European Journal of Orthopaedic Surgery & Traumatology

, Volume 22, Issue 6, pp 467–472

Role of hemiarthroplasty in intertrochanteric fractures in elderly osteoporotic patients: a case series


  • Ramchander Siwach
    • Department of OrthopaedicsPGIMS
    • Department of OrthopaedicsPGIMS
  • Roop Singh
    • Department of OrthopaedicsPGIMS
  • Kapil Sangwan
    • Department of OrthopaedicsPGIMS
Original Article

DOI: 10.1007/s00590-011-0870-2

Cite this article as:
Siwach, R., Jain, H., Singh, R. et al. Eur J Orthop Surg Traumatol (2012) 22: 467. doi:10.1007/s00590-011-0870-2



Management of intertrochanteric fractures in elderly osteoporotic patients continues to be a challenging problem for orthopedic surgeons. Cutting out of implant from the femoral head and varus malpositioning of fragments is often seen in such cases if early ambulation is allowed. Prolonged bed rest in elderly patients leads to higher risk of complications such as bed sores, pneumonia and deep vein thrombosis not uncommonly leading to fatal pulmonary embolism. The purpose of this case series study was to assess the role of hemiarthroplasty in the treatment of intertrochanteric fractures in elderly patients and study the complications.

Materials and methods

Twenty-five patients were treated at a tertiary care center with hip hemiarthroplasty in intertrochanteric fractures. Mean age of the patients in the study was 77.8 years. Young patients with stable fractures and patients with active infection were excluded from study. Seventy-two percent of patients in the study were osteoporotic and associated comorbidity was present in 18 patients. Preoperative ambulatory status of all patients was noted for comparison. If calcar was deficient, calcar was reconstructed with a cut autograft from the femoral neck.


All Patients were followed for 1 year and evaluated using Modified Harris Hip Score except one patient who expired in postoperative period probably due to embolism. He had poor cardiopulmonary reserve preoperatively and cement was used in this case. Average period of initiation of full weight bearing in the present study was 5.5 days. Excellent/good results were seen in 20 patients (80%). One of the patients, who remained bedridden even after surgery, developed decubitus ulcer on the back and was labeled as failure. This patient expired 9 months after surgery. Dislocation of prosthesis was not seen in any case. One patient had shortening more than 1.5 cm due to sinking of prosthesis.


Although majority of patients with intertrochanteric fractures can be successfully managed with osteosynthesis, older patients with severe osteoporosis and associated comorbidity may benefit from prosthetic replacement. However, large scale studies are required to prove it conclusively.


HemiarthroplastyIntertrochanteric fracturesOsteoporosis


Intertrochanteric fractures constitute almost 50% of proximal femoral fractures. Osteosynthesis is possible with satisfactory outcome in majority of patients.

Extramedullary implants like DHS, DCS and Blade Plate are common options, but vastly used implant is DHS with its inherent advantage of controlled collapse. DHS superseded earlier devices such as McLaughlin Pin and Plate and the Jewitt nail plate, but failure of fixation still occurs in up to 20% of cases. Cutting out of implant from the femoral head continues to be an important cause of mechanical failure with these implants, though its incidence is determined by factors like fracture subtype, quality of reduction, implant position and bone density [1].

Dynamic Hip Screw has higher complication rate in patients with unstable and comminuted fractures, associated severe osteoporosis, and associated comorbidity. Comminution along calcar and posterior cortex leads to varus malpositioning of the fracture fragments.

An intramedullary device (gamma nail) has some theoretical advantage over DHS and its predecessors. It is not dependent on screw fixation of a plate to the lateral cortex, which can be a problem in very osteoporotic bone. However, these devices have also been associated with increased risk of intraoperative and postoperative femoral shaft fractures [24].

When instability is a potential problem, the patient must remember and adhere to a prolonged postoperative regimen of limited weight bearing until union of the fracture. However, many elderly patients who sustain hip fractures are not capable of complying with weight-bearing restrictions during postoperative ambulation. Our primary concern is to try and perform an operation on patient that will give him greatest opportunity for early mobilization [5, 6].

Primary prosthetic arthroplasty offers great opportunity to mobilize these patients rapidly, thus preventing complications such as pneumonia, deep vein thrombosis, pulmonary embolism, pressure sores, etc. Arthroplasty per se is absolved of the risk of both fracture site nonunion and avascular necrosis [79]. Additionally, a hip prosthesis allows early full weight bearing, and thereby expediates the patient’s return to a preinjury functional level.

Failed treatment of intertrochanteric fractures, such as nonunion, posttraumatic osteoarthrosis, perforation of acetabulum by an internal fixation device and rarely avascular necrosis of femoral head, leads to profound functional disability and pain. In such cases, attempt to preserve host femoral head is preferred in young patients, but salvage treatment with hip arthroplasty should be increasingly considered for selected older patients with already poor bone quality, bone loss or articular cartilage damage [10].

Materials and methods

The present study was conducted on 25 elderly osteoporotic patients of intertrochanteric fractures treated in tertiary care center between 2006 and 2009 with hemiarthroplasty. Young patients with stable fractures and patients with active infection were excluded from the study.

The age of the patients in the present study was in the range of 65–95 years with an average age of 77.8 years. Out of these, 10 patients were in seventh and eighth decade each, three patients were in their ninth decade and two belonged to sixth decade. There was a preponderance of females patients (72%) in the present study. Simple fall was the most common mode of trauma accounting for all (100%) of the cases. The majority of patients in this study were osteoporotic with 72% patients falling in Singh’s grade III (definite osteoporosis), II (moderately advanced osteoporosis) and I (severe osteoporosis), while 28% were having borderline osteoporosis that is Singh’s index grade IV.

The ambulatory status was assessed prior to trauma based on whether or not patient required support for walking and if required then what support patient required. Based on this, patient was given points as in modified Harris Hip Score. The presurgery ambulatory status of patients is shown in Table 1. Majority of the patients (72%) in the present study had associated comorbidities as shown in Table 2.
Table 1

Previous ambulatory status

Support used


No. of patients






Cane for long walk




Cane most of time




One crutch


Two canes




Two crutches




Not able to walk


Table 2

Associated comorbidities

Associated illness

Number of patients


Cardiological insufficiency



Pulmonary insufficiency















Operative technique

Posterolateral approach was used for exposure. Larger incision was required in some cases for reattachment of lesser trochanter and greater trochanter with stainless steel wires. After capsular exposure, our technique varied depending on the nature of fracture:
  • If calcar was deficient, femoral neck osteotomy done with saw and calcar was reconstructed with a cut autograft from the femoral neck. Femoral head with remaining part of neck retrieved out. A circular graft was fashioned from the neck bone fragment to fit around the broach and replace the calcar (Fig. 1). The calcar autograft is compressed between collar of the femoral stem and proximal femur as the stem is fully inserted.
    Fig. 1

    a Intertrochantric fracture. Level of neck osteotomy marked with arrow. The horizontally shaded part to be used as autograft to construct calcar. b The stem is inserted with autograft in place. The greater and lesser trochanter reattached with stainless steel wires which are tightened after cementing

If calcar was intact: femoral head dislocated posteriorly and retrieved out. Prosthesis was inserted directly after femoral canal preparation in these cases. An osteopenic femur had a wide medullary canal, making excessive reaming unnecessary.
  • If lesser trochanter was fractured, it was reattached in its anatomical position with steel wire.

  • If greater trochanter was fractured, another steel wire was placed around proximal femur.

These steel wires were tightened once the femoral component was seated in place.

The center of prosthesis was taken at the tip of the trochanter. In cases where greater trochanter was fractured, length determination was done by repositioning the fractured greater trochanter anatomically. If the femoral head was above the trochanter’s apex, the limb would be too long and if it was lower, it would be too short. Anteversion and retroversion of prosthesis was assessed according to lateral condyle of femur. Cases in whom cement was not used, grafts prepared from femoral head were inserted to get a proper fit of stem of prosthesis.

Patients were given perioperative antibiotic cover for at least 72 h. Patients were advised to do quadriceps exercises and active movements of the limb so as to improve circulation and decrease edema. Patients were mobilized with full weight bearing with support of walker as soon as patient’s condition permitted starting from 2nd postoperative day. Average period of initiation of full weight bearing in the present study was 5.5 days. Stitches were removed between 12 and 16 days as indicated.

At each follow-up, functional evaluation of the patient was done to note the range of movements at hip and knee, limb length, wound condition and any other complaint. The occurrence of postoperative complications like wound infection, stiffness, swelling, limb length discrepancy, dislocation and sinking of prosthesis was noted. Patients were subjected to radiological evaluation at regular intervals to ascertain implant position, see any sign of implant loosening and new bone formation. Results were assessed using Modified Harris Hip Score.

Final results

Radiologically fractures were classified according to AO classification of fractures. In the present study, 14 cases were of stable (A1) type and 11 were unstable. Calcar was intact in all stable fractures (14 cases) and had to be reconstructed in six of unstable fracture cases.

Hemiarthroplasty with bipolar prosthesis was done in all patients. Neck was preserved in 6 of patients to reconstruct calcar. In 3 of patients, encirclage wires were used to reattach lesser and greater trochanters.

Cement was used only in three patients on account of high risk of embolism in elderly patients while pressurizing cement. Instead, bone grafts taken from retrieved femoral head were inserted in medullary canal so that prosthesis stem does not remain loose in the canal. In 80% of total cases, we achieved excellent/good results as assessed by Harris Hip Score. In another 12%, we achieved fair results.

One of the patients (95 years old) in whom cement was used expired 13 h after surgery. It was observed that cement should be used with extreme caution in these elderly patients with comorbidities. Sinking of prosthesis occurred in two patients leading to more than 1.5 cm shortening in one of the cases. Both these cases were of the unstable type. Results depending on the stability of fracture are given in Table 3.
Table 3

Results based on the stability of fracture

Type of fracture

No. of patients

Harris Hip Score >80




13 patients






Seventy-two percentage of patients complained of no pain, while 20% of patients complained of slight pain. Limp was unnoticeable in 56% of patients of this study while severe limp was seen in one patient. Twenty-two patients (88%) were successfully ambulated with or without support in the present study. Fifteen (60%) patients had no increase in dependency on walking aid as compared to prefracture status, while 7 patients (28%) had increased dependency on walking aid after surgery. One patient could come out of bed with support of two persons and one patient remained completely bedridden.

Intraoperative cardiorespiratory disturbances were observed in 2 patients in this study in whom cemented arthroplasty was done out of which one patient (case 7) expired 13 h after surgery. The other patient (case 9) remained well during postoperative period and entire follow-up. This death was presumed to be due to pulmonary embolism. No case of postoperative infection, dislocation of prosthesis and clinically evident deep vein thrombosis was seen in this study. One of the patients (case 10) who remained bedridden even after surgery developed decubitus ulcer on the back. This patient expired 9 months after surgery.


Intertrochanteric fractures in elderly osteopenic patients, especially those who cannot follow limited weight-bearing instructions, continue to vex orthopedic surgeons. The best treatment for intertrochanteric fractures in such patients remains controversial. Many surgeons have recommended that the hip be protected throughout the healing period in patients who have intertrochanteric or subtrochanteric fractures that have major comminution, osteoporosis or poor fixation of screws. Though fracture healing and functional restoration are important but are secondary to the overall welfare of the patient. This summarizes our philosophy in the treatment of intertrochanteric fractures in elderly patients with arthroplasty that allows immediate mobilization. Although arthroplasty in unstable fractures presents greater challenge to surgeon and may require calcar reconstruction or reattachment of fragments using encirclage wires nonunion, malunion, metal breakage or impingement are of no concern with the prosthesis [1114].

Hemiarthroplasty has been well used in the treatment of intertrochanteric fractures since 1970s though its initial use began as a salvage procedure for fractures that went into nonunion. In 1974, Tronzo [15] reported on the use of long-stem Matchett Brown endoprosthesis for intertrochanteric fractures. In 1977, Stern MB reported 86% good results in their series of 29 patients of intertrochanteric fractures treated with Leinbach prosthesis [6]. Pho et al. [7] also described their good results with Thompson prosthesis and emphasized on proper building up of calcar height with cement. There are also reports on calcar replacement hemiarthroplasties for these fractures, but implantation of calcar replacing prosthesis usually requires removal of the fractured metaphyseal fragments that produce a gap between greater trochanter and proximal femur so that trochanteric nonunion with associated abductor weakness is a common complication [16].

The mean age of patients in our series was 77.8 years. The high incidence of intertrochanteric fractures in elderly population reflects high incidence of osteoporosis in these patients. Most of the patients in the present study were osteoporotic with 72% falling in grade III, II and I of Singh’s Index of osteoporosis. As a result, simple fall was the only mode of trauma of all the elderly patients in our study. In 2000, Chan and Gurdev [17] described a series of 55 hemiarthroplasties in intertrochanteric fractures in patients with mean age of 84.2 years. Similarly, Grimsurd et al. [5] reported average age of patients in their study as 80 years though no record of degree of osteoporosis was available in these studies.

The mean operating time in the present series of 47.82 is slightly lesser than Chan (69 min) and Leinbach (80 min) but only slightly longer than 41 min reported for hip pinning under fluoroscopy [18]. In our series, we used cement in only four cases which could have saved some time. Haentgens et al. [16] reported that in their institution the operative times for bipolar arthroplasty and internal fixation for intertrochanteric fractures are not significantly different.

Comorbidity in the aged patients further complicates the management of these fractures. The comorbid illnesses did delay the surgical fitness of patients, thereby resulting in some delay in surgery. Despite this, only one of the patients with preexisting pulmonary disease, cardiac insufficiency, renal insufficiency and anemia expired in postoperative period. This patient also succumbed to intraoperative cardiorespiratory disturbances. Grafts taken from head are a good alternative to cement in such high risk patients. The in-hospital mortality rate of 4% in the present study is slightly less than 7.3 as reported by Chan and Gurdev [17]. Geiger and Heizel [19] reported that four or more comorbidities increased the risk to die by 78%. Intramedullary reaming and cementation associated with arthroplasty theoretically increase the risk of cardiopulmonary complications after hip fracture. In study by Dobbs et al. [20], intraoperative cardiopulmonary instability was noted for almost all the arthroplasty patients who succumbed to death within 30 days of surgery.

In this study, the standard stem length was used. In follow-up, there were no failures related to inadequate stem length and no case of periprosthetic fracture occured in our series. We did not find it necessary to use long-stem prosthesis. Chan and Gurdev [17] also used the standard stem length ranging from 120 to 140 mm.

Patients were trained to begin walking with walker earlier (average 5.5 days) in this study so as to reduce postoperative complications of prolonged recumbency like pneumonia, bed sore, DVT, etc. Leinbach in their study in 70 s described 25 of their 29 patients ambulating with walker in 5–7 days [6]. Eighty-eight percentages of patients in this study were successfully ambulated after surgery. Sixty percentages of patients maintained their prefracture ambulatory status, while 28% had increased dependency on walking aids but were successfully mobile. These results are comparable to previous studies as shown in Table 4.
Table 4

Comparison of results with other studies


Percentage of patients successfully ambulated

Percentage of patients with same prefracture ambulatory status

Chan and Gurdev [17]



Pho et al. [7]


Stern and Goldstein [6]


Stuart Green et al.


Present study



The patients who were ambulatory at discharge gradually improved over follow-up period and were able to transition from walker to cane and few patients without support. Two patients who were dependent at discharge did not improve and were bedridden even later. One of these patients was debilitated and too ill for walking and was limited to bed and chair prior to fracture also. He expired 9 months after surgery though death was unrelated to surgery and was due to myocardial infarction. Age, gender, prefracture health status and social dependency before fracture are important factors determining functional recovery after surgery. Other patient never followed any physiotherapy advices and showed up after 1 year with an attack of stroke and continues to remain bedridden. This indicates the importance of following of strict physiotherapy regime for good outcome of surgery in patients. Majority of patients gained good range of motion with physiotherapy.

There was no significant incidence of general complications in the present study. There was only one case of bed sore whom we could not mobilize. There was no case of pneumonia and deep vein thrombosis. This is on account of the fact that most of the patients had very short period of recumbency following surgery. Our aim was not range of motion, but basically it was ambulation in a painless hip. The greater diameter of the prosthetic head in bipolar arthroplasty might explain the decreased tendency to dislocate. Besides, all patients in our study were strictly advised not to squat and sit cross-legged. In all patients, preservation and closure of hip capsule as well as reattachment of external rotators was done meticulously.

In the present study at final follow-up, 72 percentages of patients had no pain and 24% had slight pain. One of the patients had moderate pain up to 3 months after surgery. In this patient, greater trochanter was fractured and was reattached with encirclage wire. The pain disappeared with union of greater trochanter. In 24% of patients who continued to have slight pain, no sign of infection was found and removal of prosthesis was not required so far. This slight occasional pain may be explained by loosening of stem of prosthesis in canal as the prosthesis was not cemented in these cases. In study by Green et al., 4 out of 20 patients were left with hip pain on ambulation and of these two had greater trochanter nonunion [21]. Stern and Goldstein reported one patient with painful prosthesis every time he walked because tip of the stem was rubbing against femoral cortex in bowed femur. They also reported another patient with pain subsequent to acetabular erosion [13, 16].

In the present study, sinking of prosthesis was seen in two of the patients leading to shortening of limb. One of these patients developed equinus deformity subsequently to compensate for shortening and further developed flexion deformity at hip. Limb length discrepancy was minimal (<5 mm) in 64% of cases in the present study and 28% had limb lengthening between 5 and 10 mm. Some may say that one set of problems associated with internal fixation (loss of fixation, hardware cut out) are being traded with another set of problems in arthroplasty (limb lengthy discrepancy), but if center of prosthesis is taken at level of greater trochanter tip then appropriate limb length can be maintained [13, 22]. However, sinking of prosthesis can be problem if cement is not used with calcar reconstruction as happened in two of our cases.

Excellent or good results were observed in 80% of the patients in this study. Stern and Goldstein reported excellent results in 94% of patients, while Chan and Gurdev reported good results in 83% of cases. Given the few complications and comparable functional recovery in this small study, we think that using standard arthroplasty is a reasonable alternative to sliding hip screw in intertrochanteric fractures. We would like to emphasize the careful selection of cases for this technique. The potential advantage of hemiarthroplasties for the treatment of intertrochanteric fractures warrants additional larger studies to be compared with a matched control group treated with sliding hip screw [17, 23].


No funds were received in support of this study. No benefits in any form have been or will be received from a commercial party related directly or indirectly to the subject of this manuscript.

Supplementary material

590_2011_870_MOESM1_ESM.jpg (4.1 mb)
Supplementary material 1 (JPEG 4216 kb)
590_2011_870_MOESM2_ESM.jpg (4.3 mb)
Supplementary material 2 (JPEG 4428 kb)
590_2011_870_MOESM3_ESM.jpg (3.9 mb)
Supplementary material 3 (JPEG 4013 kb)
590_2011_870_MOESM4_ESM.jpg (4.1 mb)
Supplementary material 4 (JPEG 4234 kb)
590_2011_870_MOESM5_ESM.jpg (685 kb)
Supplementary material 5 (JPEG 685 kb)
590_2011_870_MOESM6_ESM.jpg (681 kb)
Supplementary material 6 (JPEG 680 kb)

Copyright information

© Springer-Verlag 2011