Study of the proximal femoral fractures mortatlity in elderly patients

Abstracts

Foi feito um estudo de revisão sobre a mortalidade na fratura do fêmur proximal em idosos com base nas publicações mais relevantes do período de 1998 a 2002. Foram incluídos 25 artigos relacionados ao assunto, selecionados com base nos bancos de dados Medline e Cochrane, totalizando 24.062 pacientes com mais de 60 anos de idade, que tiveram fratura do fêmur proximal. Quatorze estudos foram prospectivos, oito retrospectivos e três revisões sistemáticas. As taxas médias de mortalidade foram de 5,5% durante a internação hospitalar, 4,7% ao fim de um mês de seguimento, 11,9% com três meses, 10,8% com seis meses, 19,2% com um ano e 24,9% com dois anos. Foram identificados quatro fatores intimamente relacionados com uma maior mortalidade nestes pacientes: idade avançada, grande número de doenças associadas, sexo masculino e presença de deficiências cognitivas. Outros fatores mostraram uma fraca correlação com a mortalidade como capacidade deambulatória prévia, índice de risco anestésico da Sociedade Americana de Anestesia (ASA), anemia, hipoalbuminemia, linfopenia e existência de AVC prévio. Os fatores como tempo prévio à cirurgia, tipo de anestesia utilizada e tipo de osteossíntese empregada não mostraram ter interferência.

Fraturas do fêmur; Idoso; Mortalidade

REVIEW ARTICLE

Study of the proximal femoral fractures mortatlity in elderly patients

Marcos Hideyo Sakaki^I; Arnóbio Rocha Oliveira^II; Fabrício F Coelho^III; Luiz Eugênio Garcez Leme^IV; Itiro Suzuki^V; Marco Martins Amatuzzi^VI

^IAssistant Physician at the Institute of Orthopedics and Traumatology of the Clinics Hospital of the Medicine School of the São Paulo University, Doctorate Postgraduate Student - FMUSP

^IIPhysician, Doctorate Postgraduate Student - FMUSP

^IIIMastership Postgraduate Student -FMUSP

^IVProfessor Doctor at the Medicine School of the São Paulo University; Coordinator of the Postgraduate Course of the Department of Orthopedics and Traumatology - FMUSP

^VMaster in Medicine, Clinical Director of the Institute of Orthopedics and Traumatology of the Clinics Hospital- FMUSP

^VIChairmain and Head of the Department of Orthopedics and Traumatology of the Medicine School of the São Paulo University. Coordinator of the Postgraduate Course at the Department of Orthopedics and Traumatology- FMUSP

^{Correspondence}

SUMMARY

A revision study was performed on the proximal femoral fractures mortality in elderly patients, based on the most relevant publications during the 1998 to 2002 period. Twenty-five articles related to the subject were included, selected from the Medline and Cochrane databases, totaling 24.062 patients over the age of 60 years, with fracture of the proximal femur. There were 14 prospective trials, 8 retrospective and 3 were systematic reviews. The mean in-hospital mortality was 5,5%, 1-month mortality 4,7%, 3-months 11,9%, 6-months 10,8%, 1-year 19,2% and 2-years 24,9%.

Four factors were identified in these patients closely related with a larger number of deaths: advanced age, polymorbidity, male gender and the presence of cognitive impairment. Other factors, such as a pre-fracture walking ability, ASA score, hemoglobin level, albumin level, lymphocyte count and a previous stroke demonstrated poor correlation with mortality. Finally, factors like time-to-surgery, anesthetic techniques and type of osteosynthesis, had no effect on mortality.

Keywords: Femoral fractures, Aged, Mortality

INTRODUCTION

The fracture of the proximal femur is a common and important cause of mortality and functional loss ^{(12, 14, 25)}. The incidence of this type of fracture with age⁽¹⁾, due mainly to the increase in the number of falls associated to a larger osteoporosis prevalence. It is more commonly related with senior females, resident in the urban areas and institutionalized.

With the increase of the life expectancy and consequently with the largest proportion of seniors in the population, mainly the so-called "big" seniors (those with more than 80 years), the importance of this type of fracture has been increasing in the last years ⁽⁶⁾. In England and Wales, in the biennium 1997/1998, sixty six thousand seniors were hospitalized with a fracture of the femur ⁽⁶⁾, while in the United States it was considered that 350.000 fractures of the femur occurred annually, with a total cost of approximately 6 billion dollars a year, in medical care only⁽¹⁴⁾.

The social and economical cost of the femur fracture is still higher due to the fact that after a period of variable time of hospitalization, the senior patient faces high mortality rates, requiring intensive care doctors and rehabilitation programs during long periods⁽¹⁴⁾. In recent studies, senior patients with a fracture of the proximal femur reached the previous capacity to accomplish the daily tasks in only 17% of the cases after 4 months, and only 43% reacquired the previous walking capacity⁽²⁴⁾.

Besides the social damage due to the femur fracture, the senior had reduced functional reservation and presented a large number of associated chronic diseases, with 70% of the patients tending towards at least two other diseases at the moment of the fracture⁽²⁴⁾ being, therefore, much more subject to complications in the postoperative period, so much immediate as late, presenting three complications, that were serious in 26% of the cases on average, taking an increased risk of death ⁽²⁴⁾.

The fracture of the proximal femur can be intracapsular or extracapsular. In the first type are the fractures of the collum femoris and, in the second, the trans-trochanterian fractures, and both elapse from traumas due to low energy, such as falls. The great increase in the incidence of these fractures in the age group above 65 years of age is due mainly to the installation of the osteoporosis, as well as to the largest incidence of falls in this age group. In spite of the fall prevention measures, precocious treatment of the osteoporosis, incentives to physical activities to regulate and control other diseases, the femoral fractures in the elderly are still very frequent.

The treatment of most of these fractures is surgical, the conservative treatment being reserved only for some incomplete fractures or fractures without deviation. The surgery seeks the reduction and stable fixation of the fracture, using the most varied osteosynthesis methods or, in the specific case of the fracture of the collum femoris with deviation, the prosthetic substitution. Along the time, new osteosynthesis materials have been developed. The fixed length plates of the Jewett and Muller type, formerly used in the treatment of the trans-trochanterian fractures, were substituted by the sliding plates, of the DHS type, or by the intramedular syntheses, as the gamma type pin and PFN (Femoral Proximal Nail). These syntheses provide a more rigid and safe fixation, allowing to precociously start the march, and a smaller degree of mechanical complications in the postoperative period. Besides, the necessary surgical aggression to the implant of the synthesis material is decreasing progressively. The hip prostheses, used in the treatment of the fractures of the collum femoris, are also being developing with the use of new materials and layouts, seeking a smaller index of complications during the postoperative period.

The literature about mortality in the elderly population after the several types of fractures of the proximal femur, and the association of these to the several existing types of treatments, both clinical and surgical is very extensive. The objective of this work is to do a revision of the recent literature (1998 to 2002) of the most significant works about the mortality in the elderly with fracture of the proximal femur.

METHOD

The research was performed in the MEDLINE database, using as search terms the words: mortality, morbosity, prognostic factors and geriatric patients, being limited to the last five years (articles published since 1998). Some studies were selected based on the bibliographical references of the works indicated by MEDLINE. A research was also made in the COCHRANE database, with the same parameters as referred above, seeking systematic revisions and clinical randomized assays about prognostic factors related to the mortality after femur fracture in the geriatric population.

Special attention was given so that the 40 works included in the three systematic revisions^(2,19,20) were not considered separately, what certainly would result in a systematic mistake within the data obtained.

The works found with that research were analyzed by the reviewers Sakaki, Oliveira, and Coelho independently, and selected with the preference for the works published in the higher impact magazines, according to JCR (Journal of Citation Report) criteria, of the prospective and randomized type. In case of doubt, the article was selected by consensus. Agreement evaluation was not performed. Some retrospective works were also selected because they presented a significant casuistry. Three systematic revisions were included as they dealt with important themes related to the mortality in the fracture of the femur in the elderly.

The number of deaths in several periods after fracturing the proximal femur was compiled in a table, together with the total number of senior patients studied, allowing calculating the global mortality rate. Several factors related to mortality were identified and will be discussed later.

RESULTS

The number of selected works was 25, totaling 24.062 patients with age above 60 years, which had a fracture of the proximal femur studied. There were 14 prospective works, 8 retrospective and 3 systematic revisions. The data related to the mortality are presented in Table 1. The mortality rate during hospitalization was 5,5%; one month after the fracture 4,7%; three months after 11,9%; six months after 10,8%; a year after 19,2%, and P = prospectivo, R = retrospectivo, FI = fixação interna de fratura do colo femoral, APQ = artroplastia parcial do quadril, ATQ = artroplastia total do quadril, DA = doenças associadas, CD = cognição diminuída,. U= morte, c = contrôle, c2 = contrôle de 2 anos; two years later 24,9%. In control groups the mortality rate was 3.2% six months following surgery, 5.2% one year following surgery, and 9.6¨[11, 17] two years following surgery.The factors positively correlated with mortality identified more frequently were: advanced age, presence of associated diseases, important cognitive deficiency and the masculine gender.

DISCUSSION

The mortality rate for the geriatric patients found in this revision show that there really is an important influence of the fracture of the proximal femur on the overlife of these patients, mainly if compared with the indexes found in the population without fractures. The mortality rate in the control groups at six months of follow-up is of 3,2%, at the end of a year 5,2% and after two years 9,6%^(11,17), much smaller than found in the patients with fractures, 10,8%, 19,2% and 24,9% respectively. Willig et al⁸ find mortality rates in the patients with fracture that surpass the control group-control in 4,5% in the first month, 6,0% in the third, 3,5% in the first year, 4,5% in the second, 6,5% in the fifth year, and 9,0% in the sixth year.

The high age as a decisive factor of the greater mortality in geriatric patients with a fracture of the femur is mentioned by several authors^{(3,5,7,9,10,22,24)}, and is in agreement with the literature. Two studies deserve prominence: while Alarcón et al⁽¹⁾ conclude that patients with more than 90 years don't present a mortality rate larger than usually found in the literature, Forster et al¹¹, when evaluating only patients over 100 years of age, show that 56% of the patients die at the end of one year, as this rate is 20% larger than the expected mortality for this age, for people without fractures of the femur. In this group, the incidence of deaths during hospitalization (31%) is also significantly larger than in the general geriatric group, emphasizing the need of intense medical care during hospitalization

The existence of other chronic diseases in elderly patients at the moment of the fracture is an important prognostic factor^{(3,4,5,1417,22,24)}, the heart and lung pathologies, renal diseases, diabetes mellitus and strokes (CVA) are the larger influence factors. Van Balen et al⁽²⁴⁾ mentions that 67% of the patients treated present two or more diseases associated to the femoral fracture. The mortality rate, in four months, jumps from zero (for those without any associated disease) to 4% when there is one, 15% when there are two, 26% when there are three and 47% when there are four or more diseases concomitant to the femoral fracture. Meyer et al⁽¹⁷⁾ in the work with a control group concludes that, in the absence of associated diseases, of cognition deficiency and of physical impairment, the mortality rate is not altered when the elderly fractures the femur. The scale of anesthetic risk of the American Society of Anesthesia (ASA) is directly related to the presence of a larger number of chronic diseases. As such, Dzupa et al⁽⁹⁾ find a larger incidence of deaths after the fracture of the femur in the elderly with larger ASA. Those with ASA 1 have 0%, ASA 2 have 4,3%, ASA 3 have 21,3%, ASA 4 have 42,1% and ASA 5 have 68,9% mortality in one year.

The cognitive capacity, as a predictive factor of the mortality in the elderly with fracture of the femur is frequently found in the revised studies^{(1,7,14,17,24)}. Van Balen et al⁽²⁴⁾ use Folstein's mini-mental test in their evaluations, and obtain a mortality rate, at four months, of 38% when the result is between zero and 12, 12% when between 13 and 18, 23% when between 19 and 22 and only 2% when between 23 and 29. Cree et al⁽⁷⁾ show that the average for this same test among the survivors after a fracture of the femur is of 20 points, while among those who die in three months the average is of only nine points. Iuusko et al¹⁵, in a prospective comparative and randomized study, concluded that the intense rehabilitation after the fracture in the patients with light or moderate dementia leads to a longer survival rate than in those that have a standard rehabilitation. The death risk in the dementia patients (10 to 12 points in the Engedal test) is the double than in the lucid patients (15 points)⁽¹⁷⁾. As the increase in mortality is more evident just during the first six months after the fracture, it is discussed if it is due to a temporary cognitive deficit, or delirium, and not of a reduced mental state previous to the fracture.

The masculine gender was more susceptive in relation to mortality^{(3,5,7,8,9,11)}. While a woman with a fracture of the proximal femur has 1,5 times more chance of dying than without a fracture, within the period of two years, a man has seven times more⁽⁷⁾. In a prospective cohort study with a control group, Fransen⁽¹¹⁾ finds out that only the masculine gender as predisposition factor to death, and that all the other items studied, such as age, associated diseases and cognitive deficiency don't have any influence. Davidson et al⁽⁸⁾ state that the largest masculine mortality occurs in all types of fracture of the proximal femur. Some hypotheses are placed to explain such fact⁽⁷⁾: the man's fall is more traumatic than that of a woman. The man has more associated diseases or more serious cognitive deficiency. Cree⁽⁷⁾ compares men with the same cognitive capacities and continues to obtain the same mortality difference. None of the other revised studies is capable of explaining the cause of this difference.

The walking ability previous to the patient's fracture is directly related to the mortality^(5,14,17). They are used as parameters of the marching capacity, the capacity to walk out of the residence⁽¹⁷⁾ or of walking without any type of support⁽⁵⁾. The elderly that get to walk out of house and have a fracture of the femur, present the same death risk as those that don't have a fracture⁽¹⁷⁾.

Gruson et al⁽¹³⁾ study the effect of the presence of the anemia at the initial examination at the time of the hospitalization of the elderly with a fracture of the femur. Anemia is considered as serious when less than 10 g/dl and light when among 10 g/dl and 12 g/dl in women, and between 10 g/dl and 13 g/dl in men. The mortality rate at six and twelve months after the fracture is directly proportional to the degree of anemia. An elderly with a fracture of the femur and serious anemia has five times more risks of dying than one without anemia. Because the need of blood transfusion is directly related to the degree of anemia, the risk of death can be inferred by the amount of blood⁽⁴⁾ received by the patient during hospitalization. The patients with dementia, non-walkers and institutionalized, with a clear evidence of the influence of anemia in the mortality rate, are excluded from this study.

The fractures of the collum femoris in the elderly have osteosynthesis, partial and total hip arthroplasty as treatment options. In spite of the more frequent need for revision surgeries in cases treated with the fixation of the fracture, the rate of late mortality (13 years) it is the same for all of the treatments⁽²³⁾.

Koval et al⁽¹⁶⁾ correlate hypalbuminemia (level of blood albumin below 3,5 g/dl) and the lymphopenia (dosage of total blood lymphocytes less than 1500 u/ml) found in the initial examinations at the time of hospitalization with late mortality. These laboratory alterations indirectly measure the state of the patient's malnutrition. The hypalbuminemia predisposes an increased mortality during hospitalization, while lymphopenia is a predictive of increased mortality one year after the fracture. The authors conclude that the prognostic regarding mortality can be presented through those simple laboratory exams, at the time of hospitalization.

Regional anesthesia in the surgical treatment of the fractures of the proximal femur in the elderly is used with much more frequency than the general in our area. The idea that the first option is much safer than the second one for the geriatric patient is opposed by the study of O'Hara's et al⁽¹⁸⁾, that studied 9.425 elderly patients operated from a fracture of the femur The mortality of the group submitted to the regional anesthesia (3.219 patients) is same as of the group operated with general anesthesia (6.206 patients), respectively 5,4% and 4,4% in 30 days. In spite of being a retrospective study, the large population sample confers a relevant importance to the study. Parker et al⁽¹⁹⁾ reach the same conclusion in a meta-analysis including 17 comparative and randomized studies. Some of these studies show a smaller mortality rate in the first month after the surgery when the regional anesthesia is used. However this difference does not reach a limit of significance.

In a retrospective study, Ramnemark et al⁽²¹⁾, show the influence of CVA in the evolution of elderly patients with a fracture of the collum femoris. Three hundred and ninety two patients are compared with only a fracture of the collum femoris, to 147 patients with CVA and fracture, and the mortality rate at the end of one year increases, respectively, from 16,8% to 29,3%. Even when an adjustment is made as to the gender, the difference continues significant. The increased mortality may be explained by the impaired waking ability that patients with CVA present before the fracture, a characteristic that is not individually studied in this assay.

The time elapsed between the fracture and the surgery seems not to be important in the mortality rate for the elderly with a fracture of the femur, as concluded by Grimes et al², after a study with 8.383 patients. The patients were divided into three groups. The first was operated in the period of 24 hours after the fracture. The second after the 24 hours, but presented no active pathology. The third group was also operated after 24 hours, but presented active pathology. It is understood that a patient without active pathology is the patient whose surgery is postponed for non-clinical reasons, such as, for instance, lack of vacancy in the operating room or bureaucratic impediments. The patients with an active pathology have the surgery postponed by diseases that are temporarily decompensated. The mortality found in the group operated quickly (less than 24 hours) is seemingly smaller. However, when the groups are homogenized as to the presence of associated diseases, this difference disappears.

As there seems to be a direct relationship of the mortality in elderly with a fracture of the hip and malnutrition⁽¹⁶⁾, several authors try check on influence of feeding in the long term prognostic of these patients. Food supplementation with proteins, vitamins and mineral salts was studied in a meta-analysis by Avenell et al⁽²⁾ and, in spite of the low quality of the studies in methodological terms, it was concluded that there are no differences in the mortality among patients fed regularly and patients receiving supplements.

Parker et al⁽²⁰⁾ show that there are no literary evidences that the surgical treatment of the fractures of the proximal femur in the elderly reduces the mortality, in relation to the conservative treatment. In this meta-analysis study the authors selected the randomized and quasi-randomized studies that compare those two types of treatment, and found only five studies, involving 428 patients. In addition, most of these studies present a low quality scientific methodology, according to the authors' evaluation. New comparative studies would be necessary to confirm these findings. However it is very difficult, at this time, to indicate a conservative treatment for a higid elderly. This review is important because it allows us to choose the conservative treatment in some selected cases of fracture of the hip, as for instance in the elderly with some impediment for the surgery.

CONCLUSION

We can conclude that the recent literature identified four great factors directly related with the mortality in elderly with a fracture of the proximal femur, the patient's advanced age, presence of associated diseases, of the masculine gender and the existence of cognitive deficiencies. The intra-hospital mortality, until the end of one month, three months, six months, one year and two years were 5,52%, 4,74%, 11,88%, 10,76% 19,24% and 24,94% respectively. Other decisive factors in the mortality, such as walking ability capacity previous to the fracture, ASA index, anemia, hypalbuminemia, lymphopenia and the existence of CVA were found in some isolated studies. The type of surgical treatment in the fractures of the collum femoris, the type of anesthesia used and the time elapsed before the surgery didn't present a correlation with the mortality.

REFERÊNCIAS BIBLIOGRÁFICAS

Trabalho recebido em 19/02/2004.

Aprovado em 05/07/2004

Work performed at the Institute of Orthopedics and Traumatology of the Clinics Hospital of the Medicine School of the São Paulo University - FMUSP

• 1. Alarcon T, Gonzalez-Montalvo JI, Barcena A, Saez P. Further experience of nonagenarians with hip fractures. Injury 32:555-558, 2001.
• 2. Avenell A, Handoll HH. Nutritional supllementation for hip fracture aftercare in the elderly. Cochrane Database Syst Rev 1:CD001880, 2004.
• 3. Bovy P, Jolly S, Dropsy S, Sacre F. Results of rehabilitation on quality of walking and outcome in elderly patients following femoral neck fracture. Development after one year]. Ann Readapt Med Phys 45:19-25, 2002.
• 4. Carson JL, Terrin ML, Barton FB et al. A pilot randomized trial comparing symptomatic vs. hemoglobin-level-driven red blood cell transfusions following hip fracture. Transfusion 38:522-529, 1998.
• 5. Chariyalertsak S, Suriyawongpisal P, Thakkinstain A. Mortality after hip fractures in Thailand. Int Orthop 25:294-297, 2001.
• 6. Clague JE, Craddock E, Andrew G, Horan MA, Pendleton N. Predictors of outcome following hip fracture. Admission time predicts length of stay and in-hospital mortality. Injury 33:1-6, 2002.
• 7. Cree M, Soskolne CL, Belseck E et al. Mortality and institutionalization following hip fracture. J Am Geriatr Soc 48:283-288, 2000.
• 8. Davidson CW, Merrilees MJ, Wilkinson TJ, McKie JS, Gilchrist NL. Hip fracture mortality and morbidity-can we do better? NZ Med J 114:329-332, 2001.
• 9. Dzupa V, Bartonicek J, Skala-Rosenbaum J, Prikazsky V. [Mortality in patients with proximal femoral fractures during the first year after the injury]. Acta Chir Orthop Traumatol Cech 69:39-44, 2002.
• 10. Forster MC, Calthorpe D. Mortality following surgery for proximal femoral fractures in centenarians. Injury 31:537-539, 2000.
• 11. Fransen M, Woodward M, Norton R, Robinson E, Butler M, Campbell AJ. Excess mortality or institutionalization after hip fracture: men are at greater risk than women. J Am Geriatr Soc 50:685-690, 2002.
• 12. Grimes JP, Gregory PM, Noveck H, Butler MS, Carson JL. The effects of time-to-surgery on mortality and morbidity in patients following hip fracture. Am J Med 112:702-709, 2002.
• 13. Gruson KI, Aharonoff GB, Egol KA, Zuckerman JD, Koval KJ. The relationship between admission hemoglobin level and outcome after hip fracture. J Orthop Trauma 16:39-44, 2002.
• 14. Hannan EL, Magaziner J, Wang JJ et al. Mortality and locomotion 6 months after hospitalization for hip fracture: risk factors and risk-adjusted hospital outcomes. JAMA 285:2736-2742, 2001.
• 15. Iuusko TM, Karppi P, Avikainen V, Kautiainen H, Sulkava R. Randomised, clinically controlled trial of intensive geriatric rehabilitation in patients with hip fracture: subgroup analysis of patients with dementia. BMJ 321:1107-1111, 2000.
• 16. Koval KJ, Maurer SG, Su ET, Aharonoff GB, Zuckerman JD. The effects of nutritional status on outcome after hip fracture. J Orthop Trauma, 13:164-169, 1999.
• 17. Meyer HE, Tverdal A, Falch JÁ, Pedersen JI. Factors associated with mortality after hip fracture. Osteoporos Int. 11:228-232, 2000.
• 18. O'hara DA, Duff A, Berlin JÁ. The effect of anesthetic technique on postoperative outcomes in hip fracture repair. Anesthesiology 92:947-957, 2000.
• 19. Parker MJ, Handoll HH, Griffiths R. Anaesthesia for hip fracture surgery in adults. Cochrane Database Syst Rev 4:CD000521, 2001.
• 20. Parker MJ, Handoll HH, Bhargara A. Conservative versus operative treatment for hip fractures. Cochrane Database Syst Rev 4:CD000337, 2000.
• 21. Ramnemark A, Nilsson M, Borssen B, Gustafson Y. Stroke, a major and increasing risk factor for femoral neck fracture. Stroke 31:1572-1577, 2000.
• 22. Raunest J, Engelmann R, Jonas M, Derra E. [Morbidity and mortality in para-articular femoral fractures in advanced age. Results of a prospective study]. Unfallchirurg 104:325-332, 2001.
• 23. Ravikumar KJ, Marsh G. Internal fixation versus hemiarthroplasty versus total hip arthroplasty for displaced subcapital fractures of femur-13 year results of a prospective randomised study. Injury 31:793-797, 2000.
• 24. Van Balen R, Steyerberg EW, Polder JJ, Ribbers TL, Habbema JD, Cools HJ. Hip fracture in elderly patients: outcomes for function, quality of life, and type of residence. Clin Orthop 390:232-243, 2001.
• 25. Willig R, Keinanen-Kiukaaniemi S, Jalovaara P. Mortality and quality of life after trochanteric hip fracture. Public Health 115:323-327, 2001.

Publication Dates

• Publication in this collection
  01 Mar 2005
• Date of issue
  Dec 2004

History

• Received
  19 Feb 2004
• Accepted
  05 July 2004