Association of Serum Vitamin D and Calcium Levels With the Severity of Intertrochanteric Fractures in the Elderly: A Retrospective Study


 Background: Femur intertrochanteric fractures can be classified into stable and unstable fractures according to their severity. Postoperative complications and mortality were more common in patients with unstable intertrochanteric fractures. However, there has been little effort to evaluate the risk factors of the two fracture types. This study aimed to identify the possible differences in demographic and clinical characteristics of older patients with different types of intertrochanteric fractures.Methods: The medical records of patients aged ≥60 years who presented with intertrochanteric fractures from June 2018 to March 2020 were retrospectively reviewed. Fifty-seven patients were enrolled and divided into two groups according to severity: stable (21 patients) and unstable (36 patients). Demographic data, body mass index (BMI), ambulatory ability prior to fracture, pre-fracture residence, season at fracture, bone mineral density (BMD), and serum 25-hydroxyvitamin D [25(OH)D], osteocalcin, and calcium levels were compared between the two groups. Additionally, we analyzed the correlation among variables.Results: The stable group had significantly higher serum 25(OH)D and calcium levels than the unstable group (p = 0.010, p = 0.019). There were no statistically significant differences (p > 0.05) in age, sex, height, weight, BMI, ambulatory ability prior to fracture, pre-fracture residence, season at fracture, BMD, and serum osteocalcin level between the two groups. In addition, serum 25(OH)D and calcium levels were not correlated with any of the variables in all patients.Conclusion: Low vitamin D and calcium levels are associated with unstable intertrochanteric fracture in elderly patients. Maintaining adequate vitamin D and calcium levels could avoid an increase in the severity of intertrochanteric fractures.

Intertrochanteric fractures can be classi ed into stable and unstable according to their severity. The characteristics of unstable fractures include reverse obliquity, loss of posterior medial buttress, lateral wall comminution, and subtrochanteric extension [11]. These di cult patterns are associated with longer operative time, increased need for open reduction, hardware failure, head cutout, varus malunion, and nonunion [12,13]. Medical complications such as myocardial infarction, pneumonia, and urinary tract infections frequent occur after surgery of patients with unstable intertrochanteric fractures [4]. Mortality after surgical xation was higher in patients with more unstable fracture patterns, because unstable fractures are associated with increasing age and functional dependence [4,12,14].
Although unstable intertrochanteric fractures have a variety of complications and poor prognosis, it remains unclear whether the risk factors for fracture vary by severity of intertrochanteric fracture. It is important to determine whether the interactions between variables differ by type of intertrochanteric fracture and which risk factors can be used in clinical practice. Understanding the relationships between stable and unstable fractures can help in personalizing preventive measures, improving prognosis, planning of treatment, and providing differentially targeted interventions. From this, we aimed to evaluate the possible differences in demographic and clinical characteristics of older patients with different types of intertrochanteric fractures. Additionally, we analyzed the correlation among these variables.

Study design and population
We conducted a single-center, retrospective cross-sectional study. Patients aged 60 years or older who were diagnosed with femur intertrochanteric fracture in our hospital from June 2018 to March 2020 were selected. We retrospectively reviewed the patients' medical records. Patients (1) with intertrochanteric fractures due to high-energy trauma, such as vehicular accidents or falls from a height of > 2 m; (2) with pathological fractures; (3) who had previous hip surgery or trauma on either ipsilateral or contralateral hip; and (4) were currently taking anti-osteoporotic drugs or vitamin D and/or calcium or had a past history of taking such drugs were excluded from the study. Fractures were de ned according to the Orthopedic Trauma Association (OTA) classi cation, and were categorized into two groups: the stable group, which included A1.1 through A2.1, and the unstable group, which included A2.2 through A3.3 [15].
The patients were divided into two groups according to this classi cation: stable group and unstable group.

Data collection
The following demographic and clinical outcomes data were obtained from the electronic medical records: age, sex, height, weight, body mass index (BMI), ambulatory ability prior to fracture, pre-fracture residence, season at fracture, BMD and serum 25-hydroxyvitamin D [25(OH)D], osteocalcin, and calcium levels. Body height and weight were measured, and BMI was calculated using the standard formula. The ambulatory ability prior to fracture was assessed according to the Koval's classi cation (grade I: independent community ambulator, grade II: community ambulator with cane, grade III: community ambulator with walker/crutches, grade IV: independent household ambulator, grade V: household ambulator with cane, grade VI: household ambulator with walker/crutches, and grade VII: nonfunctional ambulator) [16]. Pre-fracture residence was classi ed as living in the community (community) and residing in aged care facilities (nursing home). Seasons during the occurrence of fracture were divided into spring (March, April, and May), summer (June, July, and August), fall (September, October, and November), and winter (December, January, and February). BMD (in g/cm 2 ) was measured by performing a dual-energy X-ray absorptiometry (DXA) (Discovery W; Hologic Inc., Marlborough, MA, USA) in the femoral neck, trochanteric region, intertrochanter region, Ward's triangle region, and the total hip region of the contralateral proximal femur. However, the Ward's triangle region of the hip should not be used for this study, because this site overestimates osteoporosis [17]. BMD measurements were routinely obtained not later than 1 week following fracture.
For biochemical measurements, patients' blood samples were obtained after an overnight fast of at least 8 hours, and their serum 25(OH)D, osteocalcin, and calcium levels were measured. Serum 25(OH)D was measured by performing a chemiluminescence immunoassay method using the UniCel DxI 800 (Beckman Coulter Diagnostics, CA, USA). Serum osteocalcin was measured by performing an electrochemiluminescence immunoassay method using the N-MID osteocalcin kit (Roche Diagnostics, Mannheim, Germany). Serum calcium was measured by the O-cresolphthalein method using the 7180 Clinical Analyzer (Hitachi Ltd., Tokyo, Japan). Vitamin D status was classi ed as follows: normal vitamin D status (serum 25(OH)D level ≥ 30 ng/mL) and low serum vitamin D levels, which can be further subcategorized into insu ciency (20-30 ng/mL) and de ciency (< 20 ng/mL) [18].

Statistical analysis
Continuous variables were expressed as means ± standard deviation and categorical variables as percentages. The normality of the data was assessed using the Shapiro-Wilk test. In order to verify the differences between groups of variables, categorical variables (sex, ambulatory ability prior to fracture, pre-fracture residence, and season at fracture) were analyzed using chi-square test, while continuous variables (weight, height, BMI, BMD, and serum calcium levels) were analyzed using an independent sample t-test. Because of the uneven distribution of the data for age, serum vitamin D, and osteocalcin, the Mann-Whitney U test was performed. In order to evaluate the correlation among variables, a Spearman correlation analysis was performed. Statistical analyses were performed using IBM SPSS software, version 25.0 (SPSS Inc. Chicago, IL, USA) for Windows. For all analyses, the level of signi cance was set at p < 0.05.

Patient characteristics
A total of 68 patients with femoral intertrochanteric fractures were identi ed. Of them, 57 patients were included in the study. Eight patients with previous hip fractures and three who were taking antiosteoporotic drugs or vitamin D were excluded. The mean age of these patients was 81.93 years (range, 60-95 years), and 41 (71.9%) were women. According to the OTA classi cation, 21 patients comprised the stable group (A1.1-A2.1), while 36 comprised the unstable group (A2.2-A3.3). Approximately 57.9% of the patients showed vitamin D de ciency, 26.3% showed insu ciency, and 15.8% showed normal vitamin D levels.

Differences between the stable group and unstable group
The patients in the unstable group (82.72 ± 9.76 years) were slightly older than those in the stable group (80.57 ± 9.35 years), but this difference was not statistically signi cant (p = 0.328) ( Table 1). The femaleto-male ratio was similar between the two groups (p = 0.585). No statistically signi cant differences were observed in both groups in terms of height, weight, BMI, ambulatory ability prior to fracture, pre-fracture residence, and season at fracture (p > 0.05). Compared with patients with stable fractures, those with unstable fractures had lower BMD in the femoral neck, trochanteric region, intertrochanter region, and total hip region, but this difference was not statistically signi cant (p > 0.05). The mean serum 25(OH)D and calcium levels were 27.01 ± 15.75 ng/mL and 8.48 ± 0.45 mg/dL in the stable group and 17.52 ± 8.65 ng/mL and 8.19 ± 0.43 mg/dL in the unstable group. The stable group had signi cantly higher serum 25(OH)D and calcium levels than the unstable group (p = 0.010, p = 0.019). In the stable group, 38.1% of the patients had a vitamin D de ciency and 28.6% had a vitamin D insu ciency ( Table 2). In the unstable group, 69.4% of the patients had a vitamin D de ciency, while 25.0% had a vitamin D insu ciency. More patients in the unstable group had low vitamin D status (p = 0.012). However, there were no statistically signi cant differences in serum osteocalcin level between the two groups (p = 0.914).  The p values represent the differences between stable and unstable group.

Correlation between variables
These analyses revealed similarities and differences between variables associated with the type of intertrochanteric fractures (Table 3). In all patients, age showed a signi cant positive correlation with the ambulatory ability prior to fracture (r = 0.337, p = 0.010) and a signi cant negative correlation with height The worse the ambulatory ability prior to fracture, the lower the neck BMD. Pre-fracture residence, which were coded as community = Height was relatively low in patients living in nursing homes. Neck BMD, trochanteric BMD, intertrochanter BMD, and total BMD showed a signi cant positive correlation with each other (p < 0.001). However, serum 25(OH)D and calcium levels were not correlated with any of the variables in the total group (p > 0.05).

Discussion
Because unstable intertrochanteric fractures of the hip have a greater bony injury, technically they are much more challenging than stable fractures [4]. Unstable fractures have greater potential to displace or result in nonunion [19]. Typically closed reduction required for xation of intertrochanteric fractures, but open reduction seems to be necessary for complex fracture patterns [12]. Medical complications after operation of unstable intertrochanteric fractures frequently occur. The most common complications are myocardial infarction, pneumonia, and urinary tract infections [4]. Mortality rate after surgical xation is higher in patients with more unstable fracture patterns [12,14]. Functional outcomes may also vary based on the severity of hip fracture. Functional independence is worse in patients with unstable intertrochanteric fractures than in those with stable fracture [20]. Thus, unstable fractures are recognized as di cult problem for both surgeons and elderly patients. Hence, we investigated which variable determines the severity of the fracture when intertrochanteric fractures occur.
Previous studies have examined the relationship between BMD and hip fracture severity. Cauley et al.
found that patients with more stable fractures (intra-and extracapsular) had a mean lower hip BMD than those with unstable fractures [21]. They assumed that this could be because the lower the BMD, the weaker the mechanical forces that are needed to cause the hip fracture, thus leading to less displacement and more stable fractures. Spencer et al. reported that there was no difference in BMD measured with DXA between stable and unstable hip fractures [22]. Our results showed that patients with unstable fractures had lower BMD in all regions than those with stable fractures. However, this difference was not statistically signi cant (p > 0.05). Another study evaluated the relationship between vitamin D and the severity of hip fracture. Larrosa et al. reported an association between vitamin D level and the severity of osteoporotic hip fractures (femur neck and intertrochanteric fractures) [23]. The authors concluded that the largest vitamin D de ciency was associated with a more severe type of hip fracture. However, serum calcium levels were not different between less severe fracture and more severe fracture. The present study reported that serum 25(OH)D levels as well as serum calcium levels were signi cantly different between the stable group and unstable group. The patients with unstable fractures had signi cantly lower serum vitamin D and calcium levels. Correspondingly, more patients were categorized as having low vitamin D levels in the unstable group. In the stable group, 28.6% of the patients had vitamin D insu ciency, while 38.1% had vitamin D de ciency. In the unstable group, 25.0% of the patients had vitamin D insu ciency, while 69.4% had vitamin D de ciency. Irving et al. studied the relationship between BMI and stability of intertrochanteric fracture [24]. They reported that unstable intertrochanteric fracture were found more frequently in obese patients (BMI > 30) than in those who were not obese, but this difference did not reach statistical signi cance. Another retrospective study by Chen et al. investigated the factors affecting the stability of intertrochanteric fractures in elderly patients [25]. They concluded that the stability of intertrochanteric fractures cannot be predicted by patient's age, gender, body weight, body height, and BMI. The results of the present study demonstrate that BMI was slightly higher in the stable group than in the unstable group, but there was no statistically signi cant difference.
Vitamin D status has been associated with the development and maintenance of mineralized bone and muscle function [26]. The insu ciency of vitamin D increases the secretion of parathyroid hormone to promote bone resorption, resulting in osteopenia and osteoporosis [27]. In addition, vitamin D de ciency weakens muscle strength, increases the risk of falls, and consequently increases the risk of fracture [28,29]. Vitamin D is necessary for the regulation of calcium and phosphate in the human body, and decreased levels can alter the bone mineralization process. There is a high prevalence of vitamin D de ciency in patients with osteoporosis, and low vitamin D is associated with a reduction in bone strength and increased fracture risk [30]. Meanwhile, higher levels of 25(OH)D and vitamin D supplementation may increase hip bone strength via effects on femur geometry, including the cortical thickness, cortical volume, cross-sectional area, and cross-sectional moment of inertia [31,32]. Calcium is also a substantial component of bone. Calcium plays an important role in bone mass and bone strength [33]. Low calcium intake increases the parathyroid hormone secretion and bone resorption, thus increasing bone turnover, bone loss, and risk of fractures [34]. In addition, calcium is plays an essential role in the muscle physiology and skeletal-muscle interaction. Hence, adequate calcium levels are important for both the bones and muscles [35]. In the present study, the higher severity of fracture in patients with low vitamin D and calcium levels is a result of the combined effects of vitamin D and calcium on both bones and muscle metabolism. This relationship and its pathophysiology can be further explored in larger, dedicated trials in the future. Currently, it remains unclear how vitamin D and calcium could have in uenced the severity of fracture.
The association of vitamin D, calcium, and BMD has been studied over the past decades, but the presence of a correlation between vitamin D, calcium, and BMD remains controversial. In a study conducted in southern California among community-dwelling postmenopausal older women, 25(OH)D levels were positively and independently associated with BMD, while PTH levels were negatively and independently associated with BMD [36]. A Dutch population-based study showed that all BMD values were higher in the higher serum 25(OH)D groups than in the lowest serum 25(OH)D group, although this was only signi cant for the total hip, femoral trochanter, and total body bone mineral content [37]. A Saudi study showed no signi cant correlation between vitamin D and spine or total femoral BMD in a community-based population [38]. Japanese study found that there is positive association between serum 25(OH) D and BMD of the femoral neck, but no association was found between serum 25(OH) D and BMD of the lumber spine in home-dwelling postmenopausal women [39]. A Chinese populationbased study did not nd a signi cant correlation between 25 (OH) D and calcaneus BMD in middle-aged and elderly individuals [40]. Liu  There are several limitations to our analysis. First, an independent interpretation of the radiographs to classify the fractures was not provided. The classi cation of the fracture type and severity could have been affected by interobserver variability, although experienced orthopaedic surgeons were asked to examine the images. Second, the number of study participants was relatively small to allow a de nite conclusion. Therefore, additional large-scale, prospective studies are warranted to supplement and further validate our results. Third, there was a possibility of inadequate assessment of the facts such as injury mechanism, history of anti-osteoporosis treatment, previous trauma, and other secondary osteoporosis causes, because the review was limited to medical records. For example, if a patient has cognitive impairment due to old age, the wrong information might be recorded.

Conclusion
We believe that the lower severity of intertrochanteric fractures is important for elderly populations, allowing them to return to higher functional activities and reducing the overall morbidity and mortality. Committee. Because this study was the retrospective analysis, the requirement for obtaining an informed consent was waived.

Consent for publication
Not applicable.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
SMH and YHK designed the study. SHH collected the subject data. SMH, SHH, and YHK performed the data analysis and interpreted the results. SMH and YHK contributed to the review and editing of the manuscript and revising it critically. All authors were involved in writing the manuscript. All authors read and approved the nal submitted manuscript.