84
Injury (1990) 21,84-88 Prinfedin Great Britain
Treatment of unstable intertrochanteric fractures of the femur: a prospective trial comparing anatomical reduction and valgus osteotomy D. W. Clark1and W. J. Ribbans IWexham Park Hospital, Slough, UK “Central Middlesex Hospital, London, UK
A series of 100 consecufivepatients wifh u&able inferfrochunftic fracfures were treated by compressionhip screw#ofion; 55 patientshad an anatomical reducfion(Group 1)and 45 patientsa Sarmienfoosfeofomy and valgusreducfion(Group 2). Group 7 spent an averageof 10 days kxs in hospital than Group 2 (21 days compared with 31 days) (I’< 0.02). They alsohad a greater chance of refurning to their pre-injury accommodafionand of achieving theirpre-injury walking capabilify. Radiologicalfailure of fracfure fixation, with varus angulofion of the femoral head by cuffing out of thescrew,was seen seven timesin Group I but only once in Group 2. Anafomical reducfion provides better clinical results than valgus osfeofomy in the patient with an unstable inferfrochanfericfiacfure stabilizedby a compressionhip screw. 7he capacityfor failure of fracfure f&ion
is greater, however, in the former. Valgus osfeofomyprovia%sa
simple means of securing a stablereducfionof thefracfurewhich cannotbe satisfactorilyreduced by closedmeans.
Introduction Fractures of the hip are a major cause of morbidity and mortality in the elderly (OPCS, 1987). They represent an increasingly severe drain upon the resources of orthopaedic and geriatric departments. The stable intertrochanteric fracture can be successfully managed by closed reduction and internal fixation with a variety of implants. If there is medial and posterior comminution, it will lead to instability which will, on loading, progress to varus angulation and retroversion leading to failure of the fixation. Medial displacement osteotomy (Dimon and Hughston, 1967) and valgus osteotomy (Sarmiento and Williams, 1970) were introduced to produce bony stability and prevent this collapse. Compression hip screws allow collapse of the bony fragments in a controlled fashion, thus maintaining bony stability. The results of compression hip screw fixation after anatomical reduction have been good (Heyse-Moore et al., 1983; Rao et al., x983), but clinical and radiological failures still occur. This paper describes a prospective trial contrasting the results of anatomical reduction with valgus osteotomy, each followed by fixation with a compression hip screw. The 0 1990Butterworth & Co (Publishers) Ltd OOZO-1383/90/020084-05
objective was to establish whether stabilization of the fracture by valgus osteotomy would improve the patients’ rehabilitation by reducing the failures associated with anatomical reduction and compression.
Patientsand methods A series of 100 consecutive patients with type 1 unstable intertrochanteric fractures of the femur (Evans, 1949) were admitted to Wexham Park and Heatherwood Hospitals between January 1985 and June 1987 (Figure 7). An assessment was made of the patient’s pre-injury walking ability and accommodation. There were 86 women and 14 men with an average age of 83 years (range 5 l-100 years). Of these, 44 patients had a fracture of the right femur and 56 of the left femur. Eighty-five patients were operated on within 2 days of admission. The longest interval was 25 days in a patient with severe hyponatraemia. The patients were allocated by their year of birth to two treatment groups: odd years to anatomical reduction and compression hip screw fixation (Group l), and even years to valgus osteotomy and compression hip screw fixation (Group 2). There were 55 patients in Group 1 and 45 in Group 2. All the operations were performed by Registrars or Senior Registrars. Operative technique Group 1 - Closed anatomical reduction (Figure 2). The patient was positioned supine on the operating table and reduction of the fracture achieved under the image intensifier by traction and internal rotation of the leg. Internal fixation was performed with a compression hip screw and a 135”, 4 hole plate. Group 2 - Valgus osteotomy (Figrrre 3). The patient was positioned supine on the operating table with the injured leg in 10” of internal rotation. The greater trochanter and the upper third of the femoral shaft were exposed through a lateral incision. An osteotomy was performed at 30-45” to the transverse plane, running obliquely downwards and
Clark and Ribbans: Treatment of unstable intertrochanteric
Figure 1. Anteroposterior radiograph unstable intertrochanteric fracture.
of an Evans
Type
fractures of the femur
I
85
Figure 2. Anteroposterior reduction and compression
radiograph following hip screw fixation
b
a
Figure 3. a, Anteroposterior radiograph following valgus osteotomy and compression in the performance of a valgus osteotomy (modified from Sarmiento et al., 1970).
medially from inferior aspect
the flare of the greater trochanter to the of the lesser trochanteric fragment. The
greater trochanteric fragments were retracted cephalad exposing the base of the head and neck fragment. Under image intensifier control a short-thread compression screw was inserted into the centre of the neck so that its tip came to lie in the head 5 mm from the articular cartilage. A slot was cut in the lateral cortex of the distal fragment to accommo-
anatomical
hip screw fixation, b, c, Diagrams of principal stages
date the barrel of the plate. A 135”, 4 hole plate was secured to the screw and the osteotomy reduced. The intensifier was used to check the reduction and then the plate was secured to the femoral shaft. All wounds were closed in a standard fashion with a suction drain. Prophylactic antibiotics were given for the first 48 h.
The patients
were allowed
full weight
bearing
with a
86
Injury: the British Journal of Accident Surgery (1990) Vol. 21/No. 2
Table I. Patient accommodation before injury and at discharge (Modified from Heyse-Moore et al., 1983)
Grade
Original accommodation
Final accommodation
A B C D E
Home Not own home Home Not own home Any
Home Same as pre-injury Home Same as pre-injury Long-term geriatric care
N.B. -These
3
weeks weeks weeks weeks -
Anatomical group
Osteotomy group
45% 23% 7% 14% 11%
20% 22% 27% 12% 19%
figures are based upon the surviving 85 patients at the time of hospital discharge.
Table II. Level of mobility. Changes in mobility
following
surgery Level Level Level Level
Hospital stay
1 : Non-walker 2 : Supervised walker 3 : Independent-with walking aid 4 : Independent - no walking aid
Change in mobility at 3 months after surgery Same level of mobility Reduced by 1 level Reduced by 2 levels Reduced by 3 levels
Anatomical group
Valgus osteotomy group
60% 31% 9% 0%
49% 39% 9% 3%
48 h after the operation. They were discharged as soon as they were capable of coping in their pre-injury accommodation or alternative accommodation suitable to their new functional level could be found.
frame
Follow-up. All patients were reviewed clinically and radiologically at both 6 and 12 weeks after operation. The patient was discharged at 3 months if the clinical and radiological findings demonstrated sound fracture fixation and bony union. Further follow-up of patients deemed failures on clinical and/or radiological grounds has now been extended to an average of 28 months (range 12-42 months). Three patients returned to their distant homes and were lost to follow-up.
Results The distribution of age, sex and side of fracture for the two groups was Similar. Operating time and peroperative blood loss The average operating time in Group I was 53 minutes ( f 17 minutes) and in Group 2 56 minutes ( f IS minutes). The mean peroperative blood loss was 289 cc ( f I I 7 cc) in Group I and in Group 2 347 cc ( f 204 cc). Neither result reached a statistically significant level. Postoperative radiological assessment The fracture/osteotomy reduction was assessed on the first postoperative radiograph. Satisfactory reduction was considered to have been achieved if the gap between the major fragments was less than 5 mm (Jensen et al., 1978). Of the Group I patients, 23 (42 per cent) and of the Group 2 patients, 14 (31 per cent) had satisfactory reductions in both anteroposterior and lateral planes. Complications and mortality In Group I,24 (44 per cent) patients and in Group 2,19 (42
per cent) patients suffered from one or more postoperative complication, e.g. respiratory, urinary, or thromboembolic. Eleven patients in Group 1 and four in Group 2 died in hospital. A further five patients in each group died before their 3 month follow-up. This resulted in an overall 26 per cent mortality at 3 months. Hospital stay and discharge accommodation (X&k r) The average inpatient stay was 21 days ( f 17 days) in Group I and 31 days ( f 22 days) in Group 2. This was statistically significant (PC 0.02 using Fischer’s test). Patient discharges were studied using a modification of the method employed by Heyse-Moore et al. (1983). Of Group 1 patients, 89 per cent and of Group 2 patients, 81 per cent who survived returned to their previous accommodation. However, 77 per cent of Group 1 patients and only 52 per cent of Group 2 patients achieved this within 3 weeks. Walking ability (TableIr) This was assessed at 3 months and compared with pre-injury status. Of Group 1 patients, 21(58 per cent) and of Group 2 patients, 16 (48 per cent) alive had achieved their preoperative level. Clinical review (Figuxs 4 and 5) At 3 months, eight patients complained of hip pain on walking. In six cases, the radiographs were satisfactory and the pain resolved spontaneously. In two patients, the radiographs revealed varus angulation of the femoral head with cutting out of the screw. In one of these (Group I), conversion to a total hip replacement was required to alleviate pain. The other patient (Group 2) experienced spontaneous pain resolution when the fracture united. One deep infection occurred in a Group 1 patient and neccessitated the removal of the implant once bony union had occurred. Twelve patients had shortening of the affected leg greater four in Group I and eight in Group 2. Four thanlcmpatients had external rotation deformity of greater than 10” - one in Group 1 and three in Group 2. Radiological review No patient developed a non-union of their fracture. Sliding of the compression screw was commonly observed. It exceeded 10 mm in nine Group I patients and in seven Group 2 patients by 3 months. Using Jensen’s criteria for failure of stabilization Uensen et al., l978), varus angulation of greater than 10” of the femoral head was seen in seven (13 per cent) Group I patients but only once (2%) in Group 2 patients. There was no instance of mechanical failure of the implant or axial penetration of the femoral head by the implant. One of the Group I patients required revision to a total hip replacement for persistent pain.
Clark and Ribbans: Treatment of unstable intertrochanteric
fractures of the femur
87
Figure 4. Anteroposterior
radiograph demonstrating implant and fracture fragment position immediately following surgery (Group I patient).
Discussion The aim of surgery for intertrochanteric fractures is to return the patient to his or her pre-injury functional level as quickly as possible. Stable reduction of the fracture, adequate internal fixation, minimal anaesthetic time and blood loss, and early mobilization all contribute to this aim. Anatomical reduction and compression hip screw fixation produce excellent results in stable fractures, but failures still occur in the unstable fracture. Heyse-Moore et al. (1983) reported a 14 per cent radiological and an 11 per cent clinical failure rate in these fractures. Harrington and Johnson (‘1973) reported improved radiographic results and fewer reoperations with the use of a medial displacement osteotomy and compression hip screw fixation, although they made no direct comparison with anatomical reduction and compression hip screw fixation. Stabilization of the unstable fracture by osteotomy and medial displacement has been tested biomechanically (Chang et al., 1987). Accurate anatomical reduction was found to produce a more physiological strain distribution and increased medial load transmission than did medial displacement osteotomy. Clinically completely accurate anatomical reduction of these comminuted fractures is very difficult. Only 42 per cent of our Group I even approached accuracy. Pun et al. (1987) used a !3armiento valg.us osteotomy, sliding hip screw and plate fixation, and augmented their reduction with acrylic bone cement posteriorly. Operating time and blood loss were considerably greater than our experience, and radiological failure still occurred in 8.6 per cent of cases. We feel that bone cement augmentation prolongs and complicates valgus osteotomy, and offers no
Figure 5. Anteroposterior radiograph of the same patient in Figure 4 demonstrating varus angulation 3 months following surgery.
benefits in terms of enhanced fracture stability. Functionally, 28.6 per cent of their patients had a fair or poor result. No details were given of their patients’ length of stay in hospital or ultimate accommodation. The valgus osteotomy has several attractive biomechanical advantages. It converts the fracture plane to a more horizontal one, allowing for better fragment compression and reducing the shear stresses at the fracture site. The more valgus position reduces the varus producing moment arm and partially compensates for the bone removal and inevitable ‘fracture settling’, although overcorrection can weaken abductor power. The medial displacement and valgus osteotomies were described in an era where the standard methods of fixation employed rigid metal implants, e.g. Jewett nail-plate, I-beam nail-plate, which had notoriously high incidences of failure of stabilization. The sliding hip screw has now superseded these implants in most centres. It is well recognized that it provides better proximal fragment control and is biomechanically a stronger total implant-fragment assembly. It provides less bone stress shielding and allows for controlled fragment collapse, with less varus angulation and reduced incidence of axial penetration and implant breakage.
Conclusion We have found no functional clinical advantage in the routine use of a valgus osteotomy with sliding hip screw fixation. In the short term the patients treated by anatomical reduction mobilized more quickly and we believe that this
88
Injury: the British Journal of Accident Surgery (1990) Vol. 21/No.
accounts for their shorter stay in hospital and early return to their pre-injury accommodation. There is no significant difference between the two groups 3 months after injury. There has been no increase in patient morbidity associated with the use of the more complex osteotomy in terms of operating time, peroperative blood loss or postoperative complications. Radiographic failure occurred more frequently in the anatomical reduction patients, suggesting that the valgus osteotomy is biomechanically superior in providing a more stable, stronger configuration. We conclude that unstable intertrochanteric fractures are best treated by reducing the fracture as anatomically as possible and using compression hip screw fixation. Valgus osteotomy does not, however, disadvantage the patient in the long term and should be reserved for those fractures which prove difficult or impossible to reduce.
Acknowledgements The authors would lie to thank Mr M. Swann, Mr N. C. Roles, Mr G. Deane, and Mr R. L. Allurn for permission to report their patients. No benefits in any form have been received from a commercial party related directly or indirectly to the subject of this article.
2
Dimon J. H. and Hughston J. C. (1967) Unstable intertrochanteric fractures of the hip. J. BoneJoint Surg. 49A, 440. Evans E. M. (1949) The treatment of intertrochanteric fractures of the femur. J. BoneJoint Srrrg SIB, 190. Harrington K. D. and Johnston J. D. (1973) The management of comminuted unstable intertrochanteric fractures. 1. Bone Joint Surg 55A, 1367. Heyse-Moore G. H., MacEacbem A. G. and Jarneson Evans D. C. (1983) Treatment of intertrochanteric fractures of the femur-A comparison of the Richards screw-plate with the Jewett nail-plate. J. BoneJoint SUR. 65B, 262. Jensen J. Steen, Tondevold E. and Mossing N. (1978) Unstable trochanteric fractures treated with the sliding screw-plate system - A biomecbanical study of unstable trochantertc fractures. Acfa orthop. Scattd.49,392. Office of Population Censuses and Surveys. (OPCS) (1987) Haspifaf In-patient Enquiy. London: HMSO. Pun W. K., chow S. P, Chan K. C. et al. (1987) Treatment of unstable inter-trochanteric fractures with Sarmiento valgus osteotomy and acrylic cement augmentation. Injny 18,384. Rao J. P., Banzon M. T., Weiss A. B. et al. (1983) Treatment of unstable intertrochanteric fractures with anatomic reduction and compression hip screw fixation. C&z. Orthop. 175, 65. Sanniento A. and Williams E. M. (1970) The unstable intertrochanteric fracture: Treatment with a valgus osteotomy and I-beam nail-plate. 1. BoneJoint Surg. 52A, 1309.
Paper accepted 20 September
1989.
References Chang
W. S., Zuckerman J. D., Kumer F. J. et al. (1987) Biomecbanical evaluation of anatomical reduction versus medial displacement in unstable intertrochanteric fractures. Clin. Orfhop. 225, 141.
Requt& for reprints should be addressed to: Mr D. W. Clark, Senior Orthopaedic Registrar, Wexham Park Hospital, Slough, Berkshire SL2 4HL, UK.
Injury (1990) 21,84-88 Prinfedin Great Britain
Treatment of unstable intertrochanteric fractures of the femur: a prospective trial comparing anatomical reduction and valgus osteotomy D. W. Clark1and W. J. Ribbans IWexham Park Hospital, Slough, UK “Central Middlesex Hospital, London, UK
A series of 100 consecufivepatients wifh u&able inferfrochunftic fracfures were treated by compressionhip screw#ofion; 55 patientshad an anatomical reducfion(Group 1)and 45 patientsa Sarmienfoosfeofomy and valgusreducfion(Group 2). Group 7 spent an averageof 10 days kxs in hospital than Group 2 (21 days compared with 31 days) (I’< 0.02). They alsohad a greater chance of refurning to their pre-injury accommodafionand of achieving theirpre-injury walking capabilify. Radiologicalfailure of fracfure fixation, with varus angulofion of the femoral head by cuffing out of thescrew,was seen seven timesin Group I but only once in Group 2. Anafomical reducfion provides better clinical results than valgus osfeofomy in the patient with an unstable inferfrochanfericfiacfure stabilizedby a compressionhip screw. 7he capacityfor failure of fracfure f&ion
is greater, however, in the former. Valgus osfeofomyprovia%sa
simple means of securing a stablereducfionof thefracfurewhich cannotbe satisfactorilyreduced by closedmeans.
Introduction Fractures of the hip are a major cause of morbidity and mortality in the elderly (OPCS, 1987). They represent an increasingly severe drain upon the resources of orthopaedic and geriatric departments. The stable intertrochanteric fracture can be successfully managed by closed reduction and internal fixation with a variety of implants. If there is medial and posterior comminution, it will lead to instability which will, on loading, progress to varus angulation and retroversion leading to failure of the fixation. Medial displacement osteotomy (Dimon and Hughston, 1967) and valgus osteotomy (Sarmiento and Williams, 1970) were introduced to produce bony stability and prevent this collapse. Compression hip screws allow collapse of the bony fragments in a controlled fashion, thus maintaining bony stability. The results of compression hip screw fixation after anatomical reduction have been good (Heyse-Moore et al., 1983; Rao et al., x983), but clinical and radiological failures still occur. This paper describes a prospective trial contrasting the results of anatomical reduction with valgus osteotomy, each followed by fixation with a compression hip screw. The 0 1990Butterworth & Co (Publishers) Ltd OOZO-1383/90/020084-05
objective was to establish whether stabilization of the fracture by valgus osteotomy would improve the patients’ rehabilitation by reducing the failures associated with anatomical reduction and compression.
Patientsand methods A series of 100 consecutive patients with type 1 unstable intertrochanteric fractures of the femur (Evans, 1949) were admitted to Wexham Park and Heatherwood Hospitals between January 1985 and June 1987 (Figure 7). An assessment was made of the patient’s pre-injury walking ability and accommodation. There were 86 women and 14 men with an average age of 83 years (range 5 l-100 years). Of these, 44 patients had a fracture of the right femur and 56 of the left femur. Eighty-five patients were operated on within 2 days of admission. The longest interval was 25 days in a patient with severe hyponatraemia. The patients were allocated by their year of birth to two treatment groups: odd years to anatomical reduction and compression hip screw fixation (Group l), and even years to valgus osteotomy and compression hip screw fixation (Group 2). There were 55 patients in Group 1 and 45 in Group 2. All the operations were performed by Registrars or Senior Registrars. Operative technique Group 1 - Closed anatomical reduction (Figure 2). The patient was positioned supine on the operating table and reduction of the fracture achieved under the image intensifier by traction and internal rotation of the leg. Internal fixation was performed with a compression hip screw and a 135”, 4 hole plate. Group 2 - Valgus osteotomy (Figrrre 3). The patient was positioned supine on the operating table with the injured leg in 10” of internal rotation. The greater trochanter and the upper third of the femoral shaft were exposed through a lateral incision. An osteotomy was performed at 30-45” to the transverse plane, running obliquely downwards and
Clark and Ribbans: Treatment of unstable intertrochanteric
Figure 1. Anteroposterior radiograph unstable intertrochanteric fracture.
of an Evans
Type
fractures of the femur
I
85
Figure 2. Anteroposterior reduction and compression
radiograph following hip screw fixation
b
a
Figure 3. a, Anteroposterior radiograph following valgus osteotomy and compression in the performance of a valgus osteotomy (modified from Sarmiento et al., 1970).
medially from inferior aspect
the flare of the greater trochanter to the of the lesser trochanteric fragment. The
greater trochanteric fragments were retracted cephalad exposing the base of the head and neck fragment. Under image intensifier control a short-thread compression screw was inserted into the centre of the neck so that its tip came to lie in the head 5 mm from the articular cartilage. A slot was cut in the lateral cortex of the distal fragment to accommo-
anatomical
hip screw fixation, b, c, Diagrams of principal stages
date the barrel of the plate. A 135”, 4 hole plate was secured to the screw and the osteotomy reduced. The intensifier was used to check the reduction and then the plate was secured to the femoral shaft. All wounds were closed in a standard fashion with a suction drain. Prophylactic antibiotics were given for the first 48 h.
The patients
were allowed
full weight
bearing
with a
86
Injury: the British Journal of Accident Surgery (1990) Vol. 21/No. 2
Table I. Patient accommodation before injury and at discharge (Modified from Heyse-Moore et al., 1983)
Grade
Original accommodation
Final accommodation
A B C D E
Home Not own home Home Not own home Any
Home Same as pre-injury Home Same as pre-injury Long-term geriatric care
N.B. -These
3
weeks weeks weeks weeks -
Anatomical group
Osteotomy group
45% 23% 7% 14% 11%
20% 22% 27% 12% 19%
figures are based upon the surviving 85 patients at the time of hospital discharge.
Table II. Level of mobility. Changes in mobility
following
surgery Level Level Level Level
Hospital stay
1 : Non-walker 2 : Supervised walker 3 : Independent-with walking aid 4 : Independent - no walking aid
Change in mobility at 3 months after surgery Same level of mobility Reduced by 1 level Reduced by 2 levels Reduced by 3 levels
Anatomical group
Valgus osteotomy group
60% 31% 9% 0%
49% 39% 9% 3%
48 h after the operation. They were discharged as soon as they were capable of coping in their pre-injury accommodation or alternative accommodation suitable to their new functional level could be found.
frame
Follow-up. All patients were reviewed clinically and radiologically at both 6 and 12 weeks after operation. The patient was discharged at 3 months if the clinical and radiological findings demonstrated sound fracture fixation and bony union. Further follow-up of patients deemed failures on clinical and/or radiological grounds has now been extended to an average of 28 months (range 12-42 months). Three patients returned to their distant homes and were lost to follow-up.
Results The distribution of age, sex and side of fracture for the two groups was Similar. Operating time and peroperative blood loss The average operating time in Group I was 53 minutes ( f 17 minutes) and in Group 2 56 minutes ( f IS minutes). The mean peroperative blood loss was 289 cc ( f I I 7 cc) in Group I and in Group 2 347 cc ( f 204 cc). Neither result reached a statistically significant level. Postoperative radiological assessment The fracture/osteotomy reduction was assessed on the first postoperative radiograph. Satisfactory reduction was considered to have been achieved if the gap between the major fragments was less than 5 mm (Jensen et al., 1978). Of the Group I patients, 23 (42 per cent) and of the Group 2 patients, 14 (31 per cent) had satisfactory reductions in both anteroposterior and lateral planes. Complications and mortality In Group I,24 (44 per cent) patients and in Group 2,19 (42
per cent) patients suffered from one or more postoperative complication, e.g. respiratory, urinary, or thromboembolic. Eleven patients in Group 1 and four in Group 2 died in hospital. A further five patients in each group died before their 3 month follow-up. This resulted in an overall 26 per cent mortality at 3 months. Hospital stay and discharge accommodation (X&k r) The average inpatient stay was 21 days ( f 17 days) in Group I and 31 days ( f 22 days) in Group 2. This was statistically significant (PC 0.02 using Fischer’s test). Patient discharges were studied using a modification of the method employed by Heyse-Moore et al. (1983). Of Group 1 patients, 89 per cent and of Group 2 patients, 81 per cent who survived returned to their previous accommodation. However, 77 per cent of Group 1 patients and only 52 per cent of Group 2 patients achieved this within 3 weeks. Walking ability (TableIr) This was assessed at 3 months and compared with pre-injury status. Of Group 1 patients, 21(58 per cent) and of Group 2 patients, 16 (48 per cent) alive had achieved their preoperative level. Clinical review (Figuxs 4 and 5) At 3 months, eight patients complained of hip pain on walking. In six cases, the radiographs were satisfactory and the pain resolved spontaneously. In two patients, the radiographs revealed varus angulation of the femoral head with cutting out of the screw. In one of these (Group I), conversion to a total hip replacement was required to alleviate pain. The other patient (Group 2) experienced spontaneous pain resolution when the fracture united. One deep infection occurred in a Group 1 patient and neccessitated the removal of the implant once bony union had occurred. Twelve patients had shortening of the affected leg greater four in Group I and eight in Group 2. Four thanlcmpatients had external rotation deformity of greater than 10” - one in Group 1 and three in Group 2. Radiological review No patient developed a non-union of their fracture. Sliding of the compression screw was commonly observed. It exceeded 10 mm in nine Group I patients and in seven Group 2 patients by 3 months. Using Jensen’s criteria for failure of stabilization Uensen et al., l978), varus angulation of greater than 10” of the femoral head was seen in seven (13 per cent) Group I patients but only once (2%) in Group 2 patients. There was no instance of mechanical failure of the implant or axial penetration of the femoral head by the implant. One of the Group I patients required revision to a total hip replacement for persistent pain.
Clark and Ribbans: Treatment of unstable intertrochanteric
fractures of the femur
87
Figure 4. Anteroposterior
radiograph demonstrating implant and fracture fragment position immediately following surgery (Group I patient).
Discussion The aim of surgery for intertrochanteric fractures is to return the patient to his or her pre-injury functional level as quickly as possible. Stable reduction of the fracture, adequate internal fixation, minimal anaesthetic time and blood loss, and early mobilization all contribute to this aim. Anatomical reduction and compression hip screw fixation produce excellent results in stable fractures, but failures still occur in the unstable fracture. Heyse-Moore et al. (1983) reported a 14 per cent radiological and an 11 per cent clinical failure rate in these fractures. Harrington and Johnson (‘1973) reported improved radiographic results and fewer reoperations with the use of a medial displacement osteotomy and compression hip screw fixation, although they made no direct comparison with anatomical reduction and compression hip screw fixation. Stabilization of the unstable fracture by osteotomy and medial displacement has been tested biomechanically (Chang et al., 1987). Accurate anatomical reduction was found to produce a more physiological strain distribution and increased medial load transmission than did medial displacement osteotomy. Clinically completely accurate anatomical reduction of these comminuted fractures is very difficult. Only 42 per cent of our Group I even approached accuracy. Pun et al. (1987) used a !3armiento valg.us osteotomy, sliding hip screw and plate fixation, and augmented their reduction with acrylic bone cement posteriorly. Operating time and blood loss were considerably greater than our experience, and radiological failure still occurred in 8.6 per cent of cases. We feel that bone cement augmentation prolongs and complicates valgus osteotomy, and offers no
Figure 5. Anteroposterior radiograph of the same patient in Figure 4 demonstrating varus angulation 3 months following surgery.
benefits in terms of enhanced fracture stability. Functionally, 28.6 per cent of their patients had a fair or poor result. No details were given of their patients’ length of stay in hospital or ultimate accommodation. The valgus osteotomy has several attractive biomechanical advantages. It converts the fracture plane to a more horizontal one, allowing for better fragment compression and reducing the shear stresses at the fracture site. The more valgus position reduces the varus producing moment arm and partially compensates for the bone removal and inevitable ‘fracture settling’, although overcorrection can weaken abductor power. The medial displacement and valgus osteotomies were described in an era where the standard methods of fixation employed rigid metal implants, e.g. Jewett nail-plate, I-beam nail-plate, which had notoriously high incidences of failure of stabilization. The sliding hip screw has now superseded these implants in most centres. It is well recognized that it provides better proximal fragment control and is biomechanically a stronger total implant-fragment assembly. It provides less bone stress shielding and allows for controlled fragment collapse, with less varus angulation and reduced incidence of axial penetration and implant breakage.
Conclusion We have found no functional clinical advantage in the routine use of a valgus osteotomy with sliding hip screw fixation. In the short term the patients treated by anatomical reduction mobilized more quickly and we believe that this
88
Injury: the British Journal of Accident Surgery (1990) Vol. 21/No.
accounts for their shorter stay in hospital and early return to their pre-injury accommodation. There is no significant difference between the two groups 3 months after injury. There has been no increase in patient morbidity associated with the use of the more complex osteotomy in terms of operating time, peroperative blood loss or postoperative complications. Radiographic failure occurred more frequently in the anatomical reduction patients, suggesting that the valgus osteotomy is biomechanically superior in providing a more stable, stronger configuration. We conclude that unstable intertrochanteric fractures are best treated by reducing the fracture as anatomically as possible and using compression hip screw fixation. Valgus osteotomy does not, however, disadvantage the patient in the long term and should be reserved for those fractures which prove difficult or impossible to reduce.
Acknowledgements The authors would lie to thank Mr M. Swann, Mr N. C. Roles, Mr G. Deane, and Mr R. L. Allurn for permission to report their patients. No benefits in any form have been received from a commercial party related directly or indirectly to the subject of this article.
2
Dimon J. H. and Hughston J. C. (1967) Unstable intertrochanteric fractures of the hip. J. BoneJoint Surg. 49A, 440. Evans E. M. (1949) The treatment of intertrochanteric fractures of the femur. J. BoneJoint Srrrg SIB, 190. Harrington K. D. and Johnston J. D. (1973) The management of comminuted unstable intertrochanteric fractures. 1. Bone Joint Surg 55A, 1367. Heyse-Moore G. H., MacEacbem A. G. and Jarneson Evans D. C. (1983) Treatment of intertrochanteric fractures of the femur-A comparison of the Richards screw-plate with the Jewett nail-plate. J. BoneJoint SUR. 65B, 262. Jensen J. Steen, Tondevold E. and Mossing N. (1978) Unstable trochanteric fractures treated with the sliding screw-plate system - A biomecbanical study of unstable trochantertc fractures. Acfa orthop. Scattd.49,392. Office of Population Censuses and Surveys. (OPCS) (1987) Haspifaf In-patient Enquiy. London: HMSO. Pun W. K., chow S. P, Chan K. C. et al. (1987) Treatment of unstable inter-trochanteric fractures with Sarmiento valgus osteotomy and acrylic cement augmentation. Injny 18,384. Rao J. P., Banzon M. T., Weiss A. B. et al. (1983) Treatment of unstable intertrochanteric fractures with anatomic reduction and compression hip screw fixation. C&z. Orthop. 175, 65. Sanniento A. and Williams E. M. (1970) The unstable intertrochanteric fracture: Treatment with a valgus osteotomy and I-beam nail-plate. 1. BoneJoint Surg. 52A, 1309.
Paper accepted 20 September
1989.
References Chang
W. S., Zuckerman J. D., Kumer F. J. et al. (1987) Biomecbanical evaluation of anatomical reduction versus medial displacement in unstable intertrochanteric fractures. Clin. Orfhop. 225, 141.
Requt& for reprints should be addressed to: Mr D. W. Clark, Senior Orthopaedic Registrar, Wexham Park Hospital, Slough, Berkshire SL2 4HL, UK.
