Cell Biochem Biophys DOI 10.1007/s12013-014-0440-4
ORIGINAL PAPER
Total Hip Arthroplasty: Leg Length Discrepancy Affects Functional Outcomes and Patient’s Gait Yin Zhang • Wenye He • Tao Cheng Xianlong Zhang
•
Ó Springer Science+Business Media New York 2014
Abstract Our study was targeted to investigate the relationship between postoperative leg length discrepancy (LLD) and the outcomes of total hip arthroplasty (THA) including gait, hip function, and lower back pain (LBP). Ninety-two patients who underwent primary THA during Jan 2009 to Apr 2011 in our medical center were enrolled in this study. We measured postoperative LLD of the patients both directly and from the leg radiographs. Six months after the surgery, we evaluated the hip function with Harris Hip Score (HHS), analyzed the gait, and recorded the degree of LBP. Patients with LLD between 10 and 20 mm were given the block footbeds to correct the difference in the leg length. Then 1 year after the surgery, the above-mentioned parameters were evaluated again. (1) LLD: Patients were divided into three groups according to the magnitude: less than 10 mm were grouped as A, those between 10 and 20 mm as B, and more than 20 mm as C. (2) Gait analysis: patients with larger LLD showed slower gait speed (t = 6.527; p \ 0.01), longer single support time (t = -2.665; p = 0.01), and shorter foot-off time (t = -8.502; p \ 0.01). After half a year of the surgery, Group B patients showed recovery and their functional performance was not significantly different from that of the patients in Group A (t = -0.686; p = 0.49). (3) HHS: In the first half of the follow-up year, patients with smaller LLD showed an improved function (t = 6.56; p \ 0.01). At the end of year one, the HHS of Group B patients was
Yin Zhang and Wenye He have contributed equally to this work. Y. Zhang W. He T. Cheng X. Zhang (&) Department of Orthopedics, Shanghai Jiao Tong University Affiliated Sixth People’s Hospital, Shanghai 200233, People’s Republic of China e-mail:
[email protected] not significantly different from that of Group A (t = 1.4; p = 0.16), suggesting a good recovery, however, Group C patients showed no improvement in HHS scores. (4) LBP: Initially, Group B patients suffered from more severe LBP than Group A patients, however, the examination conducted in the end of year 1 exhibited no difference in the LBP levels of the two groups (t = 0.683; p = 0.01, t = -0.85; p = 0.40). After THA surgery, the outcome was found to be significantly associated with the degree of different length of lower limbs. The use of block such as footbed could partly relieve the symptoms of hip dysfunction and LBP. However, the recovery was less significant in patients with leg length difference more than 20 mm. Keywords Arthroplasty Complications of hip joint surgery Functional recovery Neck of femur fracture Lower limbs length Lower back pain Gait analysis
Introduction Total hip arthroplasty (THA) is one of the most effective treatments for the severe hip diseases such as traumatic hip arthritis and avascular necrosis of the femoral head (ANFH). The treatment is meant to repair the defect in anatomical morphology and restore the function of hip joint [1]. However, the leg length discrepancy (LLD) is often a troublesome complication that is difficult to avoid completely even with most precise surgical methods. According to American Hip Society, LLD has been ranked as the 2nd most common reason of medical lawsuits after the THA in America. The post-surgery LLD often leads to low back pain, abnormal gait, and nerve palsies [2]. As a common complication of THA, LLD has become an important criterion for the evaluation of THA surgery.
123
Cell Biochem Biophys
Preoperative planning has been considered to play an important role in preventing post THA complications [3]. Postoperative LLD can be caused by a variety of reasons such as scoliosis, improper offset, and prosthesis size [4]. Local anesthesia in comparison with general anesthesia has been indicated as a more likely factor of determining LLD [5]. The operation errors, which are often difficult to correct include stretching of the peripheral nerve, and Crown IV developmental dysplasia of the hip. After recovery from anesthesia, the patients are mostly concerned if their leg length has been correctly fixed. Their concern is reasonable as LLD could lead to back pain; nerve damage caused paralysis and crippling leading to decreased satisfaction [6, 7]. A slight LLD can usually be remedied by raising the shorter leg using a shoe lifting block, however, the larger LLD needs a revision surgery to readjust the leg length. It remains controversial to date as to how much postoperative LLD is acceptable and the curative effectiveness of shoe lifting is also debatable. The purpose of this followup study was to determine a relatively safe level of LLD and to find out the therapeutic effectiveness of shoe lifting.
Material and Method Subjects This follow-up study was carried out on 92 patients (48 female, 44 male) who underwent unilateral THA in our medical center (Shanghai 6th people’s hospital) between January 2009 and April 2011. Their mean age was 63.1 years (ranged from 59 to 84 years). The unilateral THA was performed on the right hip of 50 and on the left hip of 42 patients. The criteria for including the subjects in this study were: (1) primary arthritis was diagnosed by more than two chief physicians based on the diagnostic standards provided by American Rheumatism Association; (2) based on both clinical and imaging examination, no osteoarthritis in contralateral hip joint or other lower extremity joints existed; (3) pre-surgery LLD was less than 2 cm; (4) without any systemic disease, such as severe cardiorespiratory disorder and poorly controlled diabetes. All the surgeries were conducted by a group of surgeons using standard posterior-lateral approach. An informed consent of the participating patients was obtained.
To avoid postoperative dislocation, the prostheses were implanted with a tension in the gluteus medius. As a result, the leg in the reconstructed side was lengthened. Patients with LLD of less than 10 mm were grouped as A, those with 10–20 mm as B, and with more than 20 mm as C. Gait Analysis We analyzed the gait of all the patients using the VICON612 3D system comprised four components: the infrared reflective ball, six far infrared cameras (working frequency in 120 Hz), A/D converter controller, and a computer with analysis software. This workstation was also connected with a 16-channel Kistler 9286 force measure platform to collect kinematic and dynamical parameters synchronously. The instrument was routinely calibrated for the required parameters every morning before conducting the formal tests. A total of 16 reflective markers were separately placed on the anterior–superior iliac spine, posterior–superior iliac spine, 15 cm above lateral femoral condyle, lateral femoral condyle, 15 cm above lateral ankle point, lateral ankle point, between the first and the second tarsometatarsal joint and tendo calcaneus on both sides. After making the punch mark the patients were allowed to walk adaptively on the force measure platforms in the middle of a walkway. The test was considered as a successful record when both feet contacted the two force plates while walking through the walkway. For every patient, 6–8 gait tests were recorded by an experienced surgeon and then averaged. The step frequency, single support, and foot off were used as outcome indices. Evaluation of Therapeutic Efficiency The functional hip recovery and patient satisfaction were measured as HHS whereas the level of lower back pain was evaluated by visual analog scale (VAS) of pain score. These parameters were recorded at 6 and 12 months after the THA. After the 6 months of examination, the patients in Groups B and C were recommended to use a block as a footbed. Statistics The data were analyzed by the SPSS 22.0 program. T test, both independent and paired, was used to compare the means of two groups. The difference was considered significant when p was \ 0.05.
Measurement of the Leg Length and Grouping Length of the whole leg was measured directly from anterior–superior iliac spine to the center point of ankle (accurate to 1 mm) with a tape and also on X-ray image of the leg as mentioned by Sabharwal [8]. The LLDs measured by these two methods were averaged to obtain the true value of LLD.
123
Results Overall, 1 year follow-up showed good results. None of the prostheses were found to be dislocated or loosened. No fatal complication occurred during the follow-up period.
Cell Biochem Biophys
Gait Analysis We used gait speed, single support, and foot off as indicators of the gait performance. The results are shown in Table 1. At the half year of follow-up, the smaller LLD was found to be associated with a faster gait (Groups A and B t = 6.527; p \ 0.01, Group C was numerically slower than Group B). Patients who had smaller postoperative LLD and used footbed were evidenced to walk faster at the end of 1 year follow-up examination (Group A and B t = 2.526; p = 0.01, Group C was numerically slower than Group B). At the follow-up of 1 year, all of these patients walked faster than they had in the half year follow-up examination (Group A t = -6.922; p \ 0.01; Group B: t = 16.3; p \ 0.01). Group B patients showed more improvement in the gait speed than Group A patients (t = -2.92; p \ 0.01). However, Group C patients showed less improvement in the numerical gait frequency than Group B patients. In the half year follow-up, the leg on the THA side showed a longer support in the patients with smaller LLD (Group A–B t = 2.665; p = 0.01, Group C was numerically longer than Group B), whereas in the 1 year followup, this situation was improved (Group A–B t = -0.686; p = 0.49). However, for the patients in Group C, it remained numerically longer than Group B but within 95 % Cl. In the half year follow-up, foot-off time of the leg on the reconstructed side hip was shorter in patients with smaller LLD (Group A–B t = -8.502; p \ 0.01), whereas in patients of Group C, it was numerically longer than that of Group B. This parameter was not found to change in the follow-up examination conducted at the end of 1 year (Group A–B t = 2.526; p = 0.01, and Group C was numerically longer than Group B). Harris Hip Score (HHS) Functional recovery of hip joint was evaluated by the HHS. The results are shown in Table 2. In the follow-up of half year, patients with smaller LLD showed better hip function (Group A–B t = 6.56;
Table 1 Gait analysis after 6 months and 1 year follow-up
Group Group A Group B
Group A LLD of less than 10 mm; Group B LLD 10–20 mm; Group C LLD more than 20 mm
Group C
Follow-up period
p \ 0.01) and by the end of 1 year it was completely normalized (Group A–B t = 1.4; p = 0.16). The recovery of Group B patients was partly attributed to the use of footbed (Group B: t = -9.95; p \ 0.01), Group C patients were numerically worse than Group C individuals at both half a year and one year follow-ups. The HHS of Group A patients was stable through the first and second half of the year (Group A: t = -1.35; p = 0.183). Lower Back Pain (LBP) Lower back pain occurs quite often after the surgery and it lasts for a long time. We evaluated this parameter using the VAS score. The results are shown in Table 3. In the first follow-up at half a year of the surgery, patients with larger LLD showed severer LBP (Group A–B t = 0.683; p = 0.01, Group C was even more severe than Group B). In the second half of the year, Group B patients showed a complete relief from the pain and the score was not different from that of Group A (Group A–B: t = -0.85; p = 0.40). However, Group C patients continued to show higher LBP scores.
Discussion Leg length discrepancy is one of the most common complications of THA surgery and it has been known to cause back pain, sciatica, gait abnormality, and hip dislocation [9]. It may also result in decreased satisfaction of the patients and influence the functional and gait recoveries. Studies investigating the relationship between the THA efficacy and LLD in the long-term follow-up based on large sample size have rarely been conducted. We found that limb length difference and the gait parameters determined using gait analysis method were correlated. The results showed that postoperative LLD significantly affected the gait. Postoperative LLD has been extensively studied, however, the magnitude of discrepancy acceptable in terms of functional outcomes remains controversial. Austin et al. [10] suggested that LLD of less than 7 mm to be
Step frequency (step/second)
Single support (second)
Foot off (second)
6 month
0.993 ± 0.056
1.030 ± 0.082
0.990 ± 0.413
1 year
1.031 ± 0.066
0.987 ± 0.094
0.960 ± 0.079
6 month
0.930 ± 0.034
1.086 ± 0.113
1.068 ± 0.451
1 year
0.996 ± 0.067
1.000 ± 0.086
1.017 ± 0.110
6 month 1 year
0.91 and 0.89 0.95 and 0.92
1.10 and 1.12 1.08 and 1.10
1.10 and 1.13 1.03 and 1.07
123
Cell Biochem Biophys Table 2 HHS after 6 months and 1 year follow-up Group Group A Group B Group C
Follow-up period
HHS
6 month
81.2 ± 6.5
1 year
82.1 ± 6.5
6 month
73.3 ± 4.8
1 year
80.4 ± 4.7
6 month
71 and 69
1 year
75 and 74
Group A LLD of less than 10 mm; Group B LLD 10–20 mm; Group C LLD more than 20 mm
Table 3 LBP after 6 months and 1 year follow-up Group Group A Group B Group C
Follow-up period
LBP
6 month
4.04 ± 1.93
1 year
3.58 ± 1.57
6 month
5.13 ± 1.99
1 year
3.90 ± 1.93
6 month
7.9 and 5.1
1 year
6.5 and 5.0
Group A LLD of less than 10 mm; Group B LLD 10–20 mm; Group C LLD more than 20 mm
acceptable, however, they emphasized more on stability of the joint rather than the length inequality. They suggested that, informing early of the problem can avoid post-surgical LLD complication. In a study based on 149 cases, Wylde et al. [11] found that 1/3 of the patients were identified with LLD and they performed poorly in the mid-term follow-up. May be they were too cautious, as other researchers thought the influence wasn’t that obvious. Edeen et al. [12] showed that 32 % of their patients showed leg length inequality (mean of 14.9 mm), and only half of them thought it to be unacceptable. White et al. [11] did not find a connection between the LLD and dissatisfaction. Maloney et al. [6] also considered LLD of less than 10 mm as acceptable. Della Valle [13] thought that one of the major concerns to both the surgeon and the patient after THA, is the preservation of leg length equality and an inequality of more than 10 mm increased the risk of loosening, and led to an abnormal gait, obliquity of pelvis, and patient satisfaction. In our gait analysis, patients with smaller LLD were found to exhibit faster gait speed. Moreover, by the end of a year, all of these patients walked faster indicating recovery, however, differing in degree. After the surgery, Group B patients walked slower than those of Group A in the first follow-up of half a year, but their recovery was improved in the second half of the year. On the contrary, Group C patients showed a limited recovery. According to the trend, it is expected that the gap between Group A and
123
B will be minimized over the next few years. However, it’s not certain whether the Group C patients would catch up. Similar results were obtained in single support time and foot off. It seems that patients with 10–20 mm LLD were partially recovered and their gait was affected. The LLD of more than 20 mm, resulted in slower gait recovery even when footbed was used in the second half of the year, and the hip joint function was also affected. In this study, we found that the abnormal gait caused by LLD could partly be corrected by balancing the leg length with a shoe lift. While an inequality of more than 20 mm slowed the speed and reduced the step length; the patients who underwent leg length equalization in addition to THA showed a faster gait and improved bilateral symmetry [14]. Although the gait can be normalized by balancing the leg length inequality with a footbed, this strategy alone does not seem to be a complete panacea. An artificial leg length inequality of 2 cm created by shoe lift has been shown to cause a significant increase in oxygen consumption and other physiological parameters. In particular, the elderly patients with a limb inequality of 2 cm have been found to experience difficulty in walking and a marked compromise of pulmonary, cardiac, and neuromuscular function [15]. Despite the unaffected HHS and patient satisfaction, small differences in the leg length have been reported to be negatively associated with limping and pain [11, 16]. However, rehabilitation has been considered to improve the debilitation. We observed that HHS of the patients was significantly related to the magnitude of LLD. Whereas Group A patients showed obvious strength in the first follow-up of half a year, Group B patients were found to catch up in the second half of the follow-up. In agreement with other published reports, patients with more than 20 mm LLD (Group C) performed poorly throughout the follow-up period. Lower back pain is another outcome of LLD caused by the compensated pelvis positioning under the stress of unbalanced posture. A limb length discrepancy of 5.5 % has been suggested to increase the mechanical work of the longer limb leading to a vertical displacement of body’s center of gravity [17]. In order to equalize the functional leg lengths and to deviate altered center of gravity, dropping of the pelvis with leaning of the ipsilateral position of the trunk occurs as a compensatory mechanism. The force of altered pelvic position is finally conduced to the lumbus inevitably causing LBP. However, the pain has been shown to be released by lifting up the smaller leg using a shoe lift [17]. In this study, patients from Groups B and C suffered from severe LBP in the first half of the year (VAS [ 5). Group B patients recovered in the second half of the year and their VAS scores were found to be close to those of Group A patients. Group C patients however, continued to complain of LBP and had higher VAS scores.
Cell Biochem Biophys
Our observations suggest that LLD larger than 20 mm is unacceptable and it results from problems in prosthesis alignment or bone cutting. Pelvic compensation mostly appeared in the first half year. The symptoms experienced by the Group C patients indicate that LLD more than 20 mm cannot be compensated and it leads to long-lasting LBP. The only solution to this problem is to use the footbed and lengthen the shorter calf. However, the postoperative LLD mostly affects the length of thigh and a 2 cm difference in knee height might feel uncomfortable. The limitations in our study are that we did not have postoperative pelvic plain films for all the patients. Therefore, we only compared the absolute value of LLD whereas postoperative LLD might have occurred due to the compensation of pelvic and lumbar obliquity. Through comparison, we found that taller patients had much longer legs than shorter patients. The magnitude of difference could even reach to 1.5 fold accounting for a big discrepancy in the compensatory amplitude. We also reckon that if percentage of the lower limb extension is taken into account, results can be more accurate. In conclusion, the results of this follow-up study showed that the outcome of THA surgery was significantly associated with equality of leg lengthening. The use of footbed could partially correct the gait, improve the functions, and release LBP. Whereas, the leg length inequality of more than 20 mm was a serious hurdle in recovery, LLD smaller than 20 mm could be managed and therefore it was acceptable.
References 1. Siopack, J., & Jergesen, H. (1995). Total hip arthroplasty. Western Journal of Medicine, 162, 243. 2. Upadhyay, A., York, S., Macaulay, W., McGrory, B., Robbennolt, J., & Bal, B. S. (2007). Medical malpractice in hip and knee arthroplasty. The Journal of Arthroplasty, 22(2–7), e4. 3. Della Valle, A. G., Padgett, D. E., & Salvati, E. A. (2005). Preoperative planning for primary total hip arthroplasty. Journal of the American Academy of Orthopaedic Surgeons, 13, 455–462.
4. Gurney, B. (2002). Leg length discrepancy. Gait & Posture, 15, 195–206. 5. Sathappan, S. S., Ginat, D., Patel, V., Walsh, M., Jaffe, W. L., & Di Cesare, P. E. (2008). Effect of anesthesia type on limb length discrepancy after total hip arthroplasty. The Journal of Arthroplasty, 23, 203–209. 6. Maloney, W. J., & Keeney, J. A. (2004). Leg length discrepancy after total hip arthroplasty. The Journal of Arthroplasty, 19, 108–110. 7. Clark, C. R., Huddleston, H. D., Schoch, E. P., & Thomas, B. J. (2006). Leg-length discrepancy after total hip arthroplasty. Journal of the American Academy of Orthopaedic Surgeons, 14, 38–45. 8. Sabharwal, S., & Kumar, A. (2008). Methods for assessing leg length discrepancy. Clinical Orthopaedics and Related Research, 466, 2910–2922. 9. Konyves, A., & Bannister, G. (2005). The importance of leg length discrepancy after total hip arthroplasty. Journal of Bone & Joint Surgery, British Volume, 87, 155–157. 10. Austin, M. S., Hozack, W. J., Sharkey, P. F., & Rothman, R. H. (2003). Stability and leg length equality in total hip arthroplasty. The Journal of Arthroplasty, 18, 88–90. 11. Wylde, V., Whitehouse, S. L., Taylor, A. H., Pattison, G. T., Bannister, G. C., & Blom, A. W. (2008). Prevalence and functional impact of patient-perceived leg length discrepancy after hip replacement. International Orthopaedics, 24, 210–216. 12. Edeen, J., Sharkey, P., & Alexander, A. (1995). Clinical significance of leg-length inequality after total hip arthroplasty. American Journal of Orthopedics, 24, 347–351. 13. Della Valle, C. J., & Di Cesare, P. E. (2001). Complications of total hip arthroplasty: neurovascular injury, leg-length discrepancy, and instability. Bulletin of Hospital Joint Disease, 60, 134–142. 14. Lai, K. A., Lin, C. J., Jou, I., & Su, F. C. (2001). Gait analysis after total hip arthroplasty with leg-length equalization in women with unilateral congenital complete dislocation of the hip–comparison with untreated patients. Journal of Orthopaedic Research, 19, 1147–1152. 15. Gurney, B., Mermier, C., Robergs, R., Gibson, A., & Rivero, D. (2001). Effects of limb-length discrepancy on gait economy and lower-extremity muscle activity in older adults. The Journal of Bone and Joint Surgery, 83, 907–915. 16. Plaass, C., Clauss, M., Ochsner, P. E., & Ilchmann, T. (2010). Influence of leg length discrepancy on clinical results after total hip arthroplasty—a prospective clinical trial. Hip International, 21, 441–449. 17. Bhave, A., Paley, D., & Herzenberg, J. E. (1999). Improvement in gait parameters after lengthening for the treatment of limb-length discrepancy. The Journal of Bone and Joint Surgery, 81, 529–534.
123