Intramedullary fixation of intertrochanteric fractures has gained increased interest. Such a fixation is significantly stronger and more rigid than that with an extramedullary screw plate device. 7 Nevertheless, increased stresses at the distal tip of the nail frequently are reported. These may appear as local cortical hypertrophy, midthigh pain, 9,16 or fractures of the femoral shaft around the bend in the nail or around the distal locking screws. 2,4–6,8,10,11,16,20,24
In the current authors’ experience with the Intramedullary Hip Screw (IMHS, Smith and Nephew, Memphis, TN), the prevalence of cortical hypertrophy is approximately 20% and is correlated with the fracture pattern (lack of medial buttress), the use of large diameter nails (14 and 16 mm), and with two locking screws. 12 The stresses of weightbearing are transferred from the fracture to the interlocking screws. The stressed screws, in turn, stress the surrounding bone. The use of large diameter nails and the use of two locking screws increases the stiffness of the fixation at the distal end of the nail, with a subsequent increase of the gradient of stresses between the supported and the unsupported part of the femur.
Because the current authors did not want to abandon the use of locking screws, the current study was designed to explore another mode of locking.
MATERIALS AND METHODS
Eighty-eight consecutive patients were considered for inclusion in this trial. All patients had a fracture featuring loss of the medial buttress (Type IV and Type V of the classification of Jensen and Michaelsen 14) or a reversed oblique fracture. Three patients were not enrolled in the study because they were unable to complete the entire followup (one was living abroad and two were homeless). Four patients refused to be enrolled. The study was approved by the local ethics committee and written consent was obtained from the patients or their families.
The remaining 81 patients were allocated randomly into two groups for treatment with the Intramedullary Hip Screw statically locked with two screws (39 patients, Group A) or dynamically locked with one screw passing through a slot at the distal end of the nail (42 patients, Group B) (Fig 1). In one patient in Group B, a dynamically locked nail was used but erroneously locked with two screws. This patient subsequently was not included in the trial, leaving 39 patients (26 women and 13 men) in Group A and 41 patients (24 women and 17 men) in Group B. The average age at time of operation was 75.8 years (standard deviation, 11.8) in Group A and 78.4 years (standard deviation, 13.05) in Group B.
An additional round hole was added below the slot of the dynamically locked nail to allow the surgeon to lock the nail if the screw that has to be placed in the slot failed, but this never occurred. All of the nails had a 4° valgus angle and a 135° angle between the lag screw and the nail. The diameter of all the nails was 12 mm. The locking screws were inserted with the appropriate jig. When the nail jammed, the medullary cavity was overreamed 2 mm using power reamers.
The preoperative variables included the patient’s age, gender, the mobility score of Parker and Palmer, 18 Jensen’s autonomy index, 13 mental status assessment with the Qureshi and Hodkinson test, 19 and the American Society of Anesthesiologists rating of operative risk. 1
The intraoperative variables included type of anesthesia, surgical time, experience of the surgeon, estimated blood loss, and intraoperative complications. Except for the osteosynthesis device, all other parameters of the therapeutic regimen were similar. Antibiotics (2 g cefazolin given at induction of anesthesia) were administered to all the patients, and prophylactic doses of low molecular weight heparin (calcium nadroparin, 7500 IU Anti-Xa, Sanofi-Winthrop, France) were given once a day for 15 days to patients who did not have a history of myocardial infarction or stroke in the previous 6 months, hemorrhage (genitourinary tract, gastrointestinal), tendency to bleed (thrombocytes < 150 × 106 / mL or prothrombin time < 65%), or known allergies to low molecular weight heparin. All patients were allowed to weightbear as tolerated on the injured extremity, with the goal to return to the functional status that existed before trauma. Transfusion requirements, ambulatory status at discharge, and place to which the patient was discharged (private domicile, geriatric facility, or rehabilitation center) were recorded for each patient.
The patients were followed up after 1 month, 3 months, 6 months, 1 year, 2 years, and 3 years. Ambulatory status as measured with the mobility score of Parker and Palmer, 18 independence on social welfare as determined by Jensen’s autonomy index, 13 use of ancillary devices, accommodation, and leg length discrepancy were assessed at each visit. Special attention was given to the presence of pain in the hip area and in the midthigh zone. Pain was assessed using a four-point scale: 0, no pain; 1, slight pain that did not affect the patient’s ability to walk; 2, moderate to severe pain, requiring administration of analgesics and reducing the patient’s ability to walk; and 3, untractable and constant pain. The examinations frequently were repeated and cross-matched with those of the family practitioner or of the doctor in charge of rehabilitation. The presence of pain was considered real if two successive examinations 1 month apart revealed the same pattern of pain. This restriction was important because of the poor mental status of many patients, which impaired the reliability of their verbal reactions.
Radiographs were taken intraoperatively, 2 days and 15 days after surgery, and at each followup. All of the radiographs were reviewed by the authors. The location of the lag screw in the head, the tip-apex distance as defined by Baumgaertner et al, 3 the length of the unengaged part of the lag screw, and the presence of cortical hypertrophy at the level of the tip of the nail were estimated on each set of radiographs. Attention was given to the presence of a fracture of the lateral cortex at the level of the introduction point of the sleeve. Telescoping of the fracture was determined by comparing the length of the unengaged root of the lag screw on two consecutive radiographs. Subsidence of the nail into the femur was calculated by the apparent migration of the locking screw in the slot (Fig 2). These last two measurements were corrected for the magnification factor.
The results are presented as the mean with standard deviation to the mean and range. Differences were evaluated for statistical significance using the Fisher-Irvin test and the chi square test for categoric data and the Student’s t test for interval data. A probability less than 0.05 was considered significant.
Randomization provided similar groups as shown by the division of treatment according to the preoperative variables (Table 1).
Most of the patients (35 in Group A and 37 in Group B) were operated on under spinal anesthesia (Table 2). External reduction was successful in all patients, except in two patients in Group A and one patient in Group B. In these three patients, the fracture site had to be approached and a hook had to be used for an irreducible flexion of the cervical fragment. The nail jammed in the medullary canal in 28 patients (17 in Group A, 11 in Group B; p = 0.116), requiring power reaming of the medullary cavity. The median length of the lag screw was 95 mm in each group. Estimated intraoperative and postoperative blood loss, operating time, and transfusion requirements were similar in both groups. The modes of reduction and the tip-apex distance were comparable in both groups (Table 2).
Few intraoperative complications were observed. The greater trochanter split in three patients during insertion of the nail, but healed uneventfully several months later. One locking screw was missed in one patient in Group A, leaving one statically locked nail with only one screw. One malreduction was seen in one patient in Group B, despite a perfect reduction obtained during installation (Fig 3).
Three patients in Group A and no patients in Group B were unable to walk after surgery. Two patients in Group A and three patients in Group B died during their hospital stay. Ten patients in Group A and eight in Group B were discharged to their homes, whereas the other patients were transferred to a rehabilitation center (15 patients in Group A and 13 patients in Group B) or their previous nursing care facility (12 patients in Group A and 13 patients in Group B).
The mortality was high in both groups, but did not differ significantly between them. At 3 months, five patients in Group A and four patients in Group B had died. At 6 months, six patients in Group A and seven patients in Group B had died. After 1 year, nine patients in Group A and seven patients in Group B had died; and after 2 years, 12 patients in Group A and 11 patients in Group B had died.
Only the patients still alive at 1 year were considered for late assessment of tolerance to the implant. Among these patients (30 in Group A and 34 in Group B), one patient in Group B was lost to followup because she was incarcerated. The mobility score of Parker and Palmer, 18 accommodation, and Jensen’s autonomy index 13 were similar in both groups. Eight patients (Group A, six patients; Group B, two patients) had pain at the midthigh area. Of these patients, four (Group A) stated that this pain impaired their walking capacity. Among these eight patients with pain, six (Group A) had some degree of cortical hypertrophy at the level of the locking screws (Fig 4). In one additional patient (Group B) cortical hypertrophy was seen on radiographs. This patient did not have any pain in the midthigh area; whereas all the other patients in Group B did not have any sign of cortical hypertrophy (Fig 5). These seven femoral thickenings were visible only on anteroposterior (AP) radiographs, and not on the lateral views, even if the tip of the nail impinged against the anterior cortex. The breakdown of these seven patients between Group A (six of 30 patients) and Group B (one of 34 patients) was statistically significant (p = 0.029). The occurrence of a fracture of the lateral cortex at the level of the sleeve was correlated with these bone hypertrophies and also with the presence of midthigh pain. Three of the seven patients with cortical hypertrophy and three of the 56 patients without cortical hypertrophy had a sleeve unsupported by the lateral cortex (p = 0.001). These fractures of the lateral cortex were seen in the reverse oblique fractures and some of the Type V fractures with fracture of the greater trochanter at the level of its base. Of these seven patients, two required hardware removal (Group A).
The position of the lag screw in the femoral head was similar. Most of the lag screws were central (32 screws in Group A and 37 screws in Group B) or inferior (five screws in Group A and three screws in Group B). Two screws in Group A and one in Group B were erroneously placed in an upper position.
Lag screw sliding was similar in both groups. Subsidence of the nail in the femoral shaft was seen in nine patients in Group B (and none in Group A). For these nine patients, subsidence averaged 4.6 mm (standard deviation, 1.65) (range, 2–7 mm). As a result of these relative motions of the hardware, an average limb shortening of 6.3 mm (standard deviation, 5.07) was seen in Group A, versus 7.6 mm (standard deviation, 5.37) in Group B (p = 0.3985). These data were obtained from 26 patients in Group A and 21 patients in Group B who were able to do a radiographic measurement of the length of their limbs.
In Group B, one lag screw cut out and created a secondary fracture of the femoral head. Total hip arthroplasty was done as the revision surgery. The intertrochanteric fracture was healed at the time of this secondary intracapital fracture. One case of femoral head rotation around the lag screw also was seen in Group B. The lag screw was placed eccentrically during the operation (superiorly on the AP view and centrally on the lateral view). Fifteen days after loading, the lag screw appeared superior on the AP radiographs and posterior on the lateral views, without any signs of cutting out of the lag screw.
One late femoral shaft fracture occurred in a patient in Group A. This patient had midthigh pain for 4 months and presented with hypertrophy of the femoral cortex at the level of the two locking screws. This fracture occurred 16 months after the index procedure and the patient was treated successfully with an anterograde femoral interlocking nail.
The current study showed that the use of two static locking screws is correlated with a relatively high rate of cortical hypertrophy and that the use of a dynamically locked nail significantly reduces the prevalence of this complication. To the current authors’ knowledge, benefits of this feature have not been reported in any clinical study, despite the fact that several of these new generation short nails designed for fractures of the upper femur include a slot instead of two holes.
Distal cortical hypertrophy is the radiologic mark that proximal stress shielding exists, with load concentrations at the end region of the nail. As a bone callus is forming in the area of the proximal femur, proximal osteopenia never appears, contrary to what is found with femoral endoprostheses. Bone overloading at the level of the tip of the nail is involved in two main complications seen with intramedullary fixation of intertrochanteric fractures: midshaft fractures and midthigh pain. The current authors were able to correlate the prevalence of cortical hypertrophy with the lack of medial buttress, with use of large diameter nails and with the use of two locking screws. 12 These findings explain why the current authors have included only the patients with Types IV and V fractures using the classification of Jensen and Michaelsen, 14 and reversed oblique fractures. However, the diameter of the nail was similar in both groups.
These findings were consistent with various biomechanical investigations done with the Gamma nail. Rosenblum et al 21 loaded instrumented femurs from cadavers and showed that the load transfer to the calcar decreased with decreasing fracture stability and that the device, because of its inherent stiffness, did not impart physiologic loads to the proximal femur. Using similar static models, Shaw and Wilson 22 and Mahomed et al 17 found similar unloading of the proximal femur and overstress at the level of the distal tip of the nail. This stress concentration at the implant’s distal point of fixation is attributable to the presence of the locking screw(s) and to the 10° valgus angle of the Gamma nail, responsible for a three-point fixation in the proximal femur. Using the Intramedullary Hip Screw instead of Gamma nails, Kummer et al 15 found a minimal decrease of the medial strains when fixing a stable osteotomy (75% of the strains measured for an intact femur), whereas the decrease in medial strains was considerably higher (13% of the strains measured for an intact femur) when fixing an unstable osteotomy. Controversy exists regarding the role of the locking screws. Sim et al, 23 with a finite element analysis of stress distribution reported that the stresses measured in the proximal femur were smaller when the implant was locked than when it was freely removable. Rosenblum et al 21 did not observe any significant effect on load distribution along the proximal femur when locking screws were added to their construct. In a prospective series of Gamma nails, Forthomme et al 9 observed that six of 92 patients had midshaft pain. Five of these six patients had a nail locked with two screws and the authors concluded a possible negative effect of these locking screws. However, those experiments were done with the Gamma nail, the shape of which is not similar to the Intramedullary Hip Screw. Additional investigations with the Intramedullary Hip Screw therefore are needed.
Intramedullary Hip Screws are a valuable method for fixing intertrochanteric three-part fractures with separation of the medial cortex (Type IV), four-part fractures (Type V), and reversed oblique fractures.
The aforementioned described advantages (minimal limb shortening, absence of nonunions), are seen in the two groups of the current trial. The low rate of late femoral shaft fractures (one of 80, 1.25%), makes this device attractive among the other existing intramedullary implants. Use of a slotted nail allows one to reduce the problems of late intolerance to the implant to an acceptable rate and renders this device recommendable for treatment of intertrochanteric fractures.
The authors thank Phillippe Delincé MD for authorization to do this protocol, the staff assistants for help in reviewing the patients, and Paul Smets MD, of the Pacheco Institute of Rehabilitation, for encouragement and valuable help in supervising the followup of many patients.
1. American Society of Anesthesiologists: New classification of physical status. Anesthesiology 24:111, 1963.
2. Aune AK, Ekeland A, Ødegaard B, Grøgaard B, Alho A: Gamma nail versus compression screw for trochanteric femoral fractures: 15 reoperations in a prospective, randomized study of 378 patients. Acta Orthop Scand 65:127–130, 1994.
3. Baumgaertner MR, Curtin SL, Lindskog DM, Keggi JM: The value of the tip-apex distance in predicting failure of fixation of peritrochanteric fractures of the hip. J Bone Joint Surg 77A:1058–1064, 1995.
4. Benum P, Grøntvedt T, Bråten M, et al: Gamma nail versus CHS in intertrochanteric and subtrochanteric femoral fractures: A preliminary report of a prospective randomized study. Acta Orthop Scand 63(Suppl 247):7–8, 1992.
5. Bridle SH, Patel AD, Bircher M, Calvert PT: Fixation of intertrochanteric fractures of the femur: A randomised prospective comparison of the Gamma nail and the dynamic hip screw. J Bone Joint Surg 73B:330–334, 1991.
6. Butt MS, Krikler SJ, Nafie S, Ali MS: Comparison of dynamic hip screw and Gamma nail: A prospective, randomized, controlled trial. Injury 26:615–618, 1995.
7. Curtis MJ, Jinnah RH, Wilson V, Cunningham BW: Proximal femoral fractures: A biomechanical study to compare intramedullary and extramedullary fixation. Injury 25:99–104, 1994.
8. Fornander P, Thorngren KG, Törnqvist H, Ahrengart L, Lindgren U: Swedish experience of the first 209 randomised patients with Gamma nail vs. Screw-plate. Acta Orthop Scand 63(Suppl 248):90, 1992.
9. Forthomme JP, Costenoble V, Soete P, Docquier J: Traitement des fractures trochantériennes du fémur par le clou Gamma (A propos d’une série de 92 cas). Acta Orthop Belg 59:22–29, 1993.
10. Goldhagen PR, O’Connor DR, Schwarze D, Schwartz E: A prospective comparative study of the compression hip screw and the Gamma nail. J Orthop Trauma 8:367–372, 1994.
11. Guyer P, Landolt M, Eberle C, Keller H: Der Gamma-Nagel als belastungsstabile Alternative zur DHS bei der instabilen proximalen Femurfraktur des alten Menschen. Helv Chir Acta 58:697–703, 1992.
12. Hardy DCR, Descamps PY, Krallis P, et al: Use of an intramedullary hip-screw compared with a compression hip-screw with a plate for intertrochanteric femoral fractures: A prospective, randomized study of one hundred patients. J Bone Joint Surg 80A:618–630, 1998.
13. Jensen JS: Determining factors for the mortality following hip fractures. Injury 15:411–414, 1984.
14. Jensen JS, Michaelsen M: Trochanteric femoral fractures treated with McLaughlin osteosynthesis. Acta Orthop Scand 46:795–803, 1975.
15. Kummer FJ, Olsson O, Pearlman CA, et al: Intramedullary versus extramedullary fixation of subtrochanteric fractures: A biomechanical study. Acta Orthop Scand 69:580–584, 1998.
16. Leung KS, So WS, Shen WY, Hui PW: Gamma nails and dynamic hip screws for peritrochanteric fractures: A randomised prospective study in elderly patients. J Bone Joint Surg 74B:345–351, 1992.
17. Mahomed N, Harrington I, Kellam J, et al: Biomechanical analysis of the Gamma nail and sliding hip screw. Clin Orthop 304:280–288, 1994.
18. Parker MJ, Palmer CR: A new mobility score for predicting mortality after hip fracture. J Bone Joint Surg 75B:797–798, 1993.
19. Qureshi KN, Hodkinson HM: Evaluation of a ten-question mental test in the institutionalized elderly. Age Ageing 3:152–157, 1974.
20. Radford PJ, Needoff M, Webb JK: A prospective randomised comparison of the dynamic hip screw and the gamma locking nail. J Bone Joint Surg 75B:789–793, 1993.
21. Rosenblum SF, Zuckerman JD, Kummer FJ, Tam BS: A biomechanical evaluation of the Gamma nail. J Bone Joint Surg 74B:352–357, 1992.
22. Shaw JA, Wilson S: Internal fixation of proximal femur fractures: A biomechanical comparison of the Gamma locking nail and the Omega Compression hip screw. Orthop Rev 22:61–68, 1993.
23. Sim E, Freimüller W, Reiter TJ: Finite element analysis of the stress distributions in the proximal end of the femur after stabilization of a pertrochanteric model fracture: A comparison of two implants. Injury 26:445–449, 1995.
24. Valverde JA, Alonso MG, Porro JG, et al: Use of the Gamma nail in the treatment of fractures of the proximal femur. Clin Orthop 350:56–61, 1998.