Loading...
Visit our Evidence-Based Covid-19 Website and Stay Up to Date with the latest Research.
Volume:4 Issue:4 Number:1 ISSN#:2563-559X
OE Original

Full Weight Bearing Immediately After Uncemented Total Hip Arthroplasty: A Systematic Review of Randomized Trials

Authored By: OrthoEvidence

April 5, 2021

How to Cite

OrthoEvidence. Full Weight Bearing Immediately After Uncemented Total Hip Arthroplasty: A Systematic Review of Randomized Trials. OE Original. 2021;4(4):1. Available from: https://myorthoevidene.com/Blog/Show/122


Highlights


  • - Controversy exists over the application of immediate full (FWB) vs. partial weight bearing (PWB) in patients who had uncemented total hip arthroplasty (THA). Opponents against FWB were concerned that FWB might negatively impact the initial stability of the uncemented prostheses and bone ingrowth. Advocates, on the other hand, believed that FWB would lead to relatively fast functional recovery, whereas PWB may slow down the rehabilitative process, pose stresses on upper extremities and the contralateral hip, and increase the risk of muscle atrophy.

  • - From Ovid MEDLINE, Ovid Embase, and Cochrane Controlled Register of Trials (CENTRAL), 13 eligible randomized clinical trials (RCTs), which compared FWB vs. PWB immediately following uncemented THA, were identified. No new RCTs were published in the past three years: 11 were published before 2012; and the other two in 2014 and 2017, respectively.

  • - Our meta-analysis suggested that FWB immediately after uncemented THA neither significantly improved nor worsened most of the clinical outcomes, such as the Harris hip score (HHS), the Merle d’Aubigne hip score, a normalized functional score, and pain.

  • - Our narrative summary of the safety outcomes (i.e., femoral subsidence and prosthesis loosening) showed no significant differences between patients who received FWB vs. PWB immediately after THA.

  • - Using the GRADE approach, the quality of current RCT evidence was rated as very low due to issues in the risk of bias, inconsistency, and imprecision domains. Very low quality suggests that current evidence is insufficient to permit any conclusion. Future trials with a high-quality methodological design and large sample size are required.




Total hip arthroplasty (THA) can be broadly divided into cemented and uncemented THA. Cemented THA uses polymethylmethacrylate, which functions as a grout, to help affix the joint prosthesis to the bone, while uncemented THA uses specially textured joint prosthesis to allow the bone to grow onto and adhere to the prosthesis (Maggs et al., 2017). Although there is an ongoing debate on which approach is more beneficial to patients, uncemented THA has become a popular treatment option in countries such as the United States (USA) and Canada (Zhang et al., 2017).


Weight bearing exercise has been considered as a crucial component of the rehabilitation following THA (Di Monaco et al., 2009). Postoperative weight bearing can be full (FWB, also called unrestricted) or partial (PWB, can also be called protected or restricted) weight bearing. Controversy exists over FWB and PWB immediately following THA. Advocates for FWB believe that FWB would be beneficial toward the goals of relatively fast functional recovery, while PWB after uncemented THA may slow the rehabilitative process, pose stresses on upper extremities and contralateral hip, and increase the risk of muscle atrophy (Hol et al., 2010; Rao et al., 1998). Opponents are concerned that FWB might negatively affect the initial stability of the uncemented prostheses and bone ingrowth, which are key to patients’ recovery from THA.


Systematic reviews have been conducted in the past to address the issue between FWB and PWB following uncemented THA (Hol et al., 2010; Tian et al., 2017). Hol et al., (2010) narratively synthesized evidence and recommended the use of FWB as soon as possible after uncemented THA. Tian et al. (2017) conducted a meta-analysis, which mainly focused on the safety outcomes such as femoral subsidence and prosthesis loosening. Tian et al. (2017) concluded that immediate FWB post uncemented THA was a safe treatment.


In this OE Original, we conduct an up-to-date systematic review and meta-analysis to examine evidence from RCTs with our major focus on clinical outcomes (e.g., pain and function) as well as safety outcomes between FWB vs. PWB immediately after uncemented THA.



Methods


We searched OrthoEvidence, Ovid MEDLINE, Ovid Embase, and Cochrane Controlled Register of Trials (CENTRAL) from inception to March 22nd, 2021 for RCTs published in English and comparing the clinical outcomes (i.e., patient-reported function or pain outcomes) and safety outcomes (i.e., femoral subsidence, loosening, or other adverse events) between patients receiving FWB immediately following uncemented THA vs. those who received PWB. Only studies with full texts were included. We searched both indexed terms and free text terms relating to total hip arthroplasty and weight bearing. We adopted the Cochrane risk-of-bias tool and the GRADE approach to determine the risk of bias (RoB) and the quality of evidence for included RCTs, respectively. Two reviewers independently worked on the study screening and selection processes.




Results


1. Characteristics of included RCTs


In total, 259 records were retrieved, among which 14 RCTs were eligible and therefore included (Andersson et al., 2001; Bodén et al., 2004; Bottner et al., 2005; Kishida et al., 2001; Markmiller et al., 2011; Monticone et al., 2014; Shabana et al., 2017; Ström et al., 2006; Ström et al., 2007a, 2007b; Thien et al., 2007; Unver et al., 2004; Wolf et al., 2010; Wolf et al., 2012). The characteristics of the included RCTs were presented in Table 1. The sample size of each included RCT was small. Most RCTs were published before 2012.


In terms of RoB assessment (Figure 1), none of the included RCTs could blind participants and providers due to the nature of the intervention. Most of the RCTs did not provide sufficient information to determine whether outcome assessors were blinded or there was selective reporting. There were also some concerns about the randomization and allocation process among 25% of the included RCTs.


Table 1. Characteristics of included RCTs

Study ID

Country

Sample Size (Hips)

Condition

Age (years)

Intervention

Control

Andersson et al. (2001)

Sweden

24

Patients with unilateral primary arthrosis undergoing uncemented THA*

Intervention: Mean: 55, Range: 44 to 63; Comparator: Mean: 54, Range: 44 to 59

Full weight bearing immediately after THA

Partial weight bearing for 12 weeks immediately after THA (< 10% body weight)

Boden et al. (2004)

Sweden

23

Patients with unilateral primary arthrosis undergoing uncemented THA

Intervention: Mean: 54, Range: 44 to 59; Comparator: Mean: 55, Range: 44 to 63

Full weight bearing immediately after THA

Partial weight bearing for 12 weeks immediately after THA (< 10% body weight)

Bottner et al. (2005)

United States

32 (41)

Patients with hip osteoarthritis undergoing uncemented THA

Mean: 47, Range: 24 to 59

Full weight bearing immediately after THA

Toe-touching weight bearing for 6 weeks immediately after THA and then full weight bearing

Kishida et al. (2001)

Japan

33 (37)

Patients undergoing uncemented THA

Intervention: Mean: 52, SD**: 13; Comparator: Mean: 51, SD: 12

Full weight bearing immediately after THA

No weight bearing for 3 weeks immediately after THA, then partial weight bearing for 3 weeks, then full weight bearing

Markmiller et al. (2011)

Germany

100

Patients with osteoarthritis undergoing uncemented THA

Intervention: Mean: 61.2, SD: 13.1; Comparator: Mean: 60.6, SD: 12.5

Full weight bearing immediately after THA

Partial weight bearing for 6 weeks immediately after THA (15 kg) then full weight bearing

Monticone et al. (2014)

Italy

100

patients with osteoarthritis underwent uncemented THA

Mean: 69, SD: 8

Full weight bearing immediately after THA + task-oriented exercises

Partial weight-bearing within 3 months immediately after THA + open chain kinetic exercises

Shabana et al. (2017)

Egypt

20

Patients undergoing uncemented THA

Mean: 57.5, Range: 50 to 65

Full weight bearing immediately after THA

Partial weight bearing immediately after THA and then full weight bearing

Ström et al. (2006)
Sweden

36

Patients with osteoarthritis undergoing uncemented THA

Mean: 54.3, Range: 25 to 63

Full weight bearing immediately after THA including intensive training

Partial weight bearing for 3 months including short self-training program

Ström et al. (2007a)

Sweden

29

patients with osteoarthritis underwent uncemented THA

Mean: 55, Range: 26 to 63

Full weight bearing immediately after THA including intensive training

Partial weight bearing (15 kg) for 3 months immediately after THA + conservative training

Ström et al. (2007b)

Sweden

43

Patients with osteoarthritis undergoing uncemented THA

Intervention: Mean: 54.5; Comparator: Mean: 55.6

Full weight bearing immediately after THA including intensive training

Partial weight bearing (15 kg) for 3 months immediately after THA + self-training

Thien et al. (2007)

Sweden

43

Patients undergoing uncemented THA

Mean: 53, Range: 41 to 63

Full weight bearing immediately after THA

Partial weight bearing for 6 weeks immediately after THA

Unver et al. (2004)

Turkey

51 (60)

Patients undergoing uncemented THA

Intervention: Mean: 49.9, SD: 10.0; Comparator: Mean: 48.9, SD: 12.9

Accelerated rehabilitation with full weight bearing

Accelerated rehabilitation with partial weight bearing

Wolf et al. (2010)

Sweden

42

Patients with osteoarthritis undergoing uncemented THA

Mean: 54, Range: 25 to 63

Full weight bearing immediately after THA including intensive training

Partial weight bearing for 3 months immediately after THA  including restricted rehabilitation program

Wolf et al. (2012)

Sweden

43

Patients with osteoarthritis undergoing uncemented THA

Mean: 54, SD: 10

Full weight bearing immediately after THA including intensive training

Partial weight bearing for 3 months immediately after THA  including restricted rehabilitation program

* THA: total hip arthroplasty; ** SD: standard deviation












2. Meta-analysis


2.1 Clinical outcomes


2.1.1 Harris hip score (HHS)


The HHS is used to evaluate pain, activity, and function of patients following THA. The score of HHS ranges between 0 and 100. A higher score indicates a better outcome.


Four included RCTs reported the HHS score (Bodén et al., 2004; Bottner et al., 2005; Shabana et al., 2017; Unver et al., 2004). No significant differences were found in the HHS score at six weeks, three months, and two years postoperatively between patients who received FWB immediately after THA and those who had PWB (Figure 2).






2.1.2 Merle d'Aubigne hip score


The Merle d’Aubigne hip score is a score used to assess the functional outcomes of patients following THA. The range of the Merle d’Aubigne hip score is between 0 and 18. A higher score indicates a better outcome.


Only two RCTs (Kishida et al., 2001; Markmiller et al., 2011) reported the total Merle d’Aubigne hip score, and one additional RCT (Ström et al., 2006) reported the subscale scores of the Merle d’Aubigne hip score.


Kishida et al. (2001) found that FWB immediately post THA led to a significant improvement in the total Merle d’Aubigne hip score at six months [mean difference (MD): 0.50, 95% confidence interval (CI): 0.09 to 0.91] postoperatively, compared to patients who had PWB (Figure 3).


No significant differences were identified in the total Merle d’Aubigne hip score at three months and two to five years postoperatively between patients who received FWB immediately after THA and those who received PWB (Figure 3).


Ström et al. (2006), which examined the subscale scores of the Merle d’Aubigne hip score including pain, walking ability, and range of motion, found no significant difference between the intervention group and the comparison.





2.1.3 Patient-reported efficacy outcomes on a normalized scale


We also incorporated the HHS and the Merle D’Aubigne hip score to a normalized scale (0 - 100, a higher score indicates a better outcome).


We found no statistically significant differences in the normalized scale between patients receiving FWB immediately following THA and those who received PWB at three months and two to five years postoperatively (Figure 4).




2.1.4 Pain on a normalized scale


Three included RCTs reported pain either on a visual analogue scale (VAS) or numeric rating scale (NRS) (Andersson et al., 2001; Markmiller et al., 2011; Monticone et al., 2014). We converted them to a normalized scale with a range of 0 to 100. A higher score on the normalized scale indicates worse pain.


As shown in Figure 5, there were no significant differences in pain between the intervention group and the comparison at three months, six months, and twelve months after THA.





2.2 Safety outcomes


2.2.1 Femoral subsidence


Three RCTs reported femoral subsidence as a dichotomous variable (Bodén et al., 2004; Kishida et al., 2001; Markmiller et al., 2011). Bodén et al. (2004) and Markmiller et al. (2011) defined subsidence as a migration of 5 mm or more in the vertical direction, while Kishida et al. (2001) considered a migration of 4 mm as definite subsidence. Results from the three RCTs showed that no subsidence that met the definitions occurred neither in the short term (e.g., three months) nor in the long term (e.g., five years) after THA (Bodén et al., 2004; Kishida et al., 2001; Markmiller et al., 2011).


Four included RCTs reported femoral subsidence as a continuous outcome (Ström et al., 2007a, 2007b; Thien et al., 2007; Wolf et al., 2010). However, we were not able to conduct meta-analysis on subsidence (continuous outcome) because we could not rule out the possibility that Ström et al. (2007a), (2007b), and Wolf et al. (2010) might overlap in the research population. As a result, we provided a narrative summary.


Results from Ström et al. (2007a), (2007b), and Wolf et al. (2010) all showed no significant differences in subsidence between the early weight bearing group vs the comparator up to five years postoperatively.


Thien et al. (2017) found that the mean subsidence [proximal (+) or distal (–)] in the FWB and PWB groups at three months post THA was -0.31 mm (95% CI: -0.34 to 0.67) and -0.14 mm (95% CI: -0.33 to 0.67), respectively. At one year after THA, the mean subsidence in the FWB and PWB groups was -0.28 mm (95% CI: -0.37 to 0.58) and -0.17 mm (95% CI: -0.38 to 0.59), respectively. There was no statistical significance between the FWB and PWB groups at either follow-up points.


2.2.2 Loosening


Loosening refers to the migration or breakage of the prosthesis. Five included RCTs reported loosening (Bottner et al., 2005; Ström et al., 2007a, 2007b; Wolf et al., 2010; Wolf et al., 2012). No loosening happened in the RCT conducted by Bottner et al. (2005).


Aseptic loosening occurred in one patient in the FWB group at 1.5 years post THA was reported by Ström et al. (2007a), (2007b), Wolf et al. (2010), and (2012). Again, we were not able to rule out the possibility that they reported the same incident.


Both Wolf et al. (2010) and (2012) reported one cup revision due to loosening in the PWB group within 20 months post THA.


Discussion


In this OE Original, we conducted an up-to-date systematic review and meta-analysis to examine current available evidence from randomized controlled trials (RCTs) comparing full weight bearing (FWB) vs. partial weight bearing (PWB) immediately after uncemented total hip arthroplasty (THA). In addition to outcomes (i.e., subsidence and loosening) synthesized in the past systematic reviews (Hol et al., 2010; P. Tian et al., 2017), we focused more on the synthesis of clinical outcomes, including the Harris hip score (HHS), the Merle d’Aubigne hip score, a normalized score incorporating HHS and the Merle d’Aubigne hip score, and a normalized pain scale incorporating the visual analogue scale (VAS) and the numeric rating scale (NRS). The summary of our meta-analysis was presented in Table 2.


Overall, our meta-analysis found no statistically significant differences in most of the clinical outcomes except that immediate FWB after THA resulted in a significant improvement in the Merle d’Aubigne hip score at 6 months (MD: 0.50, 95% CI: 0.09 to 0.91), compared to patients who had PWB.







Table 2. Summary of results

Outcome

Statistical Significance

Quality of Evidence

HHS (6 weeks after THA)

No

Very Low

HHS (3 months after THA)

No

Very Low

HHS (2 years after THA)

No

Very Low

Merle d'Aubigne Hip Score (3 months after THA)

No

Very Low

Merle d'Aubigne Hip Score (6 months after THA)

Yes, MD: 0.50 (95% CI: 0.09 to 0.91), favoring FWB

Very Low

Merle d'Aubigne Hip Score (2 to 5 years after THA)

No

Very Low

Efficacy outcome on a normalized scale (3 months after THA)

No

Very Low

Efficacy outcome on a normalized scale (2 to 5 years after THA)

No

Very Low

Pain (= 3 months after THA

No

Very Low

Pain (6 months after THA)

No

Very Low

Pain (= 12 months after THA)

No

Very Low


HHS: Harris hip score; THA: total hip arthroplasty; FWB: full weight bearing; MD: mean difference; 95% CI: 95% confidence interval




A certain conclusion, however, could not be drawn from the clinical outcomes due to the very low quality of evidence in each clinical outcome we meta-analyzed (Table 2). The major concerns were from risk of bias (RoB), inconsistency, and imprecision domains. For instance, as we described previously for RoB assessment (Figure 1), none of the RCTs blinded participants, and there were concerns regarding the randomization and allocation process, blinding outcome assessors, and selective reporting in a considerable proportion of included RCTs.


In terms of inconsistency, for example, both Shabana et al. (2017) and Unver et al. (2004) reported HHS at 3 months post THA (Figure 2). However, their point estimates were located in a completely opposite direction with Shabana et al. (2017) favoring PWB and Unver et al. (2004) favoring FWB. There was also no overlap between the 95% CIs, indicating a very high degree of inconsistency between the two RCTs.


Imprecision was a significant concern in all of the included studies due to 95% CI crossing the no effect line or sample size not meeting the threshold of the optimal information size (OIS), which should be over 400 for a continuous outcome (Guyatt et al., 2011). For details about how to assess and interpret the quality of evidence, please see our past OE Original: Grading a Body of Evidence: How certain are we in the results? Our evaluation of the quality of evidence strongly warranted that more RCTs with higher methodological quality and larger sample sizes were urgently needed.


In this OE Original, we only narratively synthesized safety outcomes (i.e., subsidence and loosening) because we were not able to rule out the possibility that there was a high overlap in the research population among the RCTs which reported femoral subsidence and prosthesis loosening. Given this context, it might be misleading if we provide a quantitative synthesis.


Our narrative summary found that there were no statistically significant differences in femoral subsidence (see section 2.2.1) and loosening (see section 2.2.2) between FWB and PWB in patients who had uncemented THA, corroborating the conclusions from past systematic reviews (Hol et al., 2010; Tian et al., 2017). However, more RCTs investigating these safety outcomes are required.


Bottom Line


The present up-to-date systematic review and meta-analysis suggested that full weight bearing immediately after uncemented total hip arthroplasty, as opposed to partial weight bearing, neither significantly improved nor worsened the clinical outcomes (i.e., the Harris hip score, the Merle d’Aubigne hip score, a normalized score incorporating the Harris hip score and the Merle d’Aubigne hip score, and pain) or safety outcomes (i.e., femoral subsidence and prosthesis loosening).


Following the GRADE approach, we rated the quality of current RCT evidence as very low due to the issues in the risk of bias, inconsistency, and imprecision domains. As a result, we considered current evidence was insufficient and a clear conclusion could not be drawn. Future trials with a high-quality methodological design and large sample size are warranted.




Reference


Andersson, L., et al. (2001). Immediate or late weight bearing after uncemented total hip arthroplasty: A study of functional recovery. Journal of Arthroplasty, 16(8), 1063-1065. doi:http://dx.doi.org/10.1054/arth.2001.27253

Bodén, H., et al. (2004). No adverse effects of early weight bearing after uncemented total hip arthroplasty: a randomized study of 20 patients. Acta orthopaedica Scandinavica, 75(1), 21-29. doi:10.1080/00016470410001708040

Bottner, F., et al. (2005). Implant migration after early weightbearing in cementless hip replacement. Clinical Orthopaedics and Related Research(436), 132-137. doi:http://dx.doi.org/10.1097/01.blo.0000160380.15429.fb

Di Monaco, M., et al. (2009). Rehabilitation after total hip arthroplasty: a systematic review of controlled trials on physical exercise programs. Eur J Phys Rehabil Med, 45(3), 303-317.

Guyatt, G. H., et al. (2011). GRADE guidelines 6. Rating the quality of evidence--imprecision. J Clin Epidemiol, 64(12), 1283-1293. doi:10.1016/j.jclinepi.2011.01.012

Hol, A. M., et al. (2010). Partial versus unrestricted weight bearing after an uncemented femoral stem in total hip arthroplasty: recommendation of a concise rehabilitation protocol from a systematic review of the literature. Archives of orthopaedic and trauma surgery, 130(4), 547-555. doi:https://dx.doi.org/10.1007/s00402-009-1017-3

Kishida, Y., et al. (2001). Full weight-bearing after cementless total hip arthroplasty. International Orthopaedics, 25(1), 25-28. doi:http://dx.doi.org/10.1007/s002640000221

Maggs, J., et al. (2017). The Relative Merits of Cemented and Uncemented Prostheses in Total Hip Arthroplasty. Indian journal of orthopaedics, 51(4), 377-385. doi:10.4103/ortho.IJOrtho_405_16

Markmiller, M., et al. (2011). Partial weightbearing is not necessary after cementless total hip arthroplasty A two-year prospective randomized study on 100 patients. International Orthopaedics, 35(8), 1139-1143. doi:http://dx.doi.org/10.1007/s00264-010-1089-2

Monticone, M., et al. (2014). Task-oriented exercises and early full weight-bearing contribute to improving disability after total hip replacement: a randomized controlled trial. Clinical rehabilitation, 28(7), 658-668. doi:http://dx.doi.org/10.1177/0269215513519342

Rao, R. R., et al. (1998). Immediate Weightbearing After Uncemented Total Hip Arthroplasty. Clinical Orthopaedics and Related Research®, 349.

Shabana, M., et al. (2017). Immediate unrestricted versus graduated weight bearing following primary cementless total hip arthroplasty: a randomized controlled trial. Annals of Joint; Vol 2, No 2 (February 2017): Annals of Joint.

Ström, H., et al. (2006). Unrestricted weight bearing and intensive physiotherapy after uncemented total hip arthroplasty. Scandinavian Journal of Surgery, 95(1), 55-60. doi:http://dx.doi.org/10.1177/145749690609500111

Ström, H., et al. (2007a). Early migration pattern of the uncemented CLS stem in total hip arthroplasties. Clinical orthopaedics and related research, 454, 127-132. doi:10.1097/01.blo.0000238785.98606.9d

Ström, H., et al. (2007b). The effect of early weight bearing on migration pattern of the uncemented CLS stem in total hip arthroplasty. Journal of arthroplasty, 22(8), 1122-1129. doi:10.1016/j.arth.2006.11.015

Thien, T. M., et al. (2007). Immediate weight bearing after uncemented total hip arthroplasty with an anteverted stem: A prospective randomized comparison using radiostereometry. Acta Orthopaedica, 78(6), 730-738. doi:http://dx.doi.org/10.1080/17453670710014491

Tian, P., et al. (2017). Partial versus early full weight bearing after uncemented total hip arthroplasty: a meta-analysis. Journal of orthopaedic surgery and research, 12(1), 31. doi:https://dx.doi.org/10.1186/s13018-017-0527-x

Tian, P., et al. (2017). Partial versus early full weight bearing after uncemented total hip arthroplasty: a meta-analysis. Journal of orthopaedic surgery and research, 12(1), 31. doi:http://dx.doi.org/10.1186/s13018-017-0527-x

Unver, B., et al. (2004). Comparison of two different rehabilitation programmes for thrust plate prosthesis: a randomized controlled study. Clinical rehabilitation, 18(1), 84-91. doi:10.1191/0269215504cr705oa

Wolf, O., et al. (2010). Periprosthetic bone mineral density and fixation of the uncemented CLS stem related to different weight bearing regimes: A randomized study using DXA and RSA in 38 patients followed for 5 years. Acta Orthopaedica, 81(3), 286-291. doi:http://dx.doi.org/10.3109/17453674.2010.487238

Wolf, O., et al. (2012). The effects of different weight-bearing regimes on press-fit cup stability: a randomised study with five years of follow-up using radiostereometry. International orthopaedics, 36(4), 735-740. doi:https://dx.doi.org/10.1007/s00264-011-1413-5

Zhang, C., et al. (2017). Cemented or cementless fixation for primary hip arthroplasty—evidence from The International Joint Replacement Registries. Annals of Joint; Vol 2, No 10 (October 2017): Annals of Joint.







Please Login or Join to leave comments.