005).

Conclusion: Surgical treatment for NCHF yields better outcomes compared to conservative treatment.

Figure 1

Flowchart for selection of study participants. mECM, modified Elixhauser’s Comorbidity Measure.

Figure 2

Dose-effect relationship between the mECM and UFO among total study participants. mECM, modified Elixhauser’s Comorbidity Measure. Evaluated by using a restricted cubic spline model with four knots and adjusted by all perioperative covariables with P-value <0.1, including treatment method, age group, insurance, marital status, hypoproteinemia, respiratory disease, liver disease, kidney disease, and severe anemia. P-nonlinearity showed no statistical significance, which was estimated using the likelihood ratio test comparing the restricted cubic spline model with the linear model. Relative risks were indicated by blue solid line and 95% CIs by blue dashed lines, in which the reference point was 3 for the mECM.

Figure 3

Kaplan–Meier survival curves of total study participants receiving surgical or nonsurgical interventions before (A) and after (B) PSM. PSM, propensity score matching.

Figure 4

Risk of PFO and death in relation to treatment group and mECM with considering the competing risk of death in total participants. mECM, modified Elixhauser’s Comorbidity Measure; PFO, poor functional outcomes. Univariate competing risk analyses by using the Fine and Gray’s hazard model outcomes in total patients showed the treatment methods was not statistically significant on PFO (A) while statistically significant on death (B). mECM was a statistically significant risk factor for PFO (C) while there was no statistical difference when it comes to death (D).

Figure 5

Nomogram for predicting the survival probability at 36-months (A), 48-months (B), 60-months (C), and 72 months (D) of total participants by using multivariate competing risk analyses. ASA, American Society of Anesthesiologists; mECM, modified Elixhauser’s Comorbidity Measure.