Impact of Wiberg patellar morphology on patient-reported outcomes following unicompartmental knee arthroplasty: A 3 ~ 5 year retrospective cohort analysis

Abstract

Background

While unicompartmental knee arthroplasty (UKA) effectively treats medial compartment osteoarthritis, patient-reported outcomes remain variable. The patellofemoral joint is a recognized source of persistent symptoms, yet the influence of specific patellar morphologies, as classified by the Wiberg system, on functional recovery and satisfaction is not well defined. This study aimed to evaluate the impact of Wiberg patellar types on patient-reported outcomes following UKA.

Methods

In this retrospective cohort study, 327 consecutive patients who underwent fixed-bearing UKA between 2020 and 2023 were categorized by Wiberg type: 59 Type I, 200 Type II, and 68 Type III. Preoperative and postoperative outcomes were assessed using the Forgotten Joint Score-12 (FJS-12), Feller score (and its components: anterior knee pain, stair-climbing, sit-to-stand), Hospital for Special Surgery (HSS) score, Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), and patient satisfaction.

Results

Although all groups showed significant improvement in knee function postoperatively, distinct intergroup disparities emerged. A significant main effect of patellar type was observed for FJS-12 (F = 17.827, p < 0.001, η²=0.099) and Feller scores (F = 16.585, p < 0.001, η²=0.093). Post-hoc analysis revealed that patients with Type III patellae reported significantly inferior FJS-12 scores compared to Type II (mean difference: -6.36 points, 95% CI [-8.92, -3.80], p < 0.001) and greater anterior knee pain. Consequently, the Type III group exhibited a significantly higher dissatisfaction rate (13.2%) versus the Type II group (3.5%; p < 0.0125). No significant differences were found in HSS, WOMAC or baseline demographics among the groups.

Conclusion

UKA provides significant functional improvement across all Wiberg patellar types. However, patients with Wiberg Type III morphology reported consistently inferior patient-reported outcomes and lower satisfaction. This strong association highlights the clinical importance of preoperative identification of Type III patella, as these patients may benefit from tailored surgical planning and specific postoperative rehabilitation protocols to optimize their results.

Background

Knee osteoarthritis (OA) poses a major health burden for middle-aged and elderly populations owing to its high disability rate, and its incidence is projected to continue rising [1]. For patients with isolated medial compartment OA, unicompartmental knee arthroplasty (UKA) has emerged as an effective surgical solution [2]. The predominance of medial compartment involvement is attributed to the complex interplay of anatomical, bony, and soft-tissue factors that lead to uneven load distribution across the knee [3].

When considering surgical options, UKA presents distinct characteristics compared to total knee arthroplasty (TKA). While TKA replaces all three compartments of the knee, UKA is a bone- and ligament-preserving procedure that addresses only the diseased medial compartment. Historically, strict contraindications for UKA, as proposed by Kozinn and Scott, included conditions such as patellofemoral arthritis and an incompetent anterior cruciate ligament (ACL) [4, 5] However, with advancements in surgical techniques and prosthetic design, the functional outcomes and satisfaction following UKA have been shown to be comparable to TKA, with UKA offering specific advantages such as earlier postoperative ambulation, faster recovery, and better preservation of native knee kinematics [5–7].

The choice of prosthesis in UKA, primarily between mobile-bearing and fixed-bearing designs, involves a trade-off. Fixed-bearing prostheses are associated with a simpler surgical technique and lower risk of bearing dislocation, whereas mobile-bearing designs may offer superior flexion-extension mobility [8]. Beyond implant selection, postoperative outcomes are influenced by multiple patient-specific factors. Notably, the status of the patellofemoral joint remains a critical consideration. Recent evidence suggests that mild patellofemoral degeneration may not be an absolute contraindication for UKA [9]. Nevertheless, the specific morphology of the patella, classified by the Wiberg system, may impact functional results. The Wiberg classification categorizes patellar shapes into three types: Type I (symmetrical medial and lateral facets), Type II (a dominant lateral facet), and Type III (a severely hypoplastic or absent medial facet). Previous research indicates that a Wiberg Type III patella, characterized by inherent instability, may be associated with poorer outcomes; for instance, Sun et al. reported that Type III patellae in conjunction with mobile-bearing platforms demonstrated inferior function, and patellofemoral instability has been identified as a cause of failure in fixed-bearing UKA [10, 11].

Despite these insights, a direct investigation into the influence of different Wiberg patellar types on postoperative knee function and patient satisfaction following UKA is lacking. Therefore, this study aimed to address this gap in knowledge. We retrospectively collected clinical data from 327 patients who underwent UKA and, through an average follow-up of 35 months, investigated the impact of patellar morphology on functional outcomes and satisfaction.

Methods

Design and setting

This single-center, retrospective cohort study was conducted at Department of joint orthopedics, People’s Hospital of Ningxia Hui Autonomous Region We consecutively included all patients who underwent primary unicompartmental knee arthroplasty (UKA) between January 1, 2020, and December 31, 2023. All surgeries were performed by a single senior orthopedic surgeon in arthroplasty at a high-volume institution with an annual caseload of approximately over 150 UKA procedures. (Fig. 1).

Fig. 1.

Flowchart of participants through research

Participants

The study focused on the surgical outcomes of individual knees. For patients who underwent bilateral procedures, each knee was assessed and included separately based on its own eligibility. Inclusion criteria were: (1) radiographically and clinically confirmed isolated medial knee osteoarthritis in the planned surgical knee; (2) absence of significant coronal deformity (varus ≤ 15° or valgus < 10°); (3) patellofemoral compartment with a Kellgren-Lawrence (K-L) grade < 4; and (4) implantation of a fixed-bearing unicompartmental knee arthroplasty (Link^® Gemini MK). Exclusion criteria comprised: (1) inflammatory arthropathies (e.g., rheumatoid arthritis); (2) severe coronal malalignment beyond the specified range; (3) advanced patellar degeneration deemed intraoperatively as a contraindication; or (4) loss to follow-up or revision surgery during the study period. (Fig. 2).

Fig. 2.

Wiberg Classfication. a: Knee 90° axial view b: Knee CT c: knee MRI

Rationale for prosthesis selection

The fixed-bearing Link^® Gemini MK prosthesis was selected for all cases in this cohort. This decision was based on two principal factors: firstly, the institutional preference and consistent clinical experience with this implant; and secondly, the desire to mitigate the potential risk of bearing dislocation associated with mobile-bearing designs, which could negatively impact patient satisfaction and implant longevity. (Fig. 3).

Fig. 3.

Comparison of Postoperative Feller and FJS-12 Scores by Wiberg Patellar Type. (a:feller score b：FJS-12 score * < 0.05, **< 0.01, *** < 0.001)

Radiographic evaluation and group allocation

Preoperative radiographic evaluation was performed for all patients to assess patellofemoral joint (PFJ) degeneration and to classify patellar morphology. The assessment of PFJ degeneration was conducted on standard knee radiographs (including anteroposterior, lateral, and 90° axial Merchant views) and was graded according to the Kellgren-Lawrence (K-L) classification system. As specified in the inclusion criteria, only patients with a patellofemoral K-L grade of < 4 were eligible for the study. (Fig. 4).

Fig. 4.

Feller score subgroup analysis. (a:AKP b:sit-to-sand c:stair climbing *P < 0.05, **P < 0.01, ***P < 0.001)

Patellar morphology was classified according to the Wiberg system by two independent orthopedic surgeons, both fellowship-trained in joint replacement with over 10 years of experience. The evaluation was performed using a combination of preoperative 90° axial X-rays, knee CT scans, and MRI images to obtain a comprehensive view of the articular facets. The Wiberg types were defined as follows:

• Type I: Both medial and lateral facets are essentially symmetrical and concave.

• Type II: The lateral articular surface is larger than the medial one, with a convex medial facet.

• Type III: The medial articular surface is extremely small or absent, with a predominantly convex lateral facet.

The evaluators were blinded to the clinical outcomes of the patients during this process. Initially, each surgeon performed the classification independently. In cases of disagreement, the specific images were reviewed jointly in a consensus meeting to reach a final classification. This process ensured a rigorous and unbiased allocation of patients into the three study groups (Type I, II, and III).

A medial surgical procedure apatellar approach

The consistent surgical principles included preservation of both the anterior and posterior cruciate ligaments, meticulous removal of osteophytes, and minimal bone resection. A cemented, fixed-bearing unicompartmental knee prosthesis (Link^® Gemini MK) was implanted in all cases. The target was to restore neutral limb alignment and achieve balanced flexion-extension gaps.

Results

Baseline characteristics and group comparability

The study cohort consisted of 327 patients categorized by Wiberg patellar type: 59with Type I, 200 with Type II, and 68 with Type III. Preoperative demographic and clinical characteristics were comparable across all three groups. No statisticallysignificant differences were observed in gender distribution (χ² = 0.361, p = 0.835,Cramér's V = 0.033), age (F = 0.845, p = 0.431, η² = 0.005), BMI (F = 1.444, p =0.238, η² = 0.009), or ASA classification (F = 0.029, p = 0.971, η² < 0.001).

Preoperative to Postoperative Functional Improvement

All three patellar type groups demonstrated significant and clinically important improvement from preoperative to final follow-up across key functional outcome measures. (Table 1).

Table 1.

Comparison of knee joint function scores+

		Preoperative	postoperative	t	p
HSS	I	52.63 ± 11.43	85.71 ± 6.23	19.515	< 0.001
	II	54.39 ± 11.58	84.09 ± 8.54	-28.713	< 0.001
	III	49.13 ± 11.05	84.35 ± 4.87	24.042	< 0.001
WOMAC	I	106.18 ± 19.02	44.74 ± 13.13	20.153	0.006
	II	108.06 ± 16.54	43.31 ± 11.68	32.996	0.008
	III	108.85 ± 16.29	45.01 ± 11.58	-26.336	0.039

Between-group differences in postoperative outcomes

Forgotten Joint Score-12 (FJS-12): A significant main effect of patellar type on FJS-12 was observed (F = 17.827, p < 0.001, η² = 0.099). Post-hoc analyses with Bonferroni correction revealed that patients with Type III patella reported significantly lower FJS-12 scores (75.70 ± 5.82) compared to those with Type II (82.06 ± 10.29; mean difference = -6.35 points, 95% CI [-9.32, -3.38], p < 0.001, Cohen's d = 0.78). The difference between Type I (76.62 ± 5.40) and Type II was also statistically significant (mean difference = -5.43 points, 95% CI [-8.57, -2.30], p < 0.001, Cohen's d = 0.67), while Type I and Type III did not differ significantly (mean difference = 0.92 points, 95% CI [-2.84, 4.68], p = 0.779, Cohen's d = 0.16). (Table 2).

Feller Score Components: Analysis of the Feller score components revealed distinct patterns among patellar types (Table 1). Post-hoc analyses of significant findings indicated that Type III patellae were associated with greater anterior knee pain (e.g., vs. Type II: mean difference = 1.21 points, 95% CI [0.27, 2.15], p < 0.001), worse stair-climbing ability (e.g., vs. Type II: mean difference = 0.27 points, 95% CI [0.05, 0.50], p = 0.01), and poorer sit-to-stand capacity (e.g., vs. Type II: mean difference = 0.71 points, 95% CI [0.33, 1.09], p < 0.001) compared to other types. “Although one-way ANOVA indicated an overall difference in quadriceps strength (p = 0.034), post-hoc pairwise comparisons revealed no significant differences between any specific groups after Bonferroni correction, suggesting no clear superiority or inferiority among patellar types.“(Table 1).

Table 2.

Comparison of general patient data and knee function score

		I(n = 59)	II(n = 200)	III(n = 68)	t/x2	P
gender	man	21(35.6%)	64(32%)	21(30.88%)	0.361	0.835
	female	38(64.4%)	136(68%)	47(69.12%)
satisfied	satisfied	56(94.92%)	193(96.5%)	59(86.76%)	—	0.018*
	not satisfied	3(5.08%)	7(3.5%)	9(13.24%)
ASA	2	11(18.64%)	39(19.5%)	12(17.65%)	0.118	0.943
	3	48(81.36%)	161(80.5%)	56(82.35%)
BMI(kg/m2)		25.87 ± 2.50	26.50 ± 3.02	25.96 ± 2.77	1.444	0.238
Age(y)		64.90 ± ± 7.63	65.40 ± 7.89	63.68 ± 7.37	1.253	0.287
Symptom Duration		2.67 ± 2.92	6.16 ± 2.85	5.68 ± 2.81	1.11	0.331
HSS		85.71 ± 6.23	84.09 ± 8.94	84.35 ± 4.87	0.989	0.373
WOMAC		44.37 ± 13.04	43.31 ± 11.68	45.01 ± 11.58	0.583	0.559
FJS-12		76.62 ± 5.40	82.06 ± 10.29	75.70 ± 5.82	17.827	< 0.001
Feller		24.83 ± 3.61	26.06 ± 2.79	23.71 ± 3.06	16.585	< 0.001

*:Fisher’s exact test; “—” = Not

Knee functional scores: No statistically significant differences were observed in postoperative HSS scores among the three patellar types (F = 0.989, p = 0.373, η² = 0.006). Similarly, postoperative WOMAC scores did not differ significantly across patellar types (F = 0.583, p = 0.559, η² = 0.004).(Table 1).

Patient Satisfaction:Patient satisfaction ratings differed significantly across the patellar types (p = 0.018). Post-hoc comparisons using Fisher's exact test with Bonferroni correction (significance threshold p < 0.0125) revealed that dissatisfaction was significantly more prevalent in the Type III group (13.2%) compared to the Type II group (3.5%; p < 0.0125, OR = 4.18). The differences between Type III and Type I (5.1%; p > 0.0125), and between Type I and Type II (p > 0.0125) were not statistically significant. (Table 3.)

Table 3.

Comparison of specific contents of feller score

	I(n = 59)	II(n = 200)	III(n = 68)	t	P
AKP	12.17 ± 3.73	12.9 ± 2.62	11.69 ± 2.94	4.925	0.008
Quadrice muscle strength	4.63 ± 0.78	4.84 ± 0.54	4.68 ± 0.74	3.42	0.034
Sit-To-Stand	3.78 ± 1.34	4.78 ± 1.03	4.47 ± 1.15	11.221	< 0.001
Stair Climbing	4.05 ± 0.68	4.14 ± 0.65	3.87 ± 0.69	4.294	0.014

AKP: anterior knee pain

Summary of Key Findings:In summary, while all patients demonstrated significant functional improvement following UKA regardless of patellar type:Patients with Type III patellae reported inferior joint awareness (FJS-12), specific functional limitations (stair climbing, sit-to-stand), and lower satisfaction rates.Type II patellae were associated with the most favorable outcomes across multiple measures.No significant between-group differences were found in overall knee function (HSS) or osteoarthritis-specific symptoms (WOMAC). (Table 4).

Table 4.

Satisfaction comparison results

	satisfactied	not satisfactied	x2	p
I-II	56–193	3–7	-	> 0.0125*
I-III	56–59	3–9		> 0.0125
II-III	193 − 59	7–9		< 0.0125*

*:Fisher’s exact test; “—” = Not

Statistical methods

Statistical analyses were conducted using SPSS 26.0 and GraphPad Prism 10.1.2. The normality of continuous data was assessed with the Shapiro-Wilk test. Normally distributed data are presented as mean ± standard deviation and analyzed using independent samples T-tests or one-way ANOVA, with Bonferroni-corrected post-hoc comparisons (p < 0.0125) where appropriate. Non-normally distributed data were analyzed using the Kruskal-Wallis test with Dunn’s post-hoc test. Categorical variables are presented as counts (%) and compared using Pearson’s chi-square or Fisher’s exact test. In line with contemporary standards, effect sizes (Cohen’s d for pairwise comparisons, η² for ANOVA, ε² for Kruskal-Wallis, and Cramér’s V for chi-square) are reported for significant findings. P < 0.05 was considered statistically significant for all omnibus tests.

Discussion

This study demonstrates that while UKA effectively improves knee function across all Wiberg patellar types with a 96.9% short-term prosthetic survival rate, patients with Type III morphology report significantly inferior outcomes in specific patient-reported measures. These include FJS-12 scores, anterior knee pain, stair-climbing ability, sit-to-stand capacity, and overall satisfaction, despite comparable improvement in traditional functional scores.The anatomical basis for these findings lies in the distinctive morphology of Type III patellae. First described by Wiberg in 1941, this classification reflects significant alterations in patellar surface geometry. The Type III patella’s severely hypoplastic or absent medial facet leads to disrupted patellofemoral congruence and substantially increased lateral pressure [12, 13], predisposing to lateral patellar hyperpressure syndrome and full-thickness cartilage loss.

Our findings provide a more nuanced Understanding of how this morphology impacts UKA outcomes. While sun et al. reported that patients with type III patellae demonstrated poorer outcomes following mobile-bearing UKA [11], our study utilizing fixed-bearing prostheses reveals a critical distinction

although all patellar types achieved comparable improvement in traditional functional scores (HSS and WOMAC), the Type III group reported significantly poorer patient-reported outcomes, including lower FJS-12 scores and higher rates of anterior knee pain.This divergence suggests that conventional scoring systems may be insufficiently sensitive to detect patellofemoral-specific concerns in UKA patients. In contrast, the FJS-12, being free from ceiling effects and specifically designed to assess joint awareness [14], proved more responsive to these morphological differences, supporting its validation as a core outcome measure following knee arthroplasty [15].

Our analysis reveals a distinct functional hierarchy among Wiberg patellar types. While Type II patellae demonstrated optimal outcomes, both Type I and Type III showed inferior FJS-12 scores compared to Type II. Importantly, the functional deficit in Type III was substantially greater and was uniquely associated with significantly higher dissatisfaction rates (13.2% vs. 3.5% in Type II). This pattern suggests that although Type I represents a less biomechanically efficient morphology than Type II, its functional impact remains below the clinical threshold for affecting overall satisfaction. In contrast, the pronounced functional limitations in Type III patellae directly translate to patient-reported dissatisfaction, underscoring its clinical significance in UKA outcomes.

Notably, our findings regarding patellar degeneration challenge certain conventional perspectives. While Sun et al. considered patellar degeneration a contraindication for fixed platforms [11].our research demonstrates that fixed platforms can significantly improve knee function even in the presence of mild to moderate patellar degeneration, which was present in nearly all our osteoarthritis patients. This approach proved feasible, as evidenced by our 94% satisfaction rate at average 35-month follow-up, consistent with established satisfaction trajectories [16].Several factors may explain the suboptimal outcomes in Type III patellae. First, and most importantly, our analysis confirms that quadriceps strength deficits are not responsible for the observed differences, as no significant intergroup variations were found. Instead, the inherent anatomical disruption of Type III patellae likely requires extended neuromuscular adaptation periods, while the pronounced lateral facet prominence may cause persistent abnormal tracking despite surgical intervention.“, which showed no significant intergroup differences. Rather, the inherent anatomical disruption requires extended adaptation periods, and the pronounced lateral facet prominence may cause abnormal tracking despite surgical intervention.“Konan et al.‘s reported correlation between degenerative changes on the lateral articular surface and poorer knee joint function [17]supports this mechanical explanation. Interestingly, our Feller score analysis revealed no significant differences in quadriceps strength among groups, suggesting that the inferior outcomes in Type III patellae are related to morphological factors rather than muscular deficiency.

Beyond morphological considerations, we emphasize that postoperative rehabilitation remains crucial. While knee replacement significantly improves quality of life, it may accelerate declines in muscle strength and mass [18], potentially leading to sarcopenia [19].which has been linked to periprosthetic infections [16]. Our follow-up revealed knee weakness as the second most frequent symptom after pain, underscoring the importance of quadriceps training. Incorporating progressive resistance training during the first 4 weeks post-surgery is essential [20], an approach consistent with our clinical practice.

While this study provides valuable insights into the impact of patellar morphology on UKA outcomes, several limitations should be acknowledged. First, the intermediate-term follow-up (average 35 months) primarily captures short-to-midterm satisfaction and functional outcomes. The high satisfaction rates observed likely reflect the substantial surgical experience at our high-volume center, but longer-term surveillance is necessary to determine whether the inferior patient-reported outcomes in Type III patellae persist or evolve over time, particularly regarding implant longevity and the progression of patellofemoral degenerative changes. Second, the exclusive use of a single fixed-bearing prosthesis design, while ensuring consistency, limits the generalizability of our findings to mobile-bearing UKA systems. Third, the single-center, retrospective design may introduce selection bias, though our consecutive inclusion of patients and blinded radiographic assessments helped mitigate this concern. Future multicenter studies with longer follow-up periods and diverse implant designs would help validate these findings and further elucidate the long-term clinical significance of Wiberg classification in UKA patient selection and outcomes.

In conclusion, UKA effectively alleviates knee osteoarthritis symptoms across all Wiberg classifications. However, patients with Type III patellae demonstrate inferior patient-reported outcomes, particularly regarding joint awareness, anterior knee pain, and satisfaction with specific functional activities. These findings highlight the importance of preoperative recognition of patellar morphology. Orthopaedic surgeons should consider implementing morphology-specific rehabilitation protocols and patient education for those with Type III patellae to optimize functional outcomes and satisfaction following UKA.

Abbreviations

UKA

Unicompartmental KneeArthroplasty

TKA

Total Knee Arthroplasty

BMI

Body Mass Index

ASA

American Society of Anesthesiologists

HSS

Hospital for Special Surgery Knee

WOMAC

Western Ontario and McMaster universities osteoarthritis index

FJS-12

Forgotten Joint Score-12

AKP

Anterior knee pain

ACL

Anterior Cruciate Ligament

K-L

Kellgren-Lawrence grading scale

ELPS

Lateral patellar hyperpressure syndrome

PFJ

patellofemoral joint

Authors’ contributions

Wang Cong wrote the main manuscript text , Ma Feiyang and Wang Zhencheng visit patient and collect basic information, Bai Zhigang operate surgery, Liang Yuqi responsible for the most popular review. All authors reviewed the manuscript.

Funding

This trial was partially funded by the Ningxia Health Commission (No. 2025-NWZC-B006). Ningxia Science and Technology Transformation Project 2025 (Application and promotion of remote precision medical system for hip and knee joints based on surgical robots: NO.2025CJE09011)The funding source did not play any role in the study design. See Additional.

Data availability

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

Declarations

Ethics approval and consent to participate

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of Ningxia Hui Autonomous Region(Approval No.2025-WJW-004). Informed consent was obtained from all individual participants included in the study.

Consent for publication

Not Applicable.

Competing interests

The authors declare no competing interests.