Do non-simultaneous bilateral proximal femoral fractures in the elderly have a similar pattern?

Abstract

BACKGROUND

Proximal femoral fractures (PFFs) in the elderly carry a substantial risk of a second contralateral fracture non-simultaneous bilateral (NSB) PFF.

AIM

To describe NSB-PFFs in terms of incidence, fracture pattern symmetry, timing between fractures, and associated risk factors.

METHODS

A single-center retrospective analysis was conducted on 1455 patients surgically treated for a PFF between June 2021 and June 2024. Among them, 136 patients with a documented non-simultaneous contralateral PFF were identified. Demographics, fracture type, treatment, and time interval between fractures were collected and analysed.

RESULTS

The incidence of NSB-PFF in the entire group was 10.5%. The cohort was predominantly female (79.4%), with a median age of 86.1 years. A strong correlation was found between the pattern of the first and second fracture (P < 0.0001). The median interval between fractures was 4.1 years, with most events occurring within 2-5 years. Multivariate analysis confirmed that the type of the first fracture was an independent predictor of the contralateral fracture pattern (odds ratio = 6.71; 95%CI: 3.11-14.48; P < 0.001), while age and sex were not significantly associated.

CONCLUSION

Patients sustaining a first PFF represent a high-risk group for a contralateral fracture, which often mirrors the initial fracture pattern and occurs within a vulnerability window.

Key Words: Contralateral hip fracture; Fracture pattern symmetry; Secondary fracture prevention; Osteoporosis management; Hip protectors

Core Tip: This retrospective study shows that approximately one in ten elderly patients sustaining a proximal femoral fracture (PFF) will later experience a contralateral fracture. The second fracture frequently reproduces the same fracture pattern as the first, and multivariate analysis confirms that the initial fracture type is an independent predictor of the contralateral fracture pattern. Most second fractures occur within a 2-5 years interval after the first event. These findings suggest that patients with a first PFF represent a clearly identifiable high-risk population in whom targeted secondary prevention strategies should be prioritized.

INTRODUCTION

Proximal femoral fracture (PFF) is one of the most common fractures in elderly individuals[1], whose incidence can reach more than 300 cases per 100000 women and more than 150 per 100000 men annually in some countries, as reported in a systematic review of 72 studies from 63 different nations[2].

Epidemiological data vary between countries, and although age-standardized incidence is gradually decreasing in many regions, this trend is far outweighed by population ageing. Consequently, the global number of hip fractures is expected to rise, from 1.26 million events recorded in 1990 to an estimated 4.5 million in 2050 in Europe[3,4]. A recent large cohort study reported a 74.1% increase in PFFs over 10 years within the same geographical and referral setting[5].

Beyond the increased risk of mortality in elderly patients[6], hip fractures have a substantial-often devastating-impact on quality of life, with a high risk of reduced mobility, functional decline, loss of independence, and inability to return to pre-injury living conditions[7-9].

The socio-economic impact of PFFs on public health is also considerable. The incidence and direct costs of PFFs in the elderly have been reported to be comparable to those of acute myocardial infarction[10]. Although the direct treatment costs of PFFs are similar to those of other frequent causes of hospitalization, such as myocardial infarction, the indirect social costs due to new comorbidities, sarcopenia, disability, and mortality are probably even higher[4].

An additional emerging issue is the tendency of elderly patients to sustain a second PFF on the contralateral side[11]. The incidence of a non-simultaneous bilateral (NSB) PFF has remained stable over the last three decades and ranges from 4% to 15%[12-14].

A limited number of studies have investigated the epidemiological characteristics and clinical outcomes of NSB-PFF, including cumulative incidence, age and sex differences, bone mineral density symmetry, interval between fractures, timing of occurrence, recovery of mobility, and mortality[11]. The present study focuses on NSB-PFFs in a clearly defined urban district over 3 years of observation. This study aims to evaluate the incidence, timing relative to the first fracture, the presence of a common fracture pattern in NSB-PFFs.

MATERIALS AND METHODS

A single-center retrospective analysis was performed on all patients surgically treated for PFFs over a 3-year period (June 2021-June 2024) in a trauma unit. A cohort of patients diagnosed with PFFs [31A1-3 and 31B1-3 according to the Arbeitsgemeinschaft für Osteosynthesefragen/Orthopaedic Trauma Association (AO/OTA) classification] in whom a previous contralateral fracture of the proximal femur had been documented. In consideration of the retrospective analysis, the Local Research Ethics Committee has confirmed that no ethical approval is required. All patients signed informed consent to be enrolled in the study. The exclusion criterion was simultaneous bilateral fractures. Patients who underwent contralateral total hip arthroplasty (THA) for hip osteoarthritis after a first PFF were included in the overall count of PFFs but were not considered as NSB-PFFs.

Clinical data were extracted from hospital medical records after obtaining explicit consent. Personal history and radiographs of each patient included in the study were reviewed to identify bilateral PFFs, risk factors, fracture pattern, and the type of surgical treatment performed.

The variables collected included sex, date of birth, dates of the two traumatic events, pattern of each fracture (for statistical purposes, AO/OTA 31A fractures were classified as pertrochanteric fractures and AO/OTA 31B fractures as femoral neck fractures), surgery performed (arthroplasty, intramedullary nailing, endoprosthesis, fixation with screws), and laterality of the lesion.

For each patient, the time interval between the two fractures, expressed in years, was calculated. Categorical variables were described as n (%). The distributions of patient age and the time interval between the two fractures were evaluated using the Shapiro-Wilk test. Normally distributed variables were expressed as mean ± SD, whereas non-normally distributed variables were expressed as median, interquartile range (IQR), minimum and maximum.

Associations between categorical variables were analysed using the χ² test of independence. In addition, a multivariate logistic regression analysis was performed to evaluate whether the pattern of the first fracture was independently associated with the pattern of the contralateral fracture. The dependent variable was the type of the second fracture (femoral neck vs pertrochanteric). Independent variables included the type of the first fracture, patient age at the time of the second fracture, and sex. Odds ratios (ORs) with 95%CIs were calculated to estimate the strength of the associations. Statistical significance was set at P < 0.05.

RESULTS

The total number of patients included in the study was 1455. Among them, 1292 patients presented with an isolated PFF, while 153 had NSB-PFFs. One hundred thirty-six patients with a PFFs and a positive history of a previous contralateral fracture of the same skeletal segment were analysed. Nineteen patients were considered drop-outs for lack of complete clinical documentation. The population comprised 79.4% women, with a female-to-male ratio of approximately 4:1. Age distribution was non-normal, with a median age of 86.1 years (IQR 81.6-90.1; range 52-106 years).

Regarding the pattern of the second fracture, 57.4% were pertrochanteric and 42.6% were femoral neck fractures. The distribution for the previous fracture was similar, with a prevalence of pertrochanteric fractures in 53.7% of patients and femoral neck fractures in 46.3%. The association between the type of the previous fracture and that of the most recent one was statistically significant (χ² = 24.17; P < 0.0001), indicating a tendency for patients to present an analogous fracture pattern over time.

The association between fracture pattern and surgical treatment was analysed separately for the previous contralateral fracture and for the most recent fracture (Table 1). For the previous contralateral fracture, femoral neck fractures were treated with a THA in 47.6% of cases, with an endoprosthesis in 31.7% and with screws in 20.6%; no patient with a femoral neck fracture was treated with intramedullary nailing. In contrast, pertrochanteric fractures were treated with intramedullary nailing in 97.3% of cases, while a THA and an endoprosthesis were used in only 1.4% of cases each; screws were never used. This distribution was clearly different between the two groups (χ² = 128.28; P < 0.0001). For the most recent fracture, a similar treatment was observed. Femoral neck fractures were treated with an endoprosthesis in 59.3% of cases, a THA in 8.5%, screws in 22.0% and intramedullary nailing in 8.5%. Pertrochanteric fractures, however, were treated exclusively with an intramedullary nail in 100% of cases, without any use of prosthetic replacement or screw fixation. Again, the association between fracture type and treatment adopted reached very high statistical significance (χ² = 116.88; P < 0.0001). The association between the type of surgery performed for the first fracture and that performed for the second was also highly significant (χ² = 52.85; P < 0.0001).

Table 1 Surgical treatment of the previous fracture compared with the contralateral fracture (most recent).

Type of fracture	Treatment	First fracture (%)	Second fracture (%)
Femoral neck	Total hip arthroplasty	47.6	59.3
	Endoprosthesis	31.7	22
	Screws	20.6	8.5
	Intramedullary nailing	0	8.5
Pertrochanteric	Intramedullary nailing	97.3	100
	Total hip arthroplasty	1.4	0
	Endoprosthesis	1.4	0
	Screws	0	0

The time interval between the two fractures showed a non-normal distribution, with a median of 4.1 years (IQR 1.7-7.1), a minimum of 0.06 years (approximately 23 days), and a maximum of 24.3 years. The distribution was right-skewed, with a higher concentration of cases between two and five years after the previous event (Figure 1).

Figure 1 Frequency distributions of time between non-simultaneous proximal femur fractures.

A multivariate logistic regression analysis was performed to assess whether age, sex, and the type of the first fracture were independently associated with the pattern of the second fracture. After adjustment for age and sex, the type of the first fracture remained a strong independent predictor of the contralateral fracture pattern (OR = 6.71; 95%CI: 3.11-14.48; P < 0.001). Neither age (P = 0.82) nor sex (P = 0.20) showed a significant association with the type of the second fracture.

DISCUSSION

The results of this study-demonstrating a strong tendency to repeat the same fracture pattern contralaterally and identifying a vulnerability window of 2-5 years-are consistent with and add perspective to the international literature.

The incidence of NSB-PFF observed (10.5%) is in line with most published data, which report analogous rates even with a considerable gap (6%-15%)[5,15-18]. The meta-analysis by Zhu et al[11] consolidates this estimate, reporting an average incidence of 8.5%. This confirms that patients who sustain a first hip fracture represent a population at very high risk of contralateral recurrence.

The fracture pattern symmetry observed in our study (Table 1) (significantly correlated, P < 0.0001) is a robust finding. Previous studies have reported that between two-thirds and four-fifths of patients sustain the same fracture type (femoral neck or pertrochanteric) on the contralateral side[15-17]. The multivariate analysis performed in the present study further strengthens this finding. After adjustment for age and sex, the pattern of the first fracture remained a strong independent predictor of the contralateral fracture type. This suggests that the observed symmetry between the two fractures is not simply explained by demographic characteristics but likely reflects patient-specific anatomical and biomechanical factors. These may include femoral geometry, bone mineral density distribution, presence of hip osteoarthritis, and habitual patterns of falls. Particularly, the presence of hip osteoarthritis has been reported to be mostly associated with femoral neck fractures[19].

The strong association observed between the surgical procedures used for the first and second fractures likely reflects the fact that surgical decision-making in hip fractures is largely driven by fracture morphology. Because patients tend to sustain the same fracture pattern contralaterally, the surgical treatment is consequently similar. Rather than representing a surgeon-dependent bias, this finding likely mirrors the reproducibility of the underlying fracture mechanism and anatomical characteristics.

The timing of the second fracture is a crucial element for prevention. Our median interval of 4.1 years, with a concentration between two and five years, is consistent with previous studies that reported median or mean intervals ranging from approximately 2-5.6 years[16,17]. In our study, the “window of vulnerability” for an NSB-PFF lies at the upper end of the previously identified interval. However, an important proportion of second fractures occurs very early: Some studies reported that nearly one-fifth to almost 40% of second fractures take place within the first year after the index event[5,15]. In the analyzed population, a peak in the statistical mode was observed in the second year, and 31% of NSB-PFF occurred after the first year (Figure 1). These findings confirm that the risk of a contralateral fracture is not uniformly distributed over time but is particularly elevated in the period immediately following the index fracture. This early phase likely reflects a combination of factors, including persistent frailty, impaired mobility, incomplete functional recovery, and the persistence of untreated or undertreated osteoporosis[20,21]. At the same time, the occurrence of contralateral fractures several years after the index event suggests that fracture risk may remain elevated over time. However, when the interval becomes long, the second fracture cannot necessarily be interpreted as a direct consequence of the initial event. Rather, the first fracture may represent a marker of underlying frailty and persistent risk factors—such as osteoporosis, aging, and increased fall risk-that continue to predispose patients to fragility fractures.

A limitation of this study is the simplification of fracture patterns into a binary classification. Future studies should more thoroughly investigate the different fracture types and their potential association with the occurrence of NSB-PFF.

Regarding outcomes, some reports have described a worse prognosis in terms of residual mobility and return to pre-fracture living conditions after a second hip fracture[22,23], whereas others, particularly larger series, have not confirmed a dramatic additional functional decline[15]. This discrepancy may be partly explained by a “survivor effect”: Patients who live long enough to sustain a second fracture may represent a more resilient subset. In contrast, one-year mortality after the second fracture is consistently high, with rates above 30% in several studies[5,11,15], and dementia and cardiac comorbidities emerge as powerful predictors of both second fracture and mortality[11].

Comorbidities, in fact, are key drivers identified in the literature. Meta-analytic and population-based data indicate that dementia, malnutrition/cachexia, heart disease, Parkinson’s disease, and visual impairment can double or triple the risk of a second fracture[18,24]. Prevention represents the key element linking these findings. Adherence to osteoporosis therapy remains unacceptably low, despite strong evidence supporting its effectiveness, and preventive strategies such as environmental modifications and structured rehabilitation are still not systematically implemented[17-25]. In this study, the role of clinical comorbidities in the incidence of NSB-PFF was not evaluated. This represents a major limitation of the study, and further studies will be required to better investigate this relationship.

CONCLUSION

A second contralateral fracture is, to a large extent, an event with a predictable pattern and potentially preventable. Its effective prevention requires a cultural shift from an episodic, reactive model of care to a proactive, multidisciplinary one, in which the orthopaedic surgeon, geriatrician, physiatrist, and general practitioner work together to protect this highly vulnerable population from an event that too often marks the onset of irreversible functional decline and institutionalisation.