Sage Journals: Discover world-class research

Abstract

We aimed to systematically compare the clinical and functional outcomes between unicompartmental knee arthroplasty (UKA) and high tibial osteotomy (HTO) for the treatment of medial knee osteoarthritis (KOA). Literatures were searched from PubMed, EMBASE, the Cochrane library, Wanfang DATA, China National Knowledge Infrastructure (CNKI) and SinoMed database until December 2020. Studies comparing postoperative clinical and functional outcomes of UKA versus HTO were included. Totally, 38 studies were included, including 2368 patients with 2393 knees in HTO group and 6536 patients with 6571 knees in UKA group. There was significant difference in postoperative pain, revision rate, complications, and the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) score between HTO and UKA groups (p < 0.05). No significant difference was found in excellent/good surgical results, Lysholm, Hospital for Special Surgery (HSS) score, Knee Society Knee (KSS) score, knee and function score of Knee Society (KSFS) score and Tegner score between these two groups (p > 0.05). UKA produced less postoperative pain, less complications and superior WOMAC score, whereas HTO offered extended range of motion (ROM) and less revision rate.

Keywords

Unicompartmental knee arthroplasty high tibial osteotomy osteoarthritis medial compartment meta-analysis

Introduction

Knee osteoarthritis (KOA) is the common chronic joint disorder, which is generally characterized by the knee articular cartilage degeneration and the subsequent bone-to-bone wearing predominantly in the medial compartment.¹ Medial KOA may result from knee alignment disorders or injuries that damages the cartilage and overloads the medial tibiofemoral compartment.² Knee malalignment has been suggested as an important biomechanical risk factor in KOA progression and declines in physical function.^3,4 Both high tibial osteotomy (HTO) and unicompartmental knee arthroplasty (UKA) are most frequently used medial KOA joint surgeries. Both types can improve knee function, reduce pain, slow knee deterioration and possibly delay the need for total knee replacement surgery.⁵ Classically, the surgical indication of two types operations for medial KOA is well defined. HTO is particularly used in young and active patients, while UKA is the preferred treatment option for older patients with moderate or severe osteoarthritis.^6–8 Today, the inclusion criteria have been greatly expanded with improvements in surgical techniques, implant design and modern instruments.^9–11

HTO is a surgical procedure providing an offloading of the damaged medial cartilage by realignment of the knee joint and correcting the varus deformity. Patients receiving HTO can benefit from natural knee joint preservation and almost unaffected physical loading, which can delay or prevent the need for a partial or total knee replacement.^8,12 Opening-wedge HTO (OWHTO) and closing-wedge HTO (CWHTO) are the two commonly used HTO procedures.¹³ The choice of the osteotomy technique depends on the degree and location of knee malalignment and the experience of the surgeon. UKA is an alternative procedure to total knee replacement. In UKA, only the affected degenerative compartment of the knee is resurfaced and replaced with an implant prosthesis, while the nonaffected compartment is preserved. UKA allows knee bone stock and ligaments sparing and offers patients a less invasive procedure with excellent pain relief and faster recovery.¹⁴

To date, several meta-analyses have directly compared HTO and UKA methods in patients with medial KOA.^15–19 However, the comparative studies^4,20 and meta-analyses have showed conflicting conclusions. The purpose of this study was to compare the clinical and functional outcomes of UKA versus HTO for teh treatment of medial KOA.

Materials and methods

Literature search

According to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline, databases including PubMed, EMBASE, the Cochrane library, Wanfang DATA, China National Knowledge Infrastructure (CNKI) and SinoMed were electronically searched for studies published before December 2020.

All analyses were based on previously published studies, so no ethical approval and patient consent are required.

Eligibility criteria

The inclusion criteria included: (1) patients were diagnosed with medial KOA; (2) original studies directly compared HTO with UKA for the treatment of medial KOA and reported at least one of function results (e.g., knee function score, range of motion [ROM], femorotibial angle [FTA]), revision incidence, complications (e.g., infection, thrombosis, and pain); (3) study type: both randomized controlled trials (RCTs) and nonrandomized controlled trials (non-RCTs) were considered; (4) language was limited to English or Chinese.

Search strategy

We started systematic search from the databases of PubMed, EMBASE, the Cochrane library, Wanfang DATA, CNKI and SinoMed from available papers in literature up to December 2020 for potentially eligible studies.

For the literature search, we used the medical subject heading (MeSH) term of ‘Osteoarthritis,” “Osteoarthritis, Knee,” and “Arthroplasty, Replacement, Knee,” and “High tibial osteotomy” “Unicompartmental,” as well as all key words corresponding to these MeSh terms.

Data extraction and quality assessment

The selection and inclusion of studies were performed in two stages by two independent authors (** and **). This included the screening of titles/abstracts followed by the full texts. Disagreements between the two authors were resolved by a third authors (**). The basic characteristics of the data from included studies consisted of names of authors, publication year, study design, country, sample size, mean age, female percentage, prosthesis properties, outcomes, follow-up time and radiographic outcomes.

We assessed the risk of bias in the RCTs using the Cochrane risk of bias tool to determine whether biases might affect the results.²¹ The risk of bias in nonrandomized studies were assessed using the Risk of Bias in Non-Randomized Studies of Interventions (ROBINS-I) assessment tool.²²

Statistical analysis

The odds ratio (OR) and the corresponding 95% confidence interval (CI) was used to express the estimates of dichotomous outcomes. The estimates of continuous outcomes were expressed by using mean difference (MD) and the corresponding 95% CI. In all cases, p values <0.05 were considered statistically significant. Sensitivity analysis was conducted to obtain a solid conclusion and to evaluate the stability of the pooled results.

Cochran’s Q statistic²³ and Higgin’s inconsistency index²⁴ were used to assess the statistical heterogeneity among studies. Significant statistical heterogeneity was considered if I2 >50% and p < 0.1, then a random-effects model was used for meta-analysis. On the contrary, a fixed-effects model was selected if I2 ≤50% and p > 0.1. Naunyn Schmied. Arch. Exp. Pathol. Pharmakol. 5, S38). In addition, sensitivity analyses were performed according to the quality of the included studies by only studies assessed as moderate or high were included and pooled. Further sensitivity analyses were performed to identify individual study effects on pooled results and test the reliability of results by introducing leave-one-out strategy. All of the statistical analyses were performed using statistics and data (STATA) SE 14.0 software (StataCorp, College Station, Texas, USA).

Results

Study selection and characteristics of included studies

Totally, 252 relevant publications were initially retrieved. Duplicated documents, studies on irrelevant disease or intervention, case report or case series, meta or review articles, and documents of comments, protocols, thesis and surveys were subsequently removed. In addition, 16 articles were not published in English or Chinese, and the full text of four articles could not be obtained. After the screening, 51 articles were selected and assessed thoroughly. Among them, 13 articles were excluded because of insufficient data. Finally, 38 articles met the criteria were included for this meta-analysis. The eligible study selection process was shown in Figure 1. The basic characteristics of the selected 38 comparative studies were summarized in Table 1. A total of 2368 medial KOA patients with 2393 knees received HTO and 6516 patients with 6551 knees received UKA. There were six RCT studies, and the rest were all retrospectively observational studies. The mean age of the patients ranged from 43 to 77. The follow-up time ranged from 8.5 ± 1.8 months to 434.4 ± 4.8 months.

Figure 1.

Flowchart of the search process of our study.

Table 1.

Characteristics of the included studies.

ID	Country	Study design	Age	Gender (%female)	Simple size (p/k)	Prosthesis properties	Follow-up (month)	Outcomes
			HTO	HTO	HTO	HTO	HTO
			UKA	UKA	UKA	UKA	UKA
Ye, 2020	China	RCT	59.94 ± 5.13	55	20	/	/	Tegner score, Lysholm knee score
Ye, 2020	China	RCT	60.53 ± 4.49	55	20	Ⅲ prosthesis	/	Tegner score, Lysholm knee score
Zhang Y, 2019	China	RCT	/	/	13	OWHTO	/	Tegner score, Lysholm knee score
Zhang Y, 2019	China	RCT	/	/	13	Oxford Ⅲ	/	Tegner score, Lysholm knee score
Guo, 2017	China	RCT	58.76 ± 6.67	56.7	30	/	12	VAS, Lysholm knee score
Guo, 2017	China	RCT	59.02 ± 6.84	60	30	/	12	VAS, Lysholm knee score
Börjesson, 2005	Sweden	RCT	63 ± 4	44.4	18	CWHTO	60	BOA score; pain during walking, PROM
Börjesson, 2005	Sweden	RCT	63 ± 4	50	22	Parapatellar skin and retinacular incision	60	BOA score; pain during walking, PROM
Stukenborg-Colsman, 2001	Germany	RCT	67 ± 4.75	40.6	32	Lateral-based wedge	90 ± 20.4	KSS score; complication
Stukenborg-Colsman, 2001	Germany	RCT	67 ± 5	78.6	28 (30)	Tubingen pattern	90 ± 20.4	KSS score; complication
Weidenhielm, 1992	Sweden	RCT	63	56	25	CWHTO	12	BOA score, walking speed
Weidenhielm, 1992	Sweden	RCT	63	50	28	Brigham	12	BOA score, walking speed
Dahl A, 2010	Sweden	Registry review	/	/	450	Hemicallotasis	/	Revision rate
Dahl A, 2010	Sweden	Registry review	/	/	4799	Many	/	Revision rate
Wang, 2020	China	RCS	58. 3 ± 3. 5	66.7	12 (12)	OWHTO	8. 5 ± 1. 8	VAS, HSS score
Wang, 2020	China	RCS	59. 4 ± 4. 5	61.1	36 (36)	Oxford Ⅲ	8. 5 ± 1. 8	VAS, HSS score
Zhang, 2020	China	RCS	51.8 ± 6.9	78.9	109	OWHTO	24	HSS score, level-walking, VAS, low-impact sports recovery, complication, revision rate
Zhang, 2020	China	RCS	53.7 ± 5.2	79.5	83	Medial parapatellar approach	24
Jin, 2020	China	RCS	64.1 ± 4.0	100	67	OWHTO	133.2 ± 19.56	ROM, the knee and function score of knee society (KSFS), HSS score, WOMAC score, forgotten joint score
Jin, 2020	China	RCS	63.1 ± 4.9	97	67	Miller-galante fixed-bearing prosthesis with cement	142.8 ± 24.6
Jacquet, 2020	France	RCS	49.3 ± 3.9	44	50	Miniaci’s method	44.4 ± 19.2	KOOS, UCLA, KSS, time to return to sport
Jacquet, 2020	France	RCS	50.8 ± 4.4	42	50	Cemented medial	46.8 ± 21.6	KOOS, UCLA, KSS, time to return to sport
Bouguennec, 2020	France	RCS	55.1	31.4	488	O/CWHTO	93.6 ± 34.8	Survival, WOMAC score, Tegner score, return to work, satisfaction
Bouguennec, 2020	France	RCS	64.1	60.6	284	Miniaci or Dugdale or the “cable” technique	66 ± 43.2
Zhao, 2019	China	RCS	55.65 ± 6.23	86.7	30	OWHTO	12	HSS, WOMAC
Zhao, 2019	China	RCS	56.23 ± 7.56	86.7	30	Oxford Ⅲ	12	HSS, WOMAC
Zhang P, 2019	China	RCS	58.96 ± 7.62	61.9	42	/	12	VAS, HSS, KSS, Lysholm knee score, complications
Zhang P, 2019	China	RCS	57.28 ± 7.46	62.5	40	LINK	12	VAS, HSS, KSS, Lysholm knee score, complications
Zhang K, 2019	China	RCS	57.9 ± 3.5	43.5	23	/	/	Tegner score, HSS score
Zhang K, 2019	China	RCS	59.6 ± 4.5	47.8	23	/	/	Tegner score, HSS score
Ma, 2019	China	RCS	67.6 ± 6.2	52.1	74 (80)	OWHTO	24	HSS, ROM
Ma, 2019	China	RCS	66.6 ± 7.2	53.7	41 (56)	Oxford	24	HSS, ROM
Liang, 2019	China	RCS	59.6 ± 6.3	65.5	55	/	12	HSS score, Lysholm knee score, complications, ROM
Liang, 2019	China	RCS	61.3 ± 5.8	76.7	43	/	12	HSS score, Lysholm knee score, complications, ROM
Li Y, 2019	China	RCS	57.6 ± 3.5	55.6	18	OWHTO	/	VAS, KOOS, IKDC, ROM, Lysholm knee score
Li Y, 2019	China	RCS	59.2 ± 3.4	59.1	22	Oxford Ⅲ	/	VAS, KOOS, IKDC, ROM, Lysholm knee score
Li J, 2019	China	RCS	59.4 ± 3.1	65	20	OWHTO	32 ± 6.7	VAS, KSS, WOMAC
Li J, 2019	China	RCS	60.8 ± 2.9	72.2	18	Oxford	28 ± 4.4	VAS, KSS, WOMAC
Jia, 2019	China	RCS	55.7 ± 3.25	75	20	/	42 ± 4.5	ROM, HSS
Jia, 2019	China	RCS	56.3 ± 3.75	75	28	Oxford Ⅲ	43.2 ± 2.5	ROM, HSS
Deng, 2019	China	RCS	60.2 ± 2.3	77.8	18	/	12	HSS score, VAS, ROM, complications
Deng, 2019	China	RCS	60.8 ± 2.1	70.6	17	Oxford	12	HSS score, VAS, ROM, complications
Chen, 2019	China	RCS	56.1 ± 6.4	72.2	18	OWHTO	12–33.6	Tegner score, Lysholm knee score
Chen, 2019	China	RCS	56.7 ± 6.5	70	20	BIOMET	12–38.4	Tegner score, Lysholm knee score
Song, 2019	South Korea	RCS	59.7 ± 4.1	85	60K	CWHTO	128.4 ± 68.4	Survival, KSFS, WOMAC, ROM
Song, 2019	South Korea	RCS	60.8 ± 3.9	86	50K	Fixed-bearing	144 ± 85.2	Survival, KSFS, WOMAC, ROM
Koh, 2019	Korea	RCS	56.1 ± 5.6	85	123	OWHTO	48	VAS, WOMAC, KSS, satisfaction
Koh, 2019	Korea	RCS	60.8 ± 4.7	83	118	Oxford	48	VAS, WOMAC, KSS, satisfaction
An, 2018	China	RCS	55.62 ± 6.12	62.5	16	OWHTO	14.5 ± 3.75	VAS, HSS score, WOMAC
An, 2018	China	RCS	59.12 ± 3.57	74.3	35	Oxford	15.3 ± 4.5	VAS, HSS score, WOMAC
Cho, 2018	South Korea	RCS	58.4 ± 5.5	60	17 (20)	OWHTO	48.4 ± 14.3	HSS score, KSFS, ROM
Cho, 2018	South Korea	RCS	67.9 ± 9.0	95	17 (20)	Mobile-bearing prosthesis	39.7 ± 14.0	HSS score, KSFS, ROM
Ryu, 2018	Korea	RCS	57.6 ± 6.4	91.3	23	OWHTO	40.0 ± 19.4	Lysholm score, VAS, HSS score, WOMAC
Ryu, 2018	Korea	RCS	60.5 ± 3.4	86.4	22	DePuy	33.1 ± 8.7	Lysholm score, VAS, HSS score, WOMAC
Tang, 2017	China	RCS	60.7 ± 7.5	60	15	OWHTO	/	Tegner score, Lysholm knee score
Tang, 2017	China	RCS	60.5 ± 7.7	53.3	15	BIOMET	/	Tegner score, Lysholm knee score
Krych, 2017	Minnesota	RCS	43	54.9	57	O/CWHTO	86.4	Surviving, Tegner scores, Lysholm score, revision rate
Krych, 2017	Minnesota	RCS	49		183	Cemented, metal-backed, fixed-bearing components	69.6	Surviving, Tegner scores, Lysholm score, revision rate
Jeon, 2017	Korea	RCS	56.8 ± 5.5	84.6	26	OWHTO	34.7	KOOS, IKDC, VAS
Jeon, 2017	Korea	RCS	60.7 ± 6.25	81	21	Miniaci method	34.7	KOOS, IKDC, VAS
Petersen, 2016	Germany	RCS	58.9 ± 2.8	39.1	23	OWHTO	60	KOOS, HSS score, revision rates
Petersen, 2016	Germany	RCS	60.7 ± 2.5	64	25	Anteromedial	60	KOOS, HSS score, revision rates
Yang, 2015	China	RCS	50.3 ± 3.5	91.7	12	CWHTO	43.2 ± 3	ROM, Tegner score, Lysholm knee score
Yang, 2015	China	RCS	55.4 ± 3.5	93.8	16	Oxford Ⅲ	42 ± 4.5	ROM, Tegner score, Lysholm knee score
Ding, 2015	China	RCS	55 ± 5	75.5	53 (55)	/	114 ± 24	HSS score
Ding, 2015	China	RCS	61 ± 8	65.2	23 (25)	Oxford	40 ± 3	HSS score
Tuncay, 2015	Turkey	RCS	52.4 ± 3.3	79.5	88	OWHTO + Dome	36.4 ± 8.26	HSS score, revision rate, complications
Tuncay, 2015	Turkey	RCS	58.7 ± 6	84	94	Oxford	42.5 ± 15	HSS score, revision rate, complications
Karamitev, 2014	Bulgaria	RCS	52–84	48.91	92 (103)	CWHTO	48	KSS
Karamitev, 2014	Bulgaria	RCS	/	64.6	65 (66)	Endoprosthesis	48	KSS
Yim, 2013	South Korea	RCS	58.3 ± 5.4	87.9	58	OWHTO	43.2 ± 4.8	Tegner score, ranges of motion, Lysholm knee score, complication
Yim, 2013	South Korea	RCS	60.3 ± 4.5	96	50	Miller-galante fixed-bearing prosthesis	434.4 ± 4.8
Takeuchi, 2010	Japan	RCS	67 ± 7	75	24 (27)	OWHTO	61 ± 10	Average extension angle, average flexion angle, KSS, function scores
Takeuchi, 2010	Japan	RCS	77 ± 4	77.8	18 (30)	Compartment uni-knee	84 ± 4
Broughton, 1986	England	RCS	63	77.6	49	CWHTO	93.6 ± 18	Revision rate, the rate of excellent and good, pain, complications
Broughton, 1986	England	RCS	71	73.8	42	St Georg	69.6 ± 14.4

HTO, high tibial osteotomy; UKA, unicompartmental knee arthroplasty; OWHTO, opening wedge high tibial osteotomy; CWHTO, closing wedge high tibial osteotomy; VAS, Visual Analogue Scale; BOA, British Orthopaedic Association score; PROM, postoperative range of motion; KSS, Knee Society Score; HSS score, the Hospital for Special Surgery score; ROM, range of motion; WOMAC, The Western Ontario and McMaster Universities; KOOS, Knee Osteoarthritis Outcomes Score; UCLA, UCLA Activity Score; KSS, American Knee Society Knee Score; KSFS, The knee and function score of Knee Society; IKDC, The International Knee Documentation Committee; RCS, retrospective comparative study; RCT, randomized controlled trial.

Quality assessment

The quality assessment of the included studies was shown in Table 2. For RCT studies, five of the six articles provided methods for random sequence generation. All RCT studies had high risk of bias for allocation concealment and blinding of participants and personnel. Blinding of outcome assessment was often not described. All studies presented sufficient data on outcome measures. The selective reporting bias of two studies were unclear,^25,26 and the remaining studies had low bias.

Table 2.

Quality assessment of included studies.

RCT study	Random sequence generation	Allocation concealment		Blinding of participants and peraonnel	Blinding of outcome assessmnet	Incomplete outcome data	Selective reporting	Other bias
Ye, 2020	High	High		High	Unclear	Low	Low	Unclear
Zhang Y, 2019	Low	High		High	Unclear	Low	Low	Unclear
Guo, 2017	Low	High		High	Unclear	Low	Low	Unclear
Börjesson, 2005	Low	High		High	Unclear	Low	Unclear	Unclear
Stukenborg-Colsman, 2001	Low	High		High	Unclear	Low	Unclear	Unclear
Weidenhielm, 1992	Low	High		High	Unclear	Low	Low	Unclear
nRCT study	Bias due to confounding	Bias in selection of participants	Bias in classification of interventions	Bias due to Deviations from intended interventions	Bias due to missing data	Bias in measurement of outcomes	Bias in selection of the reported result	Overall bias
Wang, 2020	Low	Low	Low	Low	Low	Moderate	Moderate	Low
Zhang, 2020	Moderate	Low	Low	Low	Low	Low	Moderate	Low
Jin, 2020	Low	Low	Moderate	Low	Low	Moderate	Moderate	Moderate
Jacquet, 2020	Low	Low	Low	Low	Moderate	Low	Moderate	Low
Bouguennec, 2020	Serious	Serious	Moderate	Low	Moderate	Low	Moderate	Moderate
Zhao, 2019	Moderate	Moderate	Low	Moderate	Low	Low	Moderate	Moderate
Zhang P, 2019	Low	Low	Low	Low	Low	Low	Low	Low
Zhang K, 2019	Serious	Serious	Moderate	Serious	Low	Moderate	Moderate	Serious
Ma, 2019	Low	Low	Low	Low	Low	Low	Moderate	Low
Liang, 2019	Moderate	Moderate	Moderate	Low	Low	Low	Low	Moderate
Li Y, 2019	Moderate	Moderate	Moderate	Moderate	Low	Moderate	Low	Moderate
Li J, 2019	Low	Low	Low	Moderate	Moderate	Moderate	Low	Moderate
Jia, 2019	Moderate	Moderate	Moderate	Moderate	Low	Low	Low	Moderate
Deng, 2019	Moderate	Moderate	Serious	Moderate	Low	Low	Low	Moderate
Chen, 2019	Serious	Moderate	Moderate	Moderate	Low	Low	Low	Moderate
Song, 2019	Moderate	Moderate	Serious	Moderate	Moderate	Low	Low	Moderate
Koh, 2019	Low	Low	Low	Low	Low	Low	Low	Low
An, 2018	Moderate	Moderate	Serious	Moderate	Low	Low	Low	Moderate
Cho, 2018	Low	Low	Low	Low	Low	Moderate	Moderate	Low
Ryu, 2018	Low	Low	Low	Low	Low	Low	Low	Low
Tang, 2017	Moderate	Moderate	Moderate	Moderate	Low	Low	Low	Moderate
Krych, 2017	Low	Low	Low	Low	Moderate	Moderate	Moderate	Moderate
Jeon, 2017	Low	Low	Low	Low	Low	Low	Low	Low
Petersen, 2016	Moderate	Low	Low	Low	Moderate	Moderate	Moderate	Moderate
Yang, 2015	Moderate	Moderate	Moderate	Moderate	Moderate	Low	Moderate	Moderate
Ding, 2015	Moderate	Low	Low	Low	Moderate	Moderate	Moderate	Moderate
Tuncay, 2015	Serious	Moderate	Low	Low	Moderate	Moderate	Moderate	Moderate
Karamitev, 2014	Serious	Serious	Moderate	Low	Serious	Moderate	Serious	Serious
Yim, 2013	Low	Low	Moderate	Low	Moderate	Moderate	Moderate	Moderate
Takeuchi, 2010	Moderate	Low	Low	Low	Moderate	Low	Moderate	Moderate
Broughton, 1986	Moderate	Moderate	Moderate	Low	Moderate	Moderate	Moderate	Moderate

The other included studies were all retrospective observational studies with inherent bias including confounding factors, selection bias, missing data and other bias in the design and implementation of results. Overall, nine studies had low bias, 20 studies presented moderate bias, and two studies showed serious bias.

Meta-analysis of clinical outcomes

Lysholm score

Twelve studies, including 361 patients with 361 knees in the HTO group and 474 patients with 474 knees in the UKA group, reported Lysholm score. Pooled result revealed no significant difference between the two groups (MD: −1.086, 95% CI: −2.219 to 0.046, p = 0.06; I² = 69.6%) (Figure 2 and Table 3).

Figure 2.

Forest plots of functional results.

Table 3.

Result of the meta-analysis.

Clinical outcomes	N	OR (95%CI)	p	I-square, %	p For Heterogeneity test
The rate of excellent and good	7	0.881 (0.681, 1.141)	0.338	0	0.894
Revision rate	9	0.965 (0.782, 1.189)	0.737	70.1	0.001
Complications	12	1.421 (1.046, 1.929)	0.024	0	0.678
Functional outcomes	N	WMD (95% CI)	p	I-square, %	p For heterogeneity test
Lysholm score	12	−1.086 (−2.219, 0.046)	0.060	69.6	<0.001
VAS score	11	0.371 (0.040, 0.702)	0.028	93.9	<0.001
HSS score	12	−0.433 (−1.670, 0.803)	0.492	55.9	0.009
Knee society score	8	−0.801 (−3.606, 2.004)	0.576	85.3	<0.001
Knee society function score	5	4.787 (−4.435, 14.009)	0.309	89.5	<0.001
Tegner score	7	−0.042 (−0.624, 0.540)	0.887	94.6	<0.001
WOMAC score	8	3.627 (1.333, 5.921)	0.002	86.5	<0.001
ROM score	10	2.114 (0.176, 4.052)	0.033	82.3	<0.001

VAS, Visual Analogue Scale; HSS, Hospital for Special Surgery; WOMAC, The Western Ontario and McMaster Universities; ROM, range of motion; OR, odds ratio; MD, mean difference; CI, confidence interval.

Hospital for special surgery score

Hospital for Special Surgery (HSS) score was assessed in twelve studies, including 462 patients with 467 knees in the HTO group and 441 patients with 446 knees in the UKA group, and pooled result suggested no significant difference between the two groups (MD: −0.433, 95% CI: −1.670 to 0.803), p = 0.492; I² = 55.9%) (Figure 2 and Table 3).

Visual analogue scale

Eleven studies reported the data on Visual Analogue Scale (VAS), including 378 patients with 378 knees in the HTO group and 409 patients with 409 knees in the UKA group, and pooled result suggested that UKA was significantly associated with less pain compared to HTO (MD: 0.371, 95% CI: 0.04 to 0.702, p = 0.028; I² = 93.9%) (Figure 2 and Table 3).

Range of motion score

Ten studies analyzed range of motion (ROM) score, including 399 patients with 408 knees in the HTO group and 351 patients with 369 knees in the UKA group. Pooled result revealed better flexion and extended ROM in the HTO group when compared to the UKA group (MD: 2.114, 95% CI: 0.176 to 4.052, p = 0.033; I² = 82.3%) (Figure 2 and Table 3).

The Western Ontario and McMaster Universities score

Eight studies reported the data on the Western Ontario and McMaster Universities (WOMAC) score, including 827 patients with 827 knees in the HTO group and 624 patients with 624 knees in the UKA group, and the pooled result showed a significantly higher WOMAC score in the UKA group compared to the HTO group (MD: 3.627, 95% CI: 1.333 to 5.921, p = 0.002; I² = 86.5%) (Figure 2 and Table 3).

American knee society knee score

Totally, eight studies reported the American Knee Society Knee (KSS) score, including 319 patients with 331 knees in the HTO group and 262 patients with 294 knees in the UKA group, and pooled result suggested that there was no significant difference between the two groups (MD: −0.801, 95% CI: −3.606 to 2.004, p = 0.576; I² = 85.3%) (Figure 2 and Table 3).

Tegner score

Seven studies assessed Tegner score, including 671 patients with 671 knees in the HTO group and 591 patients with 591 knees in the UKA group, and pooled result showed no significant difference between the two groups (MD: −0.042, 95% CI: −0.624 to 0.540), p = 0.887; I² = 94.6%) (Figure 2 and Table 3).

The knee and function score of knee society

Five studies reported the data on the knee and function score of Knee Society (KSFS), including 200 patients with 206 knees in the HTO group and 180 patients with 197 knees in the UKA group, and pooled result revealed no significant difference between the two groups (MD: 4.787, 95% CI: −4.435 to 14.009, p = 0.309; I² = 89.5%) (Figure 2 and Table 3).

Rate of excellent and good

The analysis of seven studies including 296 patients with 312 knees in the HTO group and 224 patients with 241 knees in the UKA group yielded no statistically significant difference between the two groups regarding excellent/good results (OR: 0.88, 95% CI: 0.68 to 1.14, p = 0.338; I² = 0%) (Figure 3 and Table 3).

Figure 3.

Forest plots of clinical results.

Complications

Generally, more complications were noted after performing HTO with significant difference between the two groups (OR: 1.42, 95% CI: 1.05 to 1.93, p = 0.024; I² = 0%) based on the analysis on 910 HTO patients with 916 knees and 691 UKA patients with 708 knees (Figure 3 and Table 3).

Revision rate

Nine studies reported revision rates, including 1320 patients with 1323 knees in the HTO group and 5556 patients with 5570 knees in the UKA group, and the pooled data analysis showed no significant difference between the two groups (OR: 0.97, 95% CI: 0.78 to 1.19, p = 0.737; I² = 70.1%) (Figure 3 and Table 3). Considering the fact that some of the included studies had a short follow-up period, thereby inevitably affecting the reliability of the data in terms of revision rate. Therefore, we further compared the revision rates between the two groups by including only studies with follow-up period of more than 60 months, and pooled result suggested that HTO was associated with significantly lower revision rate as compared with UKA (OR: 0.54, 95% CI: 0.28 to 0.79, p < 0.001, I2 = 0.1%).

Sensitivity analysis

We then conducted sensitivity analyses according to the quality of the included studies. As showed in Supplementary Figures 1 and 2, only studies assessed as moderate or high were included and pooled. The results of clinical outcomes were comparable with the overall analyses of all studies included except for the pooled results of VAS, which was not significant with MD as 0.52 (95% CI: −0.04 to 1.09, p = 0.07, as shown in Supplementary Figure 1.

The sensitivity analysis based on the leave-one-out strategy suggested that the results were robust (Supplementary Figure 3A-K).

Publication bias

There was no significant publication bias when assessing funnel plots of Lysholm, score, HSS score, VAS score, ROM score, and complications (Figure 4(a)–(e), supplementary Table 1). All p values for Begg’s test were greater than 0.05 (supplementary Table 1).

Figure 4.

Funnel plot of (a) Lysholm score; (b) HSS score; (c) VAS; (d) ROM score; and (e) Complications. HSS, the Hospital for Special Surgery; VAS, Visual Analogue Scale; ROM, range of motion.

Discussion

There was significant difference in VAS score, WOMAC score, postoperative complications, revision rate, and ROM score. Patients receiving UKA showed less postoperative pain, less postoperative complications and superior knee function WOMAC score than patients receiving HTO. However, patients undergoing HTO obtained more ROM and less revision rate than patients undergoing UKA. No significant difference was observed between the two groups in terms of excellent/good surgical results and other knee function scores, including Lysholm, HSS score, KSS score, KSFS score, and Tegner score.

To date, several meta-analysis review articles have been published to summarize the benefits and harms of HTO and UKA surgeries on medial KOA patients. Among them, there were variation in the selection of outcome indicators and assessment system, and the results were inconsistent. Comparing with previous studies, half of the selected articles were from China (19 from 38 studies)^27–36 and half of them were updated publications in 2019 and 2020 (19 from 38 studies),^{27–29,32–36} which were not at all or unfrequently discussed in other systemically quantitative analyses, resulting in obviously different comparison results. In contrast with the meta-analyses by Cao et al.¹⁵ Santoso et al.¹⁶ and Han et al.¹⁷ which observed no significant difference on knee function scores, in this study, higher WOMAC score was found after UKA compared with HTO indicating UKA may produce better functional outcome. Similarly with traditional conclusions regarding VAS pain score, postoperative complications and ROM, UKA could cause less pain and complications while HTO could cause increased flexion and extended ROM.

In the current meta-analysis, many included studies adopted OWHTO. According to recent publications, the OWHTO is increasing its popularity. Compared to CWHTO, the OWHTO has several advantages, e.g., easier and more accurate adjustment of alignment correction, the avoidance of serious complications, and higher survival rate at 10-year follow-up.^13,37,38 UKA has one disadvantage of higher revision rate, applying the appropriate patient’s indications, modern prosthetic design and techniques would significantly decrease the revision rate.^7,39 There were various techniques in UKA in this study, mainly using Oxford III prothesis which was the preferred method currently.⁴⁰ As concluded, the pooled data analysis did not find significant difference in terms of revision rate, excellent/good surgical results, and other knee function scoring system including Lysholm, HSS, KSS, KSFS and Tegner score between HTO and UKA. However, when comparing the individual study by Krych et al. in 2017⁴¹ and Bouguennec et al. in 2020,⁴² opposite results of revision rate were illustrated in Figure 3(c), and Krych showed higher treatment failure rate for UKA. According to the demographic data analysis, both studies used O/CWHTO procedures, with similar follow-up timing about 6-7 years. Additionally, the mean age of HTO patients was 43 years, and 49 years for UKA patients in the earlier study, while HTO patients aged 55 years and UKA patients aged 64 years in the recent study, which may explain the difference in the revision rate outcome.

Except for the indicators of excellent and good surgical results and postoperative complications, all the other outcomes pooled analyses had high heterogeneity (p < 0.10) which were probably caused by multiple factors such as patient selection, surgeons performing technique, outcome measures and follow-up time. In this meta-analysis, the proportion of female patients was relatively high. For many recent publications, there were no significant difference in age when undergoing HTO and UKA, and most patients were 55–65 years old. The surgical indications of the included studies are not uniform. For the past meta-analysis, many included articles were published earlier, with older technology and higher indications requirements (HTO: mostly younger patients, UKA: mostly older patients), which may suggest clinical heterogeneity. This could also cause differences in outcome analysis results. Moreover, the follow-up time of the literature collected in this study varied. The clinical effects of follow-up were mainly medium and short-term, while the long-term clinical effects lacked research and needed further exploration.

There were several limitations about this study. First, more than half of the studies on HSS score, Lysholm score, VAS score, Tegner score and ROM score collected the data from Chinese Han population, which might lead to biased result in terms of the study population. Second, the sample size was small for certain outcome indicators including KSFS score. Third, follow-up in many update studies was relatively short, therefore the high heterogeneity resulting from varying follow-up times should significantly affect the reliability of the data in terms of revision rate. In fact, additional analysis based on studies with follow-up of more than 60 months significantly changed the pooled result, indicating a beneficial result to HTO. However, due to the limited number of eligible studies, more studies are required to validate this finding. Fourth, this study did not conduct subgroup analysis according to age and HTO types, etc. Furthermore, significant high heterogeneity was observed among the studies.

Conclusion

Our meta-analysis shows that UKA may be associated with reduced postoperative pain, less postoperative complication and good knee function, while the HTO procedure showed superior ROM and less revision rate. Both surgery options yielded satisfactory results, and the treatment options should be carefully considered based on appropriate clinical indications. Further high-quality and large-scale studies are necessary to confirm these findings.

Supplemental Material

Supplemental Material - Unicompartmental knee arthroplasty versus high tibial osteotomy for medial knee osteoarthritis: A systematic review and meta-analysis

Supplemental Material for Unicompartmental knee arthroplasty versus high tibial osteotomy for medial knee osteoarthritis: A systematic review and meta-analysis by Bin Zhang, Hanguang Qian, Hongfu Wu and Xiaofei Yang in Journal of Orthopaedic Surgery

Footnotes

Author contributions

BZ conceived and coordinated the study, designed, performed and analyzed the experiments, wrote the paper. HGQ, HFW and XFY carried out the data collection, data analysis, and revised the paper. All authors reviewed the results and approved the final version of the manuscript.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Data availability

All data generated or analysed during this study are included in this published article [and its supplementary information files].

ORCID iD

Bin Zhang

Supplemental Material

Supplemental material for this article is available online.

Appendix

References

Liu

Chen

Gao

, et al. High tibial osteotomy: review of techniques and biomechanics. J Healthc Eng 2019; 2019: 8363128. DOI: 10.1155/2019/8363128.

Sirik

. Evaluation of the relationship between the femoro-tibial angle and meniscal injury. North Clin Istanb 2017; 4: 242–246. DOI: 10.14744/nci.2017.79847.

Sharma

Song

Felson

, et al. The role of knee alignment in disease progression and functional decline in knee osteoarthritis. JAMA 2001; 286: 188–195. DOI: 10.1001/jama.286.2.188.

Jin

Song

Santoso

, et al. Survival and risk factor analysis of medial open wedge high tibial osteotomy for unicompartment knee osteoarthritis. Arthroscopy, 2020; 36: 535–543. DOI: 10.1016/j.arthro.2019.08.040.

Nwachukwu

McCormick

Schairer

, et al. Unicompartmental knee arthroplasty versus high tibial osteotomy: United States practice patterns for the surgical treatment of unicompartmental arthritis. J Arthroplasty 2014; 29: 1586–1589. DOI: 10.1016/j.arth.2014.04.002.

Brouwer

Huizinga

Duivenvoorden

, et al. Osteotomy for treating knee osteoarthritis. Cochrane Database Syst Rev 2014; 2014: CD004019. DOI: 10.1002/14651858.CD004019.pub4.

Vasso

Antoniadis

Helmy

. Update on unicompartmental knee arthroplasty: current indications and failure modes. EFORT Open Rev 2018; 3: 442–448. DOI: 10.1302/2058-5241.3.170060.

Sabzevari

Ebrahimpour

Roudi

, et al. High tibial osteotomy: a systematic review and current concept. Arch Bone Jt Surg 2016; 4: 204–212.

Jennings

Kleeman-Forsthuber

Bolognesi

. Medial unicompartmental arthroplasty of the knee. J Am Acad Orthop Surg 2019; 27: 166–176. DOI: 10.5435/JAAOS-D-17-00690.

10.

Crawford

Berend

Thienpont

. Unicompartmental knee arthroplasty: US and global perspectives. Orthop Clin North Am 2020; 51: 147–159. DOI: 10.1016/j.ocl.2019.11.010.

11.

Zuiderbaan

van der List

Kleeblad

, et al. Modern indications, results, and global trends in the use of unicompartmental knee arthroplasty and high tibial osteotomy in the treatment of isolated medial compartment osteoarthritis. Am J Orthop 2016; 45: E355–e361.

12.

Herman

Giffin

. High tibial osteotomy in the ACL-deficient knee with medial compartment osteoarthritis. J Orthop Traumatol 2016; 17: 277–285. DOI: 10.1007/s10195-016-0413-z.

13.

Wang

Zeng

She

, et al. Is opening-wedge high tibial osteotomy superior to closing-wedge high tibial osteotomy in treatment of unicompartmental osteoarthritis? a meta-analysis of randomized controlled trials. Int J Surg 2018; 60: 153–163. DOI: 10.1016/j.ijsu.2018.10.045.

14.

Longo

Loppini

Trovato

, et al. No difference between unicompartmental versus total knee arthroplasty for the management of medial osteoarthtritis of the knee in the same patient: a systematic review and pooling data analysis. Br Med Bull 2015; 114: 65–73. DOI: 10.1093/bmb/ldv009.

15.

Cao

Mai

Wang

, et al. Unicompartmental knee arthroplasty vs high tibial osteotomy for knee osteoarthritis: a systematic review and meta-analysis. J Arthroplasty 2018; 33: 952–959. DOI: 10.1016/j.arth.2017.10.025.

16.

Santoso

. Unicompartmental knee arthroplasty, is it superior to high tibial osteotomy in treating unicompartmental osteoarthritis? a meta-analysis and systemic review. J Orthop Surg Res 2017; 12: 50. DOI: 10.1186/s13018-017-0552-9.

17.

Han

Kyung

Seo

, et al. Better clinical outcomes after unicompartmental knee arthroplasty when comparing with high tibial osteotomy. Medicine 2017; 96: e9268. DOI: 10.1097/md.0000000000009268.

18.

Hao

Haibiao

Xiaoqiang

, et al. A meta-analysis of high tibial osteotomy and monocondylar replacement for treating medial interventricular osteoarthritis of the knee. Chinese J. Tissue Eng. Res 2020; 24: 4905–4913.

19.

Liu

Luo

, et al. Meta analysis of unilateral condylar replacement and high tibial osteotomy in the treatment of medial compartment osteoarthritis of the knee. Zhongguo Gu Shang 2018; 31: 1156–1163. DOI: 10.3969/j.issn.1003-0034.2018.12.017.

20.

Song

Bae

Kim

, et al. Long-term survival is similar between closed-wedge high tibial osteotomy and unicompartmental knee arthroplasty in patients with similar demographics. Knee Surg Sports Traumatol Arthrosc 2019; 27: 1310–1319. DOI: 10.1007/s00167-019-05390-w.

21.

Higgins

JPT

Altman

Gotzsche

, et al. The cochrane collaboration’s tool for assessing risk of bias in randomised trials. BMJ 2011; 343: d5928. DOI: 10.1136/bmj.d5928.

22.

Sterne

JAC

Hernán

Reeves

, et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ 2016; 355: i4919. DOI: 10.1136/bmj.i4919.

23.

Higgins

JPT

Thompson

. Quantifying heterogeneity in a meta-analysis. Stat Med 2002; 21: 1539–1558. DOI: 10.1002/sim.1186.

24.

Higgins

JPT

Thompson

Deeks

, et al. Measuring inconsistency in meta-analyses. BMJ 2003; 327: 557–560. DOI: 10.1136/bmj.327.7414.557.

25.

Börjesson

Weidenhielm

Mattsson

, et al. Gait and clinical measurements in patients with knee osteoarthritis after surgery: a prospective 5-year follow-up study. Knee 2005; 12: 121–127. DOI: 10.1016/j.knee.2004.04.002.

26.

Stukenborg-Colsman

Wirth

Lazovic

, et al. High tibial osteotomy versus unicompartmental joint replacement in unicompartmental knee joint osteoarthritis: 7-10-year follow-up prospective randomised study. Knee 2001; 8: 187–194. DOI: 10.1016/s0968-0160(01)00097-7.

27.

Ziqi

Yufeng

Lei

, et al. Therapeutic effects comparison and revision case analysis of unicompartmental knee arthroplasty and open wedge high tibial osteotomy in treating medial knee osteoarthritis in patients under 60 years: a 2-6-year follow-up study. Orthop Surg 2020; 12(6): 1635–1643. DOI: 10.1111/os.12761.

28.

Yuntao

Jianyong

Hongwu

, et al. Comparison of the efficacy of unicompartmental knee arthroplasty and high tibial osteotomy in the treatment of knee osteoarthritis. Jilin Medical 2019; 40: 20200430.

29.

Lizhi

Qiu

, et al. Comparison of short-term effectiveness between high tibial osteotomy and unicompartmental knee arthroplasty for medial compartment osteoarthritis. Chin J Integr Med 2019; 29: 20200125.

30.

Yong

Zhao

Yunsheng

, et al. Comparison of the clinical efficacy of unicompartmental knee arthroplaty and high tibial osteotomy in the treatment of knee osteoarthritis. Orthopaedic Biomechanics Materials and Clinical Study 2015; 12(2): 75–77. DOI: 10.3969/j.issn.1672-5972.2015.02.021

31.

Shuai

Jie

Zheng

, et al. Early outcome of two operative methods for knee medial compartment osteoarthritis. J. Pract. Orthod 2018; 24: 20190520.

32.

Rulin

Mingwu

Hongjun

, et al. Comparison of UKA and arthroscopy combined with HTO in the treatment of medial compartment osteoarthritis of the knee. J. Nanchang Univ (Med Sci) 2019; 59: 20200125.

33.

Ruihong

Hongqing

Wenzhi

. Comparison of one-year follow-up effectiveness between high tibial osteotomy and unicompartmental knee arthroplasty for treating medial compartment osteoarthritis of the knee. Chinese J. Tissue Eng. Res 2019; 23: 3143–3147.

34.

Qingchen

Fenglong

. Clinical observation of different surgical methods for treatment of medial compartment osteoarthritis of knee. Med. Recapitulate 2019; 25: 20190805.

35.

Ping

Jiafeng

Jiajin

, et al. Comparing the early clinical efficacy of three operation modalities in medial knee osteoarthritis. J. Wannan Med College 2019; 38: 20191110.

36.

Pengli

Jianfang

Kai

, et al. Comparison of three kinds of surgical methods in the treatment of medial compartment knee osteoarthritis. Medical J. Wuhan Univ 2019; 40(5): 801–806. DOI: 10.14188/j.1671-8852.2018.0952.

37.

Lin

Jin

, et al. Comparison of clinical and radiological outcomes between opening-wedge and closing-wedge high tibial osteotomy: a comprehensive meta-analysis. PLoS One 2017; 12: e0171700. DOI: 10.1371/journal.pone.0171700.

38.

Kim

Lee

. Survival of opening versus closing wedge high tibial osteotomy: a meta-analysis. Sci Rep 2017; 7: 7296. DOI: 10.1038/s41598-017-07856-8.

39.

Vasso

Corona

D’Apolito

, et al. Unicompartmental knee arthroplasty: modes of failure and conversion to total knee arthroplasty. Joints 2017; 5: 44–50. DOI: 10.1055/s-0037-1601414.

40.

Johal

Nakano

Baxter

, et al. Unicompartmental knee arthroplasty: the past, current controversies, and future perspectives. J Knee Surg 2018; 31: 992–998. DOI: 10.1055/s-0038-1625961.

41.

Krych

Reardon

Sousa

, et al. Unicompartmental knee arthroplasty provides higher activity and durability than valgus-producing proximal tibial osteotomy at 5 to 7 years. J Bone Joint Surg Am 2017; 99: 113–122. DOI: 10.2106/JBJS.15.01031.

42.

Bouguennec

Mergenthaler

Gicquel

, et al. Medium-term survival and clinical and radiological results in high tibial osteotomy: factors for failure and comparison with unicompartmental arthroplasty. Orthop Traumatol Surg Res 2020; 106: S223–S230. DOI: 10.1016/j.otsr.2020.08.002.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.63 MB