Low serum levels of vitamin D significantly increase the risk of death in older adults with hip fractures: a prospective cohort

ABSTRACT Objective: to evaluate the relationship between 25(OH)D3 levels and fatal outcome in patients over 60 years of age undergoing surgical repair of hip fractures. Methods: prospective cohort of patients undergoing surgical repair of hip fractures. At admission, 25(OH)D3 levels were measured, among other parameters. Patients were followed for at least 1 year, and incident mortality was recorded. Results: 209 patients were included in the study, with a mean age of 79.5 ± 7.6 years among survivors and 80.7 ± 8.2 years among those who died in the first postoperative year (p=0.346). The 25(OH)D3 levels of survivors were significantly higher than those of patients who died (p=0.003). After adjusting for confounding variables, 25(OH)D3 levels below 12.5ng/mL were significant risk factors regardless of mortality (adjusted OR: 7.6; 95% CI: 2.35 to 24.56). Conclusions: our data show that serum 25(OH)D3 levels below 12.5ng/mL significantly and independently increased the risk of mortality in the first year after surgical repair of low-energy hip fracture in patients over 60 years of age in the geographic region where this study was conducted. Low albumin also showed a significant association with mortality in these patients. All other factors had no significant associations.


INTRODUCTION
H ip fractures have a significant impact on health both at the society 1 and individual level 2,3 .Outcomes vary widely from country to country 4 , from region to region, and across ethnicities 5 .In southern Brazil, hip fractures are highly prevalent and carry high mortality rates in the white population 6 .Other factors are also important, such as the influence of climate, with a higher incidence of falls and fractures in the winter months 7,8 .
The morbidity and mortality rates of this condition vary according to several risk factors 9,10 .Among these, those related to nutritional status interfere with both clinical outcomes and mortality rates 11,12 .Low albumin levels are associated with unfavorable outcomes 13 , whereas low 25(OH)D 3 levels are associated with an increased overall risk of fractures in older persons [14][15][16][17] .Some studies have analyzed possible associations of 25(OH)D 3 with morbidity and mortality after surgical treatment of hip fractures 18,19 .
In caring for older patients with hip fracture, knowing the factors that can identify those at increased risk of death provides a better understanding of the clinical conditions that of these patients, which can improve the care of this complex patient population.Within this Guerra Low serum levels of vitamin D significantly increase the risk of death in older adults with hip fractures: a prospective cohort context, the present study was designed to assess the association between preoperative serum 25(OH)D 3 levels and 1-year mortality in older adults (age >60 years) undergoing surgical repair of hip fractures in the southern region of Brazil.

Study design
Prospective cohort.

Participants
This study builds upon a previous retrospective cohort study 17 , which after the publication of these results, other patients were included in the cohort.Since the preoperative period, all patients were followed by a team of hospitalists.Upon admission, hospital social workers assessed each patient's conditions to ensure a safe return to their place of origin (home or otherwise).Postoperatively, patients were followed by the hospital Orthopedics and Trauma Surgery team for at least one year.

Outcome Variables
The primary outcome of interest was mortality.

Assessment of Covariates
As noted above, all patients had serum levels of several markers measured upon admission.Information was collected on the following variables: smoking and alcohol intake; preexisting diagnosis of dementia, hypertension, diabetes mellitus, chronic kidney disease, AIDS, hepatitis, delirium, and deep vein thrombosis; weight and height; C-reactive protein, vitamin D, albumin, urea, creatinine, and glucose; and complete blood count with differential.As per institutional protocol, a cardiologist calculated the score for risk of cardiac complications proposed by Lee et al 20 .

Sample Size
Assuming a 1-year mortality rate of 20% for the reference group (serum 25(OH)D3 >20 ng/mL) and 40% for the exposed group (serum 25(OH)D 3 ≤15ng/ mL), 82 patients per group would provide a statistical power of 80% with a two-sided type I error of 0.05.

RESULTS
Of 289 patients, 236 met the inclusion criteria.
Of these, 209 had serum 25(OH)D 3 levels measured and were included in the present analysis (Figure 1).Among the fractures, 44% were neck of femur fractures, 47% were trochanteric, and 9% were subtrochanteric.Fifty-    Our study has limitations.The socioeconomic specificities of our region and the ethnic characteristics of our population may have caused some selection bias.However, we believe that this limitation also represents an opportunity to learn about the behavior of these factors in other scenarios.
We understand that the influence of vitamin D on the outcomes of patients with hip fractures needs to be further investigated.Some questions remain unanswered: is there room for vitamin D replacement in these patients?For all patients?At what dose?Starting when?Until when?What is the association of serum 25(OH)D 3 levels with functional outcomes?Do we have data to implement public our state is geographically located close to the 30 th South parallel and the ethnic makeup of the southern Brazilian population is overwhelmingly white (>80%) 22 , studies assessing the effects of serum 25(OH)D 3 levels in patients with hip fractures are necessary.Our sample is similar to those previously reported in the literature in terms of age and sex 18,19 .levels and all-cause mortality has been well studied 23,24 , with levels below 30ng/mL being associated with an increased risk of mortality.Other authors have also found associations between 25(OH)D 3 levels and mortality.Toldy et al. 19 , in a case-control study, had results similar to ours.Fisher et al. 21 were patients aged >60 years who underwent surgical treatment for low-energy hip fractures between January 2015 and December 2016.Patients with refracture, high-energy trauma, or concomitant infectious diseases, patients who did not have 25(OH)D 3 levels measured preoperatively, and those who were nonambulatory at baseline were excluded.After application of the inclusion and exclusion criteria, 236 additional patients were enrolled.Of these, 27 patients did not have 25(OH)D 3 levels recorded.Therefore, 209 patients were analyzed in the present study (Figure1).

Figure 2 .
Figure 2. Institutional protocol for management of hip fractures in older adults.DHS, dynamic hip screw; PFN, proximal femoral nail; THR, total hip replacement; PHR, partial hip replacement.
day of vitamin D supplementation, showed that in men, who are known to be at higher risk of mortality due to hip fractures, the decrease in mortality risk with vitamin D supplementation was not significant.In addition to the selection bias reported in the study, supplementation was neither regular nor related to serum 25(OH)D 3 levels.Lee et al.25 showed a relationship between low 25(OH)D 3 levels and mortality, in univariate analyses.However, unlike our study, they did not find a significant relationship in the multivariate analyses.Our results raise the issue of vitamin D replacement.A PubMed search combining ("hip fractures"[MeSH Major Topic]) AND ("vitamin D/ administration and dosage"[MeSH Major Topic]) returned 3 articles 26-28 .Or et al. 26 , in addition to other measures, implemented a protocol for administration of 2000IU of vitamin D/day in patients with 25(OH)D 3 levels above 20ng/mL.If 25(OH)D 3 levels were below 20ng/mL, patients received a loading dose of 75,000IU.These patients were monitored by health professionals with the purpose of maintaining adherence to the protocol.Despite a decrease in mortality rates, the authors recognized the methodological weaknesses of the study.Thorpe et al. 27 reported a decrease in the risk of hip fracture among white female vegans with vitamin D supplementation.However, they did not investigate the postoperative period and the demographic and dietary profile of the sample is very specific.Al-Khalidi et al. 28 concluded that vitamin D supplementation is effective in the prevention of hip fractures.

Figure 3 .
Figure 3. Bar plot of 1-year mortality rate and vitamin D3 level.

Table 1 .
Baseline characteristics of patients.
GuerraLow serum levels of vitamin D significantly increase the risk of death in older adults with hip fractures: a prospective cohort

Table 2 .
Incidence of 1-year mortality among patients undergoing hip fracture surgery, stratified by preoperative serum vitamin D levels, Porto Alegre, Brazil.OR obtained in a logistic regression model adjusted for the confounding effects of age, gender, and serum albumin levels ≤ 3 g/dL.
a Body mass index.bNeutrophil-to-lymphocyteratio.c C-reactive protein.d Blood glucose level.DISCUSSION Hip fracture is an important public health problem worldwide.In the past five years, more than 4.846 articles with a MeSH Major Topic of Hip Fracture have been indexed in MEDLINE.However, a search query combining MeSH Major Topic Hip Fracture AND ("Vitamin D"[MeSH Major Topic], only one result was found in the past five years 21 .To date, to our knowledge, no study has correlated serum 25(OH)D 3 levels with mortality rate after hip fracture in South America.Considering that Rev Col Bras Cir 49:e20223054 Guerra Low serum levels of vitamin D significantly increase the risk of death in older adults with hip fractures: a prospective cohort Objetivo: avaliar a relação entre os níveis de 25(OH)D 3 e o desfecho fatal em pacientes acima de 60 anos submetidos a correção cirúrgica de fratura de quadril.Métodos: coorte prospectiva de pacientes submetidos a correção cirúrgica de fraturas de quadril.À admissão, foram medidos os níveis de25(OH)D 3 , entre outros parâmetros.Os pacientes foram acompanhados por, pelo menos, um ano e a mortalidade foi registrada.Resultados: foram incluídos 209 pacientes no estudo, com média de idade de 79,5 ± 7,6 anos entre os sobreviventes e 80,7 ± 8,2 anos entre os que morreram no primeiro ano de pós-operatório (p=0,346).Os níveis de 25(OH)D 3 dos sobreviventes foram significativamente maiores do que os dos pacientes que morreram (p=0,003).Após o ajuste para variáveis de confusão, níveis de 25(OH)D 3 abaixo de 12,5ng/mL foram fator de risco significativo, independentemente da mortalidade (OR ajustado 7,6; IC 95% 2,35 24,56).Conclusões: níveis séricos de 25(OH)D 3 abaixo de 12,5ng/mL aumentaram significativa e independentemente o risco de morte no primeiro ano após o reparo cirúrgico de fratura de quadril de baixa energia em pacientes com mais de 60 anos de idade na região geográfica onde este estudo foi realizado.A albumina baixa também teve associação significativa com a mortalidade nesses pacientes.Todos os outros fatores não tiveram associações significativas.Albumina.Envelhecimento Mortalidade.Estado Nutricional.Vitamina D.Our data showed that low albumin levels are also associated with 1-year mortality.Although the logistic regression models did not associate the two factors, we understand that further studies are needed to assess the real role of albumin in these scenarios.Finally, our study is sufficiently powered to allow us to state that serum albumin levels below 3g/dL as well as serum 25(OH)D 3 levels below 12.5ng/mL significantly and independently increase the risk of mortality in the first year after surgical repair of low-energy hip fracture in patients over 60 years of age.
18in multivariate analyses, found seven variables that independently increased in-hospital mortality among patients treated for hip fracture, including serum 25(OH)D 3 levels below 25nmol/L.Whiting et al.18, using data from a cohort of 23,178 individuals divided into a group without vitamin D supplementation and a group with a minimum of 800IU/