- Email: [email protected]
Results of a method of leg-length equalization for patients undergoing primary total hip replacement
The Journal of Arthroplasty Vol. 14 No. 2 1999
Results of a Method of Leg-Length Equalization for Patients Undergoing Primary Total Hip Replacement Steven T. Woolson, MD, James M. Hartford, MD, and Aenor Sawyer, MD
Abstract: The postoperative leg-length discrepancy was determined radiographically for a consecutive series of 351 patients (408 hips) who underwent bilateral or unilateral primary total hip replacement using a single method of leg-length equalization by preoperative planning with overlay templates. The method of equalization was performed by a measurement of the femoral head and neck segment to be resected from a reference point at the superior aspect of the dislocated femoral head. The a m o u n t of femoral bone resected was determined preoperatively by determining the dimensions of the acetabular component thickness and the femoral component head and neck height that would be replacing this resected bone and adjusting this distance for any preexisting leg-length discrepancy. Using this leglength equalization method, the length of the modular femoral head neck was chosen preoperatively, rather than using soft tissue tension across the prosthetic hip joint to determine whether the leg lengths were equal. Postoperative leg lengths were determined radiologically from a measurement from the acetabular teardrop to the lesser trochanter. Ninety-seven percent of the patients had a postoperative leg-length discrepancy that was less than 1 cm, and 86% had a leg-length difference that was 6 m m (1/4 inch) or less. The average postoperative discrepancy for these 351 patients was 1 mm. Key words: total hip replacement, leg length discrepancy.
Although the primary goals of total hip replacement are to relieve pain, to improve the ability to walk without a limp, and to increase range of motion, the equalization of leg lengths is important. Two methods of leg length equalization during total hip replacement have been described. One method described by Muller [ 1] involves preoperative planning with overlay templates to determine measurements for the correct femoral neck osteotomy site, which is referenced from the lesser trochanter intraoperatively to equalize the leg lengths. The other method involves the use of a pelvic and femoral marker for direct intraoperative measure-
merit of the preoperative and postoperative leg lengths [2--4]. This latter technique requires the use of trial implants to determine whether the leg lengths have been equalized. A review of the orthopaedic literature identified only 7 studies that reported radiographic measurements of the results of leg-length equalization after primary hip replacement [5-11]. All of these studies measured the preoperative and postoperative distance between the lesser trochanter and a specified anatomic landmark on the pelvis (either the lower aspect of the ischial tuberosity or the acetabular teardrop). Only one of these published reports [10], however, described the actual method that was used to equalize leg lengths. All of the other 6 publications reported lengthening of the leg postoperatively between 3 and 16 mm [5-9,11]. One additional abstract of a paper presentation on leg-length
From Stanford University Hospital, Stanford, California. Submitted March 15, 1998; accepted July i3, 1998. Reprint requests: Steven T. Woolson, MD, 1220 University Drive, Menlo Park, CA 94025. Copyright © 1999 by Churchill Livingstone® 0883 - 5403 / 1402 -0006510.00/0
159
160
The Journal of Arthroplasty Vol. 14 No. 2 February 1999
equalization reported that 8 of 40 primary total hip replacement patients ( 18 %) had a longer leg by 1.5 cm or more [12]. The fact that there are only 7 publications on the results of leg-length equalization after total hip replacement, a study that can be readily accomplished by simple m e a s u r e m e n t s of preoperative and postoperative radiographs, is surprising, especially considering the clinical importance of this aspect of the surgical procedure. The senior a u t h o r (S.T.W.) has used a single m e t h o d of leg-length equalization since 1986, which employs the technique of Muller [ I ] with a different intraoperative reference p o i n t - - t h e superior aspect of the femoral head. The preliminary results of this n e w e r technique used in 84 primary total hip patients was published in 1990 [10]. Subsequent to this publication, a larger consecutive series of patients undergoing primary total hip replacement was studied radiographically to determine the efficacy of this technique in eliminating a preoperative leg-length discrepancy or in maintaining equal preoperative leg lengths.
Method and Patients
Method of Leg-Length Equalization The m e t h o d of leg-length equalization used in this study provides for exact positioning of the femoral neck osteotomy fr0m calculations made from preoperative planning with overlay templates. The m e t h o d is based on the fact that if the same a m o u n t of femoral head and neck and remaining joint cartilage that is r e m o v e d during the procedure is replaced with prosthetic implants that are the same height, the leg length should remain equal (Fig. 1). An important aspect of this technique is that the vertical position of the acetabular compon e n t should not be altered from the anatomic position of the true acetabulum. During a primary total hip replacement in a patient w h o has degenerative arthritis, however, it is unlikely that the surgeon would err in placement of the acetabular c o m p o n e n t in a superior position because this would require removal of the superior subchrondral bone of the acetabulum. The major difference b e t w e e n this m e t h o d and other reported methods of leglength equalization lies in the m e a s u r e m e n t of the site of the femoral neck osteotomy from the landmark of the superior aspect of the femoral head rather t h a n from the lesser trochanter. The m e t h o d involves an initial determination of any preoperative leg-length discrepancy from the clinical examination and from a m e a s u r e m e n t of the preoperative radiographs. The joint space (if
HIP JOINT CARTILAGE SPACE
FEMORAL " ~ HEAD & NECK ~,\ SEGMENT ~
( ACETABULAR COMPONENT HEIGHT
FEMORAL PROSTHESIS HEAD & NECK HEIGHT
Fig. 1. The basis for this method of leg-length equalization is to equalize the amount of femoral bone and joint cartilage that is removed during total hip replacement to the height of the acetabular and femoral components that are implanted. Any remaining joint cartilage space is measured from the preoperative radiograph (left). The femoral head and neck segment (left) that will be resected is determined by adding the heights of the acetabular component and of the femoral prosthesis head and neck segment (right), which are found by preoperative templating and then subtracting half of the remaining superior cartilage space. Adjustments to the femoral head and neck segment to be resected (left) are made to correct a preoperative leg-length discrepancy. The measurement of the starting point of the femoral neck osteotomy is referenced from the superior aspect of the femoral head rather than from the lesser trochanter, as shown on the left.
any) at the most superior aspect of the femoral head is m e a s u r e d radiographically. The acetabulum and proximal f e m u r are templated using clear overlay templates, which are supplied by all manufacturers and are enlarged by 20% (the average magnification of bone images on a plain radiograph of the hip joint). The thickness of the acetabular c o m p o n e n t is determined by subtracting the inner diameter of the acetabular c o m p o n e n t from the outer diameter and dividing this n u m b e r by 2 ([acetabular c o m p o n e n t outside d i a m e t e r - acetabular c o m p o n e n t inside diameter]/2). This calculation finds the radius of the c o m p o n e n t or the distance from the inner surface of the polyethylene to the outer surface of the implant. The vertical height of the femoral c o m p o n e n t from the top of the femoral head to the bottom surface of the collar of the prosthesis (or the point on the prosthesis that is level with the cut surface of the osteotomized femoral neck, if the prosthesis is not designed with a collar) can be measured from the
Leg-Length Equalization DuringTHR
femoral prosthesis template or may be obtained from the manufacturer. Another method of determining the femoral prosthesis head and neck height is to measure the trial implant directly. If the prosthesis has a modular femoral head, this distance may be adjusted by using a different length head, but the surgeon must choose the neck length of the implant preoperatively so that the femoral neck osteotomy may be made at the appropriate level for this specific sized head and neck segment. Once the dimensions of the acetabular component and the femoral prosthesis head and neck height are known, the appropriate amount of femoral bone to be resected can be calculated. If the preoperative leg lengths are equal, if there is no superior joint cartilage space remaining, and if the acetabular component is correctly placed in the anatomic position, the femoral neck osteotomy should be made at a point on the femoral neck that is the exact distance from the superior aspect of the femoral head to the neck as the added dimensions of the acetabular and femoral components to be inserted (femoral head and neck resection distance = acetabular component thickness or height + femoral prosthesis head and neck height). If the patient's preoperative leg length were 1 cm short, the added dimensions of the implants minus 1 cm would be the correct level for the femoral neck osteotomy. If there is an unusually wide joint space superiorly on the preoperative radiograph (>2 mm), approximately half of this remaining cartilage is present on the acetabular side of the joint, and the removal of this cartilage must be accounted for during the planning calculations. Because any remaining acetabular cartilage is removed during acetabular reaming, this amount must be subtracted from the femoral neck osteotomy measurement, or the leg will be made several millimeters too short. During the operation after the femoral head is dislocated, a caliper is used to measure the appropriate amount of femoral head and neck to be resected with the reference point at the superior aspect of the head rather than at the lesser trochanter. This measurement is made from the top of the femoral head in a direct line to the medial aspect of the neck where a mark is made with the electrocautery for the starting point for the saw. The top of the femoral head is used as the reference point for the measurement of the osteotomy site because the femoral head is always readily visible, whereas the lesser trochanter may not be, and the femoral head is thus usually a more accurate landmark from which to measure. Once the osteotomy is made, the resected femoral head and neck should be measured again to ensure that the appropriate amount of bone has
•
Woolson et al.
161
been resected. During preparation of the acetabulum, care must be taken to avoid excessive superior reaming, which may shorten the leg; however, virtually all surgical techniques recommend preservation of the superior subchondral bone of the acetabulum whether a cemented or uncemented acetabular component is used, and therefore error from superior reaming is rarely encountered.
Patients A consecutive series of 375 patients who underwent 431 primary total hip replacements between January 1988 and August I995 were studied. All total hip operations were performed using this method of leg-length equalization without using direct intraoperative measurement from a pelvic marking pin or other device and without soft tissue tensioning during the trial reduction as a determinant in changing the neck length of the prosthesis from the prosthesis size that was initially chosen during preoperative planning. An anteroposterior radiograph of the pelvis centered over the symphysis pubis that was taken within 4 weeks of the operation was available for 414 hips. All of the radiographs were taken with the x-ray tube placed 40 inches from the film and with the patient's feet positioned perpendicular to the x-ray table, but no special positioning device was used. Six patients (6 hips) were excluded because an attempt was not made to equalize leg lengths in 4 patients owing to a severe preoperative discrepancy greater than 2.5 cm and because the pelvic radiographs for the other 2 patients were inadequate for measurement of leg lengths owing to lead shielding of the patient. The remaining 351 patients (408 hips) (95% of the original group of 431 hips) composed the study group. All patients had a preoperative pelvic radiograph available at the time of the initial preoperative planning, but at the time of this study a preoperative pelvic radiograph was available for 296 (84%) of these 351 patients for a retrospective preoperative leg-length determination. Fifty-seven of the 351 patients underwent bilateral total hip replacement (7 of w hom had had 1-stage procedures during 1 anesthetic session and 50 of w h o m had staged procedures). The preoperative diagnoses of these 351 patients included osteoarthritis in 256 patients (73%), osteonecrosis in 31 (9%), failed hip fracture in 20 (6%), rheumatoid arthritis in 17 (5 %), congenital dysplasia in 11 (3%), and miscellaneous conditions in 16 (5%). None of these patients had a preoperative diagnosis of complete neglected hip dislocation. The average age of these patients was 65 years (range,
162
The Journal of Arthroplasty Vol. 14 No. 2 February 1999
22-92 years), and their average height a n d weight w e r e 169 cm (range, 127-196 cm) and 75 kg (range, 39-145 kg). There w e r e 185 w o m e n and I66 m e n . Ninety-eight percent of the procedures w e r e perf o r m e d t h r o u g h a posterolateral approach with a complete capsulectomy, with the other 10 procedures t h r o u g h a transtrochanteric approach. All but 1 of these patients had an acetabular c o m p o n e n t , which was a cementless metal-backed, porouscoated cup, and that patient had a c e m e n t e d c o m p o nent. The average outside diameter of the acetabular c o m p o n e n t s used was 54 m m (range, 4 2 - 6 8 m m ) . A c e m e n t e d femoral c o m p o n e n t was used for 310 hips, and a cementless p o r o u s - c o a t e d femoral c o m p o n e n t was used for the remaining 98 hips. All of the femoral c o m p o n e n t s had a m o d u l a r h e a d and a collar. The neck length of the m o d u l a r h e a d was m e d i u m in 306 hips (75%), long in 65 (16%), and short in 37 (9%). The femoral head d i a m e t e r was 28 m m in 363 hips (89%), and 22, 26, or 32 m m in 45 hips (11%). All but I0 (3%) of these 351 patients w e r e followed for a m i n i m u m of 1 year postoperatively; 2 patients h a d died within 1 year, and 8 patients w e r e lost to follow-up after 3 to 6 m o n t h s . The leg-length discrepancies of the 294 patients w h o had u n d e r g o n e unilateral total hip replacem e n t were a c o m p a r i s o n of the operated hip to the contralateral hip. The leg-length discrepancy of the right hip only (arbitrarily using the left leg as a reference) was recorded for the 57 patients w h o had undergone bilateral hip replacement. Therefore there w e r e a total of 351 leg-length m e a s u r e m e n t s for 408 hips (351 patients).
Radiographic Measurement of Leg Length M e a s u r e m e n t s of the radiographs to d e t e r m i n e a leg-length discrepancy w e r e m a d e by an observer w h o was not involved with a n y of the operations to eliminate bias. A horizontal line was d r a w n t h r o u g h points at the m o s t inferior aspect of the acetabular teardrop of each hemipelvis. The acetabular teardrop was used as a l a n d m a r k for these m e a s u r e m e n t s rather t h a n the m o s t inferior aspect of the ischial tuberosity because the teardrop is a m o r e discrete a n a t o m i c structure, and its vertical position is not affected significantly by rotation of the pelvis [ 13]. Two other lines w e r e d r a w n parallel to the teardrop line t h r o u g h points at the center of lesser t r o c h a n t e r for each femur. The difference b e t w e e n the distances f r o m the teardrop line to the the lesser trochanters of each f e m u r was defined as the leg-length discrepancy. All m e a s u r e m e n t s were m a d e using a ruler, w h i c h was magnified 2 0 % to account for the e n l a r g e m e n t of b o n e images on
the radiographs. The interobserver variation of these leg-length m e a s u r e m e n t s was a m e a n of 0.5 mm.
The authors assumed that equalization of the distance b e t w e e n the pelvis and the lesser trochanter on each side w o u l d equalize the patients' leg lengths because no patient h a d other lower e x t r e m ity deformities that invalidate this assumption. Furt h e r m o r e , this was the m e t h o d of determining leg lengths in all of the studies that w e r e referenced earlier.
Results Of the 296 patients w h o had a preoperative radiograph available for retrospective review, the average preoperative discrepancy was - 2 . 9 m m (range, - 5 4 - +21 m m ) . Of these 296 patients, 199 patients (67%) had a shorter leg, 17 patients (6%) had equal leg lengths, and 80 patients (27%) had a longer leg. T w e n t y - t w o (28%) of the 80 patients w h o had a longer leg had had the first procedure of staged bilateral hip r e p l a c e m e n t s p e r f o r m e d on the contralateral hip (the reference hip) before this index m e a s u r e m e n t . Thirteen percent (39 of 296 hips) h a d a preoperative discrepancy of 10 m m or more. The average postoperative leg-length discrepancy was 1 m m (range, - 2 0 - + 2 2 m m , standard deviation, _+4.7 m m ) . A total of 301 patients (86%) h a d a positive or negative discrepancy of 6 m m or less (% inch); of these 301 patients, 36 had equal leg lengths, 126 had a longer leg by 1 to 6 m m , and i39 had a shorter leg by I to 6 m m . Of the r e m a i n i n g 50 patients (14%) w h o h a d a leg length difference that was greater t h a n 6 m m , only 12 (3%) had a leg-length discrepancy that was greater or equal to 10 m m (8 of w h o m h a d a short leg and 4 of w h o m had a long leg). The average change b e t w e e n the preoperative and postoperative leg length for the 296 patients w h o h a d a preoperative radiograph was +2.5 m m (range, - 1 6 - -t-58 m m ) . C o m p a r i n g the subset of 50 patients w h o had a postoperative leg-length discrepancy greater t h a n 6 m m with the 301 patients w h o h a d a discrepancy of 6 m m or less, there was no significant difference in the preoperative leg-length discrepancy ( - 2 . 7 m m vs - 2 . 9 m m , [P = .9] ). There was also no significant difference in the preoperative diagnoses or the intraoperative leg-length change b e t w e e n these 2 groups. Of the 8 patients w h o s e leg length was 10 m m or m o r e short, 5 had a preoperative radiograph available for review, all of which s h o w e d a shorter leg.
Leg-Length Equalization DuringTHR • One of these 5 patients had a smaller negative discrepancy postoperatively by 8 mm, and the other 4 had greater negative discrepancies by 5 to 10 mm. Three of the 4 patients whose final leg length was l 0 m m or more long had had a long leg preoperatively. Two of these 3 patients had had additional leg lengthening during the procedure of 5 and 6 mm. One other patient whose leg was long preoperatively had i m p r o v e m e n t of his leg-length discrepancy by 9 mm, but still remained 12 m m long postoperatively. The m e a n postoperative leg-length discrepancy of the 39 patients w h o had a preoperative discrepancy of 10 m m or more was 1.7 m m compared with 0.1 m m for the 257 patients w h o had a preoperative discrepancy of less than 10 m m (P = .055). Four patients w h o had had unilateral hip replacem e n t (0.9% of the 408 hips) were found to have a peroneal nerve palsy postoperatively, all of which had resolved completely by 6 to 24 months. None of these patients had a postoperative leg-length discrepancy that was greater t h a n 6 mm, and 3 of the 4 had a preoperative radiograph for review. Of the 3 w h o had a preoperative leg-length m e a s u r e m e n t , 1 patient's leg was made longer by 3 mm, 1 patient's leg was shorter by 3 mm, and there was no change in the leg length of the other patient. Twenty-two of these 351 patients (5 %) sustained a postoperative dislocation of 1 hip during followup. The leg length of all 351 patients was classified as long, if the leg was 3 m m or more longer t h a n the reference leg; equal, if it was between +2 and - 2 m m of the reference leg; and short, if it was 3 m m or m o r e shorter than the reference leg. Using this classification of leg-length discrepancy (long, equal, or short), there was no correlation b e t w e e n the leg-length discrepancy of this group of patients and the occurrence of a dislocation (chi square, P = .33).
Discussion A review of the orthopaedic literature for articles on the results of leg-length equalization after primary total hip replacement f o u n d only 6 publications that reported data on postoperative leg-length discrepancies. Five of these 6 studies reported leg lengthening, which averaged b e t w e e n 9 and 16 m m [5-9]. One of these reports found that 144 of 150 extremities were l e n g t h e n e d an average of 16 m m [8]. The remaining study reported an average leg-length discrepancy of 3.4 mm; however, neither the latter study with good results nor the other 5 studies with greater discrepancies described the specific m e t h o d of leg-length equalization that was
Woolson et al.
163
used for the patients in each study. These 5 studies reported results that are clearly suboptimal because an extremity that is more than 1 cm long results in a limp and m a y require a shoe lift on the contralateral side. The results of the m e t h o d of leg-length equalization presented here are superior to all other published studies in that only 12 of 351 patients (3%) had a discrepancy greater than l0 mm, and only 4 of these 351 patients (1%) had leg lengthening that was greater than 10 mm. Statistical analysis determined that there was a trend (P = . 0 5 5 ) f o r patients w h o had a preoperative discrepancy that was greater t h a n 10 m m to have a greater postoperative leg-length difference than for those patients w h o had a preoperative discrepancy less than 10 mm. This trend indicates that this m e t h o d is more accurate for patients w h o have smaller preoperative discrepancies. This technique is ideal for the routine patient w h o has a minimal preoperative leg-length difference and norreal bone anatomy. Patients w h o have severe deformities or major leg-length discrepancies m a y require leg-length equalization by other methods, such as intraoperative m e a s u r e m e n t from pelvic and femoral markers. This simple m e t h o d requires only the use of overlay templates for preoperative planning, a simple arithmatic formula, and a caliper for intraoperative m e a s u r e m e n t . It does not require repeated trials of different neck lengths for adjustment of intraoperative measurements. We believe that the reason this m e t h o d is more accurate than other reported techniques is that the m e a s u r e m e n t for the femoral neck osteotomy is made from the superior aspect of the femoral head rather than from the lesser trochanter. Visualization of the lesser trochanter is limited during primary total hip replacement, and any m e a s u r e m e n t from it is usually made by palpation rather than by using a ruler or caliper. The superior aspect of the femoral head, however, is plainly visible during all surgical approaches. The m e t h o d of intraoperative m e a s u r e m e n t of leg length using a pelvic pin and a femoral marker may also be accurate; however, there have been no reported results of the use of this technique to date. The authors found that the postoperative leg lengths of the 4 patients w h o sustained a transcient peroneal nerve palsy were not altered more than 6 mm, and therefore these nerve injuries were probably not a result of leg lengthening. Excessive leg lengthening, however, can result in sciatic nerve palsy, as pointed out by Williamson and Reckling [81, w h o found that 3% of patients w h o had significant leg lengthening (average 16 mm) developed a nerve palsy.
164
The Journal of Arthroplasty Vol. 14 No. 2 February 1999
We believe that determining leg lengths intraoperatively solely on the basis of soft tissue tensioning by the so-called shuck m e t h o d [14], w h i c h is used by some surgeons to d e t e r m i n e intraoperative leg lengths, m a y frequently cause excessive leg lengthening, as was reported by all studies f o u n d in the literature. This technique, w h i c h encourages the surgeon to use a longer femoral prosthesis neck length to tighten the soft tissues a r o u n d the hip to i m p r o v e the stability of the trial implants, is likely to result in leg lengthening, as all hip reconstructions are m a d e m o r e stable if the hip joint space is lengthened. We believe that excessive leg lengthening is an unacceptable trade-off for increased stability. In contradiction to the rationalization of the t h e o r y that inadequate soft tissue tensioning causes increased postoperative stability, our study f o u n d that there was no increased prevalence of postoperative dislocation in patients w h o had a shorter leg c o m p a r e d with patients w h o had a longer leg or equal leg lengths. Leg-length inequality is a m a j o r cause for malpractice liability suits after total hip replacement, especially w h e n a patient has a longer leg requiring a lift on the contralateral side. Surgeons w h o p e r f o r m total hip r e p l a c e m e n t should be a w a r e that m a n y of the reports in the literature of leg-length results after total hip r e p l a c e m e n t h a v e s h o w n excessive leg lengthening, p r e s u m a b l y f r o m other surgical techniques previously reported, and should adopt a m e t h o d of equalization that is reliable and easy to perform. We believe that because this m e t h o d is simple, is effective, and requires no additional time or cost, it should be strongly considered as an alternative to other techniques that lack data proving their accuracy.
References 1. Muller ME: M.E. Muller straight stem total hip replacement system. Protek Ltd. Manual, Berne, Switzerland, 1982 2. Knight WE: Accurate determination of leg lengths during total hip replacement. Clin Orthop 123:27, 1977 3. McGee HMJ, Scott JHS: A simple method of obtaining equal leg length in total hip arthroplasty. Clin Orthop i94:269, 1985 4. Woolson ST, Harris WH: A method of intraoperative limb length measurement in total hip arthroplasty. Clin Orthop 194:201, 1985 5. Edeen J, Sharkey PF, Alexander AH: Clinical significance of leg-length inequality after total hip arthroplasty. Am J Orthop 24:347, 1995 6. Rand JA, Ilstrup DM: Comparison of Charnley and T-28 total hip arthroplasty. Clin Orthop 180:201, 1983 7. Turula KB, Friberg O, Lindholm TS, et ah Leg length inequality after total hip arthroplasty. Clin Orthop 202:163, 1986 8. Williamson JA, Reckling FW: Limb length discrepancy and related problems following total hip replacement. Clin Orthop 134:135, 1978 9. Woo RYG, Morrey BF: Dislocation after total hip arthroplasty. J Bone Joint Surg Am 64:1295, 1982 10. Woolson ST: Leg length equalization during total hip replacement. Orthopedics 13:17, 1990 11. Ranawat CS, Rodriguez JA: Functional leg-length inequality following total hip arthroplasty. J Arthroplasty 12:359, 1997 12. Love BRT, Wright K: Leg length discrepancy after total hip joint replacement. J Bone Joint Surg Br 65:103, i983 13. Goodman SB, Adler SJ, Fyhrie DP, Schurman DJ: The acetabular teardrop and its relevance to acetabular migration. Clin Orthop 236:199, 1988 14. Charnley J: Low friction arthroplasty of the hip. Springer-Verlag, Berlin, 1979
Results of a Method of Leg-Length Equalization for Patients Undergoing Primary Total Hip Replacement Steven T. Woolson, MD, James M. Hartford, MD, and Aenor Sawyer, MD
Abstract: The postoperative leg-length discrepancy was determined radiographically for a consecutive series of 351 patients (408 hips) who underwent bilateral or unilateral primary total hip replacement using a single method of leg-length equalization by preoperative planning with overlay templates. The method of equalization was performed by a measurement of the femoral head and neck segment to be resected from a reference point at the superior aspect of the dislocated femoral head. The a m o u n t of femoral bone resected was determined preoperatively by determining the dimensions of the acetabular component thickness and the femoral component head and neck height that would be replacing this resected bone and adjusting this distance for any preexisting leg-length discrepancy. Using this leglength equalization method, the length of the modular femoral head neck was chosen preoperatively, rather than using soft tissue tension across the prosthetic hip joint to determine whether the leg lengths were equal. Postoperative leg lengths were determined radiologically from a measurement from the acetabular teardrop to the lesser trochanter. Ninety-seven percent of the patients had a postoperative leg-length discrepancy that was less than 1 cm, and 86% had a leg-length difference that was 6 m m (1/4 inch) or less. The average postoperative discrepancy for these 351 patients was 1 mm. Key words: total hip replacement, leg length discrepancy.
Although the primary goals of total hip replacement are to relieve pain, to improve the ability to walk without a limp, and to increase range of motion, the equalization of leg lengths is important. Two methods of leg length equalization during total hip replacement have been described. One method described by Muller [ 1] involves preoperative planning with overlay templates to determine measurements for the correct femoral neck osteotomy site, which is referenced from the lesser trochanter intraoperatively to equalize the leg lengths. The other method involves the use of a pelvic and femoral marker for direct intraoperative measure-
merit of the preoperative and postoperative leg lengths [2--4]. This latter technique requires the use of trial implants to determine whether the leg lengths have been equalized. A review of the orthopaedic literature identified only 7 studies that reported radiographic measurements of the results of leg-length equalization after primary hip replacement [5-11]. All of these studies measured the preoperative and postoperative distance between the lesser trochanter and a specified anatomic landmark on the pelvis (either the lower aspect of the ischial tuberosity or the acetabular teardrop). Only one of these published reports [10], however, described the actual method that was used to equalize leg lengths. All of the other 6 publications reported lengthening of the leg postoperatively between 3 and 16 mm [5-9,11]. One additional abstract of a paper presentation on leg-length
From Stanford University Hospital, Stanford, California. Submitted March 15, 1998; accepted July i3, 1998. Reprint requests: Steven T. Woolson, MD, 1220 University Drive, Menlo Park, CA 94025. Copyright © 1999 by Churchill Livingstone® 0883 - 5403 / 1402 -0006510.00/0
159
160
The Journal of Arthroplasty Vol. 14 No. 2 February 1999
equalization reported that 8 of 40 primary total hip replacement patients ( 18 %) had a longer leg by 1.5 cm or more [12]. The fact that there are only 7 publications on the results of leg-length equalization after total hip replacement, a study that can be readily accomplished by simple m e a s u r e m e n t s of preoperative and postoperative radiographs, is surprising, especially considering the clinical importance of this aspect of the surgical procedure. The senior a u t h o r (S.T.W.) has used a single m e t h o d of leg-length equalization since 1986, which employs the technique of Muller [ I ] with a different intraoperative reference p o i n t - - t h e superior aspect of the femoral head. The preliminary results of this n e w e r technique used in 84 primary total hip patients was published in 1990 [10]. Subsequent to this publication, a larger consecutive series of patients undergoing primary total hip replacement was studied radiographically to determine the efficacy of this technique in eliminating a preoperative leg-length discrepancy or in maintaining equal preoperative leg lengths.
Method and Patients
Method of Leg-Length Equalization The m e t h o d of leg-length equalization used in this study provides for exact positioning of the femoral neck osteotomy fr0m calculations made from preoperative planning with overlay templates. The m e t h o d is based on the fact that if the same a m o u n t of femoral head and neck and remaining joint cartilage that is r e m o v e d during the procedure is replaced with prosthetic implants that are the same height, the leg length should remain equal (Fig. 1). An important aspect of this technique is that the vertical position of the acetabular compon e n t should not be altered from the anatomic position of the true acetabulum. During a primary total hip replacement in a patient w h o has degenerative arthritis, however, it is unlikely that the surgeon would err in placement of the acetabular c o m p o n e n t in a superior position because this would require removal of the superior subchrondral bone of the acetabulum. The major difference b e t w e e n this m e t h o d and other reported methods of leglength equalization lies in the m e a s u r e m e n t of the site of the femoral neck osteotomy from the landmark of the superior aspect of the femoral head rather t h a n from the lesser trochanter. The m e t h o d involves an initial determination of any preoperative leg-length discrepancy from the clinical examination and from a m e a s u r e m e n t of the preoperative radiographs. The joint space (if
HIP JOINT CARTILAGE SPACE
FEMORAL " ~ HEAD & NECK ~,\ SEGMENT ~
( ACETABULAR COMPONENT HEIGHT
FEMORAL PROSTHESIS HEAD & NECK HEIGHT
Fig. 1. The basis for this method of leg-length equalization is to equalize the amount of femoral bone and joint cartilage that is removed during total hip replacement to the height of the acetabular and femoral components that are implanted. Any remaining joint cartilage space is measured from the preoperative radiograph (left). The femoral head and neck segment (left) that will be resected is determined by adding the heights of the acetabular component and of the femoral prosthesis head and neck segment (right), which are found by preoperative templating and then subtracting half of the remaining superior cartilage space. Adjustments to the femoral head and neck segment to be resected (left) are made to correct a preoperative leg-length discrepancy. The measurement of the starting point of the femoral neck osteotomy is referenced from the superior aspect of the femoral head rather than from the lesser trochanter, as shown on the left.
any) at the most superior aspect of the femoral head is m e a s u r e d radiographically. The acetabulum and proximal f e m u r are templated using clear overlay templates, which are supplied by all manufacturers and are enlarged by 20% (the average magnification of bone images on a plain radiograph of the hip joint). The thickness of the acetabular c o m p o n e n t is determined by subtracting the inner diameter of the acetabular c o m p o n e n t from the outer diameter and dividing this n u m b e r by 2 ([acetabular c o m p o n e n t outside d i a m e t e r - acetabular c o m p o n e n t inside diameter]/2). This calculation finds the radius of the c o m p o n e n t or the distance from the inner surface of the polyethylene to the outer surface of the implant. The vertical height of the femoral c o m p o n e n t from the top of the femoral head to the bottom surface of the collar of the prosthesis (or the point on the prosthesis that is level with the cut surface of the osteotomized femoral neck, if the prosthesis is not designed with a collar) can be measured from the
Leg-Length Equalization DuringTHR
femoral prosthesis template or may be obtained from the manufacturer. Another method of determining the femoral prosthesis head and neck height is to measure the trial implant directly. If the prosthesis has a modular femoral head, this distance may be adjusted by using a different length head, but the surgeon must choose the neck length of the implant preoperatively so that the femoral neck osteotomy may be made at the appropriate level for this specific sized head and neck segment. Once the dimensions of the acetabular component and the femoral prosthesis head and neck height are known, the appropriate amount of femoral bone to be resected can be calculated. If the preoperative leg lengths are equal, if there is no superior joint cartilage space remaining, and if the acetabular component is correctly placed in the anatomic position, the femoral neck osteotomy should be made at a point on the femoral neck that is the exact distance from the superior aspect of the femoral head to the neck as the added dimensions of the acetabular and femoral components to be inserted (femoral head and neck resection distance = acetabular component thickness or height + femoral prosthesis head and neck height). If the patient's preoperative leg length were 1 cm short, the added dimensions of the implants minus 1 cm would be the correct level for the femoral neck osteotomy. If there is an unusually wide joint space superiorly on the preoperative radiograph (>2 mm), approximately half of this remaining cartilage is present on the acetabular side of the joint, and the removal of this cartilage must be accounted for during the planning calculations. Because any remaining acetabular cartilage is removed during acetabular reaming, this amount must be subtracted from the femoral neck osteotomy measurement, or the leg will be made several millimeters too short. During the operation after the femoral head is dislocated, a caliper is used to measure the appropriate amount of femoral head and neck to be resected with the reference point at the superior aspect of the head rather than at the lesser trochanter. This measurement is made from the top of the femoral head in a direct line to the medial aspect of the neck where a mark is made with the electrocautery for the starting point for the saw. The top of the femoral head is used as the reference point for the measurement of the osteotomy site because the femoral head is always readily visible, whereas the lesser trochanter may not be, and the femoral head is thus usually a more accurate landmark from which to measure. Once the osteotomy is made, the resected femoral head and neck should be measured again to ensure that the appropriate amount of bone has
•
Woolson et al.
161
been resected. During preparation of the acetabulum, care must be taken to avoid excessive superior reaming, which may shorten the leg; however, virtually all surgical techniques recommend preservation of the superior subchondral bone of the acetabulum whether a cemented or uncemented acetabular component is used, and therefore error from superior reaming is rarely encountered.
Patients A consecutive series of 375 patients who underwent 431 primary total hip replacements between January 1988 and August I995 were studied. All total hip operations were performed using this method of leg-length equalization without using direct intraoperative measurement from a pelvic marking pin or other device and without soft tissue tensioning during the trial reduction as a determinant in changing the neck length of the prosthesis from the prosthesis size that was initially chosen during preoperative planning. An anteroposterior radiograph of the pelvis centered over the symphysis pubis that was taken within 4 weeks of the operation was available for 414 hips. All of the radiographs were taken with the x-ray tube placed 40 inches from the film and with the patient's feet positioned perpendicular to the x-ray table, but no special positioning device was used. Six patients (6 hips) were excluded because an attempt was not made to equalize leg lengths in 4 patients owing to a severe preoperative discrepancy greater than 2.5 cm and because the pelvic radiographs for the other 2 patients were inadequate for measurement of leg lengths owing to lead shielding of the patient. The remaining 351 patients (408 hips) (95% of the original group of 431 hips) composed the study group. All patients had a preoperative pelvic radiograph available at the time of the initial preoperative planning, but at the time of this study a preoperative pelvic radiograph was available for 296 (84%) of these 351 patients for a retrospective preoperative leg-length determination. Fifty-seven of the 351 patients underwent bilateral total hip replacement (7 of w hom had had 1-stage procedures during 1 anesthetic session and 50 of w h o m had staged procedures). The preoperative diagnoses of these 351 patients included osteoarthritis in 256 patients (73%), osteonecrosis in 31 (9%), failed hip fracture in 20 (6%), rheumatoid arthritis in 17 (5 %), congenital dysplasia in 11 (3%), and miscellaneous conditions in 16 (5%). None of these patients had a preoperative diagnosis of complete neglected hip dislocation. The average age of these patients was 65 years (range,
162
The Journal of Arthroplasty Vol. 14 No. 2 February 1999
22-92 years), and their average height a n d weight w e r e 169 cm (range, 127-196 cm) and 75 kg (range, 39-145 kg). There w e r e 185 w o m e n and I66 m e n . Ninety-eight percent of the procedures w e r e perf o r m e d t h r o u g h a posterolateral approach with a complete capsulectomy, with the other 10 procedures t h r o u g h a transtrochanteric approach. All but 1 of these patients had an acetabular c o m p o n e n t , which was a cementless metal-backed, porouscoated cup, and that patient had a c e m e n t e d c o m p o nent. The average outside diameter of the acetabular c o m p o n e n t s used was 54 m m (range, 4 2 - 6 8 m m ) . A c e m e n t e d femoral c o m p o n e n t was used for 310 hips, and a cementless p o r o u s - c o a t e d femoral c o m p o n e n t was used for the remaining 98 hips. All of the femoral c o m p o n e n t s had a m o d u l a r h e a d and a collar. The neck length of the m o d u l a r h e a d was m e d i u m in 306 hips (75%), long in 65 (16%), and short in 37 (9%). The femoral head d i a m e t e r was 28 m m in 363 hips (89%), and 22, 26, or 32 m m in 45 hips (11%). All but I0 (3%) of these 351 patients w e r e followed for a m i n i m u m of 1 year postoperatively; 2 patients h a d died within 1 year, and 8 patients w e r e lost to follow-up after 3 to 6 m o n t h s . The leg-length discrepancies of the 294 patients w h o had u n d e r g o n e unilateral total hip replacem e n t were a c o m p a r i s o n of the operated hip to the contralateral hip. The leg-length discrepancy of the right hip only (arbitrarily using the left leg as a reference) was recorded for the 57 patients w h o had undergone bilateral hip replacement. Therefore there w e r e a total of 351 leg-length m e a s u r e m e n t s for 408 hips (351 patients).
Radiographic Measurement of Leg Length M e a s u r e m e n t s of the radiographs to d e t e r m i n e a leg-length discrepancy w e r e m a d e by an observer w h o was not involved with a n y of the operations to eliminate bias. A horizontal line was d r a w n t h r o u g h points at the m o s t inferior aspect of the acetabular teardrop of each hemipelvis. The acetabular teardrop was used as a l a n d m a r k for these m e a s u r e m e n t s rather t h a n the m o s t inferior aspect of the ischial tuberosity because the teardrop is a m o r e discrete a n a t o m i c structure, and its vertical position is not affected significantly by rotation of the pelvis [ 13]. Two other lines w e r e d r a w n parallel to the teardrop line t h r o u g h points at the center of lesser t r o c h a n t e r for each femur. The difference b e t w e e n the distances f r o m the teardrop line to the the lesser trochanters of each f e m u r was defined as the leg-length discrepancy. All m e a s u r e m e n t s were m a d e using a ruler, w h i c h was magnified 2 0 % to account for the e n l a r g e m e n t of b o n e images on
the radiographs. The interobserver variation of these leg-length m e a s u r e m e n t s was a m e a n of 0.5 mm.
The authors assumed that equalization of the distance b e t w e e n the pelvis and the lesser trochanter on each side w o u l d equalize the patients' leg lengths because no patient h a d other lower e x t r e m ity deformities that invalidate this assumption. Furt h e r m o r e , this was the m e t h o d of determining leg lengths in all of the studies that w e r e referenced earlier.
Results Of the 296 patients w h o had a preoperative radiograph available for retrospective review, the average preoperative discrepancy was - 2 . 9 m m (range, - 5 4 - +21 m m ) . Of these 296 patients, 199 patients (67%) had a shorter leg, 17 patients (6%) had equal leg lengths, and 80 patients (27%) had a longer leg. T w e n t y - t w o (28%) of the 80 patients w h o had a longer leg had had the first procedure of staged bilateral hip r e p l a c e m e n t s p e r f o r m e d on the contralateral hip (the reference hip) before this index m e a s u r e m e n t . Thirteen percent (39 of 296 hips) h a d a preoperative discrepancy of 10 m m or more. The average postoperative leg-length discrepancy was 1 m m (range, - 2 0 - + 2 2 m m , standard deviation, _+4.7 m m ) . A total of 301 patients (86%) h a d a positive or negative discrepancy of 6 m m or less (% inch); of these 301 patients, 36 had equal leg lengths, 126 had a longer leg by 1 to 6 m m , and i39 had a shorter leg by I to 6 m m . Of the r e m a i n i n g 50 patients (14%) w h o h a d a leg length difference that was greater t h a n 6 m m , only 12 (3%) had a leg-length discrepancy that was greater or equal to 10 m m (8 of w h o m h a d a short leg and 4 of w h o m had a long leg). The average change b e t w e e n the preoperative and postoperative leg length for the 296 patients w h o h a d a preoperative radiograph was +2.5 m m (range, - 1 6 - -t-58 m m ) . C o m p a r i n g the subset of 50 patients w h o had a postoperative leg-length discrepancy greater t h a n 6 m m with the 301 patients w h o h a d a discrepancy of 6 m m or less, there was no significant difference in the preoperative leg-length discrepancy ( - 2 . 7 m m vs - 2 . 9 m m , [P = .9] ). There was also no significant difference in the preoperative diagnoses or the intraoperative leg-length change b e t w e e n these 2 groups. Of the 8 patients w h o s e leg length was 10 m m or m o r e short, 5 had a preoperative radiograph available for review, all of which s h o w e d a shorter leg.
Leg-Length Equalization DuringTHR • One of these 5 patients had a smaller negative discrepancy postoperatively by 8 mm, and the other 4 had greater negative discrepancies by 5 to 10 mm. Three of the 4 patients whose final leg length was l 0 m m or more long had had a long leg preoperatively. Two of these 3 patients had had additional leg lengthening during the procedure of 5 and 6 mm. One other patient whose leg was long preoperatively had i m p r o v e m e n t of his leg-length discrepancy by 9 mm, but still remained 12 m m long postoperatively. The m e a n postoperative leg-length discrepancy of the 39 patients w h o had a preoperative discrepancy of 10 m m or more was 1.7 m m compared with 0.1 m m for the 257 patients w h o had a preoperative discrepancy of less than 10 m m (P = .055). Four patients w h o had had unilateral hip replacem e n t (0.9% of the 408 hips) were found to have a peroneal nerve palsy postoperatively, all of which had resolved completely by 6 to 24 months. None of these patients had a postoperative leg-length discrepancy that was greater t h a n 6 mm, and 3 of the 4 had a preoperative radiograph for review. Of the 3 w h o had a preoperative leg-length m e a s u r e m e n t , 1 patient's leg was made longer by 3 mm, 1 patient's leg was shorter by 3 mm, and there was no change in the leg length of the other patient. Twenty-two of these 351 patients (5 %) sustained a postoperative dislocation of 1 hip during followup. The leg length of all 351 patients was classified as long, if the leg was 3 m m or more longer t h a n the reference leg; equal, if it was between +2 and - 2 m m of the reference leg; and short, if it was 3 m m or m o r e shorter than the reference leg. Using this classification of leg-length discrepancy (long, equal, or short), there was no correlation b e t w e e n the leg-length discrepancy of this group of patients and the occurrence of a dislocation (chi square, P = .33).
Discussion A review of the orthopaedic literature for articles on the results of leg-length equalization after primary total hip replacement f o u n d only 6 publications that reported data on postoperative leg-length discrepancies. Five of these 6 studies reported leg lengthening, which averaged b e t w e e n 9 and 16 m m [5-9]. One of these reports found that 144 of 150 extremities were l e n g t h e n e d an average of 16 m m [8]. The remaining study reported an average leg-length discrepancy of 3.4 mm; however, neither the latter study with good results nor the other 5 studies with greater discrepancies described the specific m e t h o d of leg-length equalization that was
Woolson et al.
163
used for the patients in each study. These 5 studies reported results that are clearly suboptimal because an extremity that is more than 1 cm long results in a limp and m a y require a shoe lift on the contralateral side. The results of the m e t h o d of leg-length equalization presented here are superior to all other published studies in that only 12 of 351 patients (3%) had a discrepancy greater than l0 mm, and only 4 of these 351 patients (1%) had leg lengthening that was greater than 10 mm. Statistical analysis determined that there was a trend (P = . 0 5 5 ) f o r patients w h o had a preoperative discrepancy that was greater t h a n 10 m m to have a greater postoperative leg-length difference than for those patients w h o had a preoperative discrepancy less than 10 mm. This trend indicates that this m e t h o d is more accurate for patients w h o have smaller preoperative discrepancies. This technique is ideal for the routine patient w h o has a minimal preoperative leg-length difference and norreal bone anatomy. Patients w h o have severe deformities or major leg-length discrepancies m a y require leg-length equalization by other methods, such as intraoperative m e a s u r e m e n t from pelvic and femoral markers. This simple m e t h o d requires only the use of overlay templates for preoperative planning, a simple arithmatic formula, and a caliper for intraoperative m e a s u r e m e n t . It does not require repeated trials of different neck lengths for adjustment of intraoperative measurements. We believe that the reason this m e t h o d is more accurate than other reported techniques is that the m e a s u r e m e n t for the femoral neck osteotomy is made from the superior aspect of the femoral head rather than from the lesser trochanter. Visualization of the lesser trochanter is limited during primary total hip replacement, and any m e a s u r e m e n t from it is usually made by palpation rather than by using a ruler or caliper. The superior aspect of the femoral head, however, is plainly visible during all surgical approaches. The m e t h o d of intraoperative m e a s u r e m e n t of leg length using a pelvic pin and a femoral marker may also be accurate; however, there have been no reported results of the use of this technique to date. The authors found that the postoperative leg lengths of the 4 patients w h o sustained a transcient peroneal nerve palsy were not altered more than 6 mm, and therefore these nerve injuries were probably not a result of leg lengthening. Excessive leg lengthening, however, can result in sciatic nerve palsy, as pointed out by Williamson and Reckling [81, w h o found that 3% of patients w h o had significant leg lengthening (average 16 mm) developed a nerve palsy.
164
The Journal of Arthroplasty Vol. 14 No. 2 February 1999
We believe that determining leg lengths intraoperatively solely on the basis of soft tissue tensioning by the so-called shuck m e t h o d [14], w h i c h is used by some surgeons to d e t e r m i n e intraoperative leg lengths, m a y frequently cause excessive leg lengthening, as was reported by all studies f o u n d in the literature. This technique, w h i c h encourages the surgeon to use a longer femoral prosthesis neck length to tighten the soft tissues a r o u n d the hip to i m p r o v e the stability of the trial implants, is likely to result in leg lengthening, as all hip reconstructions are m a d e m o r e stable if the hip joint space is lengthened. We believe that excessive leg lengthening is an unacceptable trade-off for increased stability. In contradiction to the rationalization of the t h e o r y that inadequate soft tissue tensioning causes increased postoperative stability, our study f o u n d that there was no increased prevalence of postoperative dislocation in patients w h o had a shorter leg c o m p a r e d with patients w h o had a longer leg or equal leg lengths. Leg-length inequality is a m a j o r cause for malpractice liability suits after total hip replacement, especially w h e n a patient has a longer leg requiring a lift on the contralateral side. Surgeons w h o p e r f o r m total hip r e p l a c e m e n t should be a w a r e that m a n y of the reports in the literature of leg-length results after total hip r e p l a c e m e n t h a v e s h o w n excessive leg lengthening, p r e s u m a b l y f r o m other surgical techniques previously reported, and should adopt a m e t h o d of equalization that is reliable and easy to perform. We believe that because this m e t h o d is simple, is effective, and requires no additional time or cost, it should be strongly considered as an alternative to other techniques that lack data proving their accuracy.
References 1. Muller ME: M.E. Muller straight stem total hip replacement system. Protek Ltd. Manual, Berne, Switzerland, 1982 2. Knight WE: Accurate determination of leg lengths during total hip replacement. Clin Orthop 123:27, 1977 3. McGee HMJ, Scott JHS: A simple method of obtaining equal leg length in total hip arthroplasty. Clin Orthop i94:269, 1985 4. Woolson ST, Harris WH: A method of intraoperative limb length measurement in total hip arthroplasty. Clin Orthop 194:201, 1985 5. Edeen J, Sharkey PF, Alexander AH: Clinical significance of leg-length inequality after total hip arthroplasty. Am J Orthop 24:347, 1995 6. Rand JA, Ilstrup DM: Comparison of Charnley and T-28 total hip arthroplasty. Clin Orthop 180:201, 1983 7. Turula KB, Friberg O, Lindholm TS, et ah Leg length inequality after total hip arthroplasty. Clin Orthop 202:163, 1986 8. Williamson JA, Reckling FW: Limb length discrepancy and related problems following total hip replacement. Clin Orthop 134:135, 1978 9. Woo RYG, Morrey BF: Dislocation after total hip arthroplasty. J Bone Joint Surg Am 64:1295, 1982 10. Woolson ST: Leg length equalization during total hip replacement. Orthopedics 13:17, 1990 11. Ranawat CS, Rodriguez JA: Functional leg-length inequality following total hip arthroplasty. J Arthroplasty 12:359, 1997 12. Love BRT, Wright K: Leg length discrepancy after total hip joint replacement. J Bone Joint Surg Br 65:103, i983 13. Goodman SB, Adler SJ, Fyhrie DP, Schurman DJ: The acetabular teardrop and its relevance to acetabular migration. Clin Orthop 236:199, 1988 14. Charnley J: Low friction arthroplasty of the hip. Springer-Verlag, Berlin, 1979