Factors influencing preoperative outcome expectations in patients undergoing total knee arthroplasty: a cross-sectional observational study

Abstract

Background

Preoperative outcome expectations are closely associated with postoperative satisfaction and recovery in patients undergoing total knee arthroplasty (TKA). However, the relationships among knee function, psychological factors, and outcome expectations remain insufficiently understood. This study aimed to examine the associations among knee function, pain catastrophizing, anxiety/depression, and preoperative outcome expectations in patients undergoing TKA using structural equation modeling (SEM).

Methods

A cross-sectional study was conducted among patients with severe osteoarthritis undergoing primary TKA at three tertiary hospitals in Changsha, China, between September 2023 and May 2024. Participants completed the Hospital for Special Surgery Total Knee Replacement Expectations Survey (HSS-TKRES), Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), Pain Catastrophizing Scale (PCS), and Hospital Anxiety and Depression Scale (HADS). SEM was used to evaluate direct and indirect associations among variables.

Results

A total of 203 participants were included. The mean preoperative outcome expectations score was 63.58 ± 9.98, indicating a moderate level of expectation. HSS-TKRES scores were positively correlated with WOMAC scores ( r = 0.229, P < 0.01) and negatively correlated with PCS ( r = − 0.338, P < 0.001) and HADS scores ( r = − 0.362, P < 0.001). In the SEM, WOMAC was positively associated with HSS-TKRES ( β = 0.398), whereas HADS ( β = −0.362) and PCS ( β = −0.193) were negatively associated with HSS-TKRES. WOMAC also showed significant indirect associations with HSS-TKRES through PCS, HADS, and the sequential pathway WOMAC → PCS → HADS → HSS-TKRES.

Conclusion

Preoperative outcome expectations among patients undergoing TKA were associated with symptom severity and psychological distress. Worse knee function was associated with higher expectations, whereas greater anxiety/depression and pain catastrophizing were associated with lower expectations. Given the cross-sectional design, these findings should be interpreted as associational rather than causal. Comprehensive preoperative assessment and expectation-focused communication may help support more realistic and adaptive expectations before TKA.

Introduction

Knee osteoarthritis (KOA) represents a chronic degenerative disorder that impacts the joints, characterized chiefly by pain, inflammation, stiffness, and diminished functionality of the knee joint [1]. In Korea, the prevalence rate among individuals aged 65 and older can attain levels ranging from 33.6% to 38% [2]. In China, the occurrence of osteoarthritis stands at 8.1%, with a notable increase to 11.1% among individuals aged 60 and above [3]. The ailment is characterized by a significant rate of disability, a high incidence, a propensity for recurrence, and challenges in achieving a cure, resulting in considerable economic strain on patients, their families, and society at large [4]. Total knee arthroplasty (TKA) is acknowledged as a proficient intervention for mitigating knee discomfort, augmenting joint functionality, and elevating the overall quality of life for numerous individuals. TKA is typically reserved for end-stage knee osteoarthritis in patients with substantial pain and functional disability after failure of conservative management, making expectation management particularly important in this population [5]. Notwithstanding its efficacy, a notable 16% to 20% of patients may continue to encounter dissatisfaction with their postoperative results [6]. Many studies have tried to identify the elements that contribute to this heightened level of dissatisfaction. A growing collection of evidence indicates that preoperative outcome expectations play a significant role in influencing patient dissatisfaction with TKA [7, 8]. Previous studies have demonstrated that patients often hold unreasonably high and unrealistic expectations regarding TKA [9]. Tilbury et al. [10] conducted an analysis of the preoperative outcome expectations of 322 patients who underwent TKA, evaluating the degree to which these expectations were fulfilled one-year post-surgery. It was observed that more than 30% of the patients’ anticipated outcomes were not achieved, particularly in relation to “being able to kneel down” (44%) and “being able to squat” (47%). Mannion AF et al. [11] demonstrated that 85% of patients anticipated being pain-free following TKA, while 52% of patients expected to experience no functional limitations post-TKA. The anticipations of patients regarding surgical outcomes tend to surpass those held by physicians [12]. Therefore, it is essential to consider the anticipated outcomes for patients undergoing TKA.

Preoperative outcome expectations represent the psychological aspirations that individuals develop, grounded in their abilities and experiences, and embody their ambition to attain objectives [13]. For individuals undergoing TKA, these anticipations relate to their expected alleviation of pain and enhancement of joint functionality post-surgery, and are associated with postoperative satisfaction [14]. A comprehensive analysis of 46 studies demonstrated a noteworthy positive correlation between optimistic preoperative outcome expectations and enhanced postoperative results, encompassing diminished pain, improved functionality, reduced disability, and increased satisfaction among patients who underwent total knee arthroplasty [15]. Furthermore, the preoperative outcome expectations of TKA patients are associated with by various circumstances. Individuals with impaired joint function are more inclined to anticipate surgical intervention to alleviate their existing pain and possess elevated expectations for the surgical outcome [16]. Numerous studies have demonstrated a significant association between patient psychological characteristics and their expectations regarding TKA outcomes [17, 18]. The phenomenon of pain catastrophizing, characterized by an exaggerated focus on and amplification of pain experiences, serves as a significant psychological determinant that is adversely related to pain expectations and outcomes following TKA [19]. Individuals exhibiting pain catastrophizing tendencies are at an elevated risk for the persistence of chronic postoperative pain and the subsequent deterioration of joint function. Furthermore, the cognitive-emotional response to pain is deeply interconnected with, and frequently intensifies, anxiety and depression [20]. This indicates a bidirectional relationship in which catastrophizing may exacerbate these adverse emotions, while elevated levels of anxiety and depression can, in turn, heighten pain catastrophizing. In summary, a relationship exists between the preoperative knee function of patients, psychological factors—including anxiety, depression, and pain catastrophizing—and their expectations regarding outcomes. However, previous studies have seldom examined the connection between these variables.

Structural equation modeling (SEM) serves as a sophisticated multivariate statistical analysis instrument that elucidates, and measures intricate relationships among variables [21]. The current literature has examined multiple factors associated with preoperative outcome expectations of patients undergoing total knee arthroplasty, such as pain levels, joint functionality, psychological well-being, and socio-demographic factors. Although prior studies have identified associations between symptom severity, psychological factors, and expectations in TKA, most have examined these variables in isolation or using conventional regression models. This limits understanding of how symptom burden and psychological processes may jointly relate to expectations. By applying structural equation modeling (see the hypothetical model in Fig. 1), the present study evaluates a theory-informed network of associations and decomposes direct and statistical indirect effects among WOMAC, pain catastrophizing, anxiety/depression, and expectations. This approach may provide a more integrated description of how these factors cluster in the preoperative period and help identify potential targets for preoperative communication and support.

Fig. 1.

Hypothesized pathways from WOMAC to HSS-IkRES in the conceptlal model. HSS-TKRES: Hospital for Special Surgery Total Knee Replacement Expectations Survey: WOMAC: Westem Ontario and McMaster Universities Osteoarthritis Index; PCS; Pain Catastrophizing Scale; HADS: Hospital Anxiety Depression Scale.

Materials and methods

Study design

The study population comprised patients undergoing total knee arthroplasty who were admitted to the orthopedic departments of three tertiary A-level general hospitals in Changsha between September 2023 and May 2024. The selection process utilized convenience sampling. Inclusion criteria: ① Individuals aged 18 years or older; ② Those diagnosed with knee osteoarthritis who were scheduled to undergo primary total knee arthroplasty.; ③ Ability to communicate reliably and complete questionnaires with or without assistance; ④ Informed consent and voluntary participation. Criteria for exclusion: ① Individuals with significant cardiovascular, neurological, renal, or other serious conditions; ② In practice, cognitive deficits or psychological disorders were identified based on documented medical history in the electronic medical record (e.g., diagnosis of dementia, delirium, severe psychiatric illness) or clinician judgment during preoperative assessment indicating inability to provide reliable responses. Recommended sample size for SEM depends on model complexity and estimation method; as a rule of thumb, a minimum of 5–10 participants per estimated parameter is often suggested, and samples of approximately 200 are commonly considered adequate for stable estimation in many applications [22]. In this study, 203 patients scheduled for TKA were selected as the research subjects. In addition, this study has been approved by the hospital ethics committee (NO. 2023-34).

To guarantee the accuracy and consistency of the survey methodology, the principal investigator, accompanied by a cadre of skilled research coordinators and nurses, facilitated extensive training sessions. The sessions concentrated on comprehending fundamental concepts, the appropriate application of survey instruments, and the ethical implications of securing informed consent. The research team conducted the survey in person. Patients were given paper questionnaires after hospital admission, within 1–3 days prior to surgery. Written informed consent was obtained from all participants before questionnaire completion.The research team reviewed the questionnaires immediately after completion to ensure data completeness, ensuring that every question was addressed and providing support, as necessary. For individuals who faced difficulties in understanding or completing the questionnaire due to physical limitations, the research team, in conjunction with family members when appropriate, facilitated increased participation. 211 questionnaires were disseminated in this survey aimed at patients preceding Total Knee Arthroplasty (TKA). Of the returned questionnaires, 8 were considered invalid, including 3 with inconsistent responses and 5 with clear indications of random completion. Consequently, a total of 203 questionnaires were included in the final analysis (see Fig. 2), yielding an effective response rate of 96.2%.

Fig. 2.

Flowchart of participant selection

Dependent measure

Chinese version of the Hospital for Special Surgery knee arthroplasty patient expectation questionnaire

The research employed the Chinese adaptation of the Hospital for Special Surgery Total Knee Arthroplasty Expectations Survey (HSS-TKRES) as formulated by Wang et al. [23]. This scale evaluates the anticipated outcomes for patients who have experienced knee arthroplasty. The assessment comprises 19 items and employs a 5-point Likert scale, which ranges from 0 points indicating “I do not have this expectation, or this expectation does not apply to me” to 4 points denoting “return to normal or completely improved”. The overall score was converted to a percentage format to facilitate analysis and comprehension; the patient’s anticipated outcomes improved alongside the rising score. In this analysis, a score exceeding 70 is classified as an important level, a score ranging from 30 to 70 is deemed a moderate level, and a score below 30 is categorized as a low level. The Cronbach’s α coefficient for this scale was determined to be 0.881 in this study.

General information questionnaire

The collected sociodemographic information included a range of factors, such as such as age, gender, body mass index (BMI), marital status, education level, economic level, address, living style, number of children, working status, economic situation, KOA Duration, and Number of co-morbidities. Living style is characterized by the individuals with whom one cohabitates, encompassing arrangements such as such as “with spouse or with children or living alone.”

Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC)

The assessment of knee function in patients was conducted using the Chinese version of the scale developed by Xie et al. [24]. The assessment comprises three dimensions: pain, stiffness, and the challenges faced in performing daily activities, encompassing 24 items. The Likert 5-point scale is employed to evaluate the outcomes, where 0 signifies “no” issue and 4 indicates a “very serious” issue. The overall score ranges from 0 to 96, with severe points exceeding 48, moderate points falling between 21 and 48, and mild points being below 21. The Chinese iteration of the WOMAC exhibited a commendable level of reliability, evidenced by a Cronbach’s alpha of 0.922.

Pain Catastrophizing Scale (PCS)

A Chinese adaptation of the scale created by Yap et al. [25] was employed to assess the patient’s negative emotions and cognitive responses to actual or expected pain. The scale comprises three dimensions: rumination, consisting of four items; exaggeration, with three items; and helplessness, which includes six items, culminating in thirteen items. The scale employs a 5-point Likert rating system, wherein 0 points denote “never”, 1 point signifies “occasionally”, 2 points indicates “moderately”, 3 points reflects “many times”, and 4 points represents “always”. The overall score ranges from 0 to 52 points. The patient attains a state of pain catastrophizing when their total score meets or surpasses 38 points; conversely, an elevated score indicates a greater involvement of pain catastrophizing. The Chinese adaptation of the Pain Catastrophizing Scale exhibited a commendable level of reliability, evidenced by a Cronbach’s alpha of 0.93 among individuals experiencing chronic pain.

Hospital Anxiety Depression Scale (HADS)

This instrument was developed to assess anxiety and depression in general hospital populations [26]. The instrument comprises two primary dimensions: anxiety, consisting of 7 items, and depression, also encompassing 7 items, totaling 14 items overall. The total scores for each subscale vary from 0 to 21, with each item receiving a rating between 0 and 3. An initial evaluation of the patient’s psychological condition can be derived from the comprehensive score: A score ranging from 0 to 7 indicates the absence of symptoms related to anxiety or depression; a score between 8 and 10 suggests the potential presence of anxiety or depression; while a score from 11 to 21 signifies a clear inclination towards anxiety or depression. The HADS demonstrates validity and reliability within the Chinese population, with both subscales exhibiting Cronbach’s alpha coefficients of 0.79 [27].

Statistical methods

The statistical analysis was performed using SPSS 26.0 and AMOS 24.0 software. Continuous variables with approximately normal distributions are presented as mean ± standard deviation, and categorical variables are presented as frequency and percentage. Group comparisons were conducted using analysis of variance, as appropriate. To ensure consistency with the structural equation modelling framework, bivariate associations among WOMAC, PCS, HADS, and HSS-TKRES were estimated in AMOS as standardized covariances from a saturated model using maximum-likelihood estimation (equivalent to Pearson correlation coefficients) and are reported as correlation coefficients (r). A structural equation model examining the factors related to preoperative outcome expectations in patients undergoing total knee arthroplasty was developed using AMOS 24.0 software. A variety of indices were employed to evaluate the model’s fit, such as the Comparative Fit Index (CFI), the normalized fit index (NFI), the goodness of fit index (GFI), the adjusted goodness of fit index (AGFI), and the root mean square error of approximation (RMSEA). The model fit was deemed satisfactory with CFI, NFI, GFI, and AGFI values at or above 0.90, a χ2/df ratio below 5, and an RMSEA of 0.08 or less [28]. The maximum likelihood method was used to estimate the model parameters, while the bias-corrected nonparametric percentile bootstrap method was applied to assess statistical indirect effects. The bootstrap procedure was performed with 5,000 resamples to obtain the effect estimates and corresponding 95% confidence intervals (95% CIs). A two-sided test was applied, with a significance level set at 0.05.

Results

General information of respondents

TKA patients’ ages fell between 38 and 82 (65.57 ± 9.29). Table 1 shows further general information.

Table 1.

Univariate analysis of general data and preoperative outcome expectations in patients with TKA (n = 203)

Variables	N (%)	HSS-TKRES score (± SD)	Statistic
			t/F	P value
Age
<60	42 (21.2)	62.97 ± 10.58	1.176	0.311
60 ~ 70	99 (48.5)	62.83 ± 10.09
>70	62 (30.3)	65.20 ± 10.85
Gender
Male	38 (18.7)	60.70 ± 9.74	−0.661	0.510
Female	92 (45.3)	61.99 ± 10.22
BMI (kg/m2)
< 28	156 (76.8)	63.22 ± 9.73	−0.920	0.359
≥ 28	47 (21.2)	64.75 ± 10.78
Education level
Primary school and below	92 (45.3)	63.56±9.73	0.733	0.533
Middle school	59 (29.1)	64.45±10.05
High school or junior college	37 (18.2)	61.63±10.26
College and above	15 (7.4)	65.08±10.77
Marital status
Married	180 (88.7)	63.18 ± 10.07	−1.602	0.111
Divorced or widowed	23 (11.3)	66.70 ± 8.78
Address
Rural area	111 (54.7)	64.40 ± 9.42	1.614	0.108
County town and Urban area	92 (45.3)	62.34 ± 10.53
Living style
With Spouse	111 (54.7)	63.86±10.14	0.158	0.853
With children	51 (25.1)	63.55±9.56
Living alone	41 (20.2)	62.84±10.23
Number of children
0	4 (2.0)	68.75±12.64	2.600	0.053
1	66 (32.5)	61.28±9.78
2	108 (53.2)	64.01±10.20
≥ 3	25 (12.3)	66.95±7.99
Working status
Yes	63 (31.0)	60.48 ± 9.39	3.024	0.003
No	140 (70.0)	64.97 ± 9.50
Economic situation (Yuan)
<3000	69 (34.0)	62.60±9.94	0.730	0.483
3000–5000	67 (33.0)	63.49±10.35
>5000	67 (33.0)	64.67±9.67
KOA Duration (year)
≤ 5	82 (40.4)	62.53±10.25	2.323	0.101
6~10	50 (24.6)	62.39±9.54
>10	71 (35.0)	65.62±9.77
Number of co-morbidities
0	154 (75.9)	63.03±10.22	1.170	0.313
1	44 (21.6)	65.61±8.99
≥ 2	5 (2.5)	62.63±10.18

TKA Total Knee Arthroplasty, BMI Body Mass Index, KOA Knee osteoarthritis, HSS-TKRES Hospital for Special Surgery Total knee arthroplasty Expectations Survey

Current status of total knee arthroplasty patient preoperative outcome expectations and univariate analysis results

In this study involving 203 TKA patients, the Chinese HSS-TKRE score was recorded as (63.58 ± 9.98). Based on the predefined cut-points (> 70 high/important, 30–70 moderate, < 30 low), the mean HSS-TKRES score in this sample indicates a moderate level of preoperative outcome expectations. Factor analysis indicated that employment status was significantly associated with overall preoperative outcome expectations among patients undergoing TKA (P < 0.05), as shown in Table 1.

Scores of WOMAC, PCS, and HADS

The descriptive statistics (mean ± SD) for WOMAC, PCS, and HADS, including their subscales, are presented in Table 2. In brief, the mean total scores were 55.54 ± 15.78 for WOMAC, 28.36 ± 9.57 for PCS, and 17.40 ± 7.00 for HADS.

Table 2.

Scores of WOMAC, PCS, and HADS (Mean ± SD)

Instrument	Domain / Dimension	Mean ± SD
WOMAC	Total	55.54 ± 15.78
	Pain	12.67 ± 3.20
	Stiffness	3.59 ± 2.15
	Physical function	39.27 ± 12.95
PCS	Total	28.36 ± 9.57
	Rumination	9.23 ± 3.07
	Magnification	7.38 ± 4.04
	Helplessness	11.78 ± 4.19
HADS	Total	17.40 ± 7.00
	Depression	8.35 ± 3.77
	Anxiety	9.05 ± 3.96

Bivariate correlations among WOMAC, PCS, HADS, and HSS-TKRES

The bivariate correlations among WOMAC, PCS, HADS, and preoperative outcome expectations (HSS-TKRES) are presented in Table 3. To maintain consistency with the structural equation modelling approach, correlations were estimated in AMOS as standardized covariances (equivalent to Pearson correlation coefficients). HSS-TKRES was positively correlated with WOMAC (r = 0.229, P < 0.01) and negatively correlated with PCS (r = − 0.338, P < 0.001) and HADS (r = − 0.362, P < 0.001). In addition, WOMAC was positively correlated with PCS (r = 0.200, P < 0.05) and HADS (r = 0.359, P < 0.01), and PCS was positively correlated with HADS (r = 0.399, P < 0.001).

Table 3.

Bivariate correlations among variables related to preoperative outcome expectations in patients undergoing TKA(r value)

Scale	HSS-TKRES	WOMAC	PCS	HADS
HSS-TKRES	1
WOMAC	0.229**	1
PCS	-0.338***	0.200*	1
HADS	-0.362***	0.359**	0.399***	1

TKA Total Knee Arthroplasty, HSS-TKRES Hospital for Special Surgery Total Knee Replacement Expectations Survey, WOMAC Western Ontario and McMaster Universities Osteoarthritis Index, PCS Pain Catastrophizing Scale, HADS Hospital Anxiety Depression Scale. *P < 0.05,**P < 0.01, ***P < 0.001

Construction of structural equation model and path analysis of factors influencing preoperative outcome expectations in patients undergoing TKA

The hypothesised structural equation model was evaluated, and two non-significant paths from work status to HADS and to preoperative outcome expectations were removed. The revised model demonstrated acceptable goodness-of-fit indices, with χ²/df = 1.810, CFI = 0.950, NFI = 0.898, GFI = 0.949, AGFI = 0.906, and RMSEA = 0.063. These indices indicate that the model showed an adequate fit to the observed data (Fig. 3). The structural equation model showed that WOMAC scores were positively associated with preoperative outcome expectations among patients undergoing TKA ( β = 0.398; 95% bias-corrected bootstrap CI [0.205, 0.947]), whereas HADS ( β = −0.362; 95% bootstrap CI [− 2.002, − 0.525]) and PCS scores ( β = −0.193; 95% bootstrap CI [− 1.133, − 0.043]) were negatively associated with these expectations.

Fig. 3.

Structural equation modeling of WOMAC, HADS, PCS, and HSS-TKRES. Model fit: x² /df=1.810, CFI =0.950, NFI = 0. 898, GFI = 0.949, AGFI=0.906, RMSEA=0.063. HSS-TKRES: Hospital for Special Surgery Total Knee Replacement Expectations Survey: WOMAC: Western Ontario and McMaster Universities Osteoarthritis Index; PCS: Pain Catastrophizing Scale; HADS: Hospital Anxiety Depression Scale

As shown in Table 4, WOMAC also demonstrated significant indirect associations with HSS-TKRES through psychological mediators. The indirect association via PCS alone (WOMAC → PCS → HSS-TKRES) was significant ( β = −0.039; 95% bootstrap CI [− 0.121, − 0.002]), as was the indirect association via HADS alone (WOMAC → HADS → HSS-TKRES) ( β = −0.101; 95% bootstrap CI [− 0.390, − 0.020]). In addition, the sequential indirect pathway WOMAC → PCS → HADS → HSS-TKRES was statistically significant ( β = −0.032; 95% bootstrap CI [− 0.178, − 0.001]). The total indirect association of WOMAC on HSS-TKRES via all mediators was negative ( β = −0.172; 95% bootstrap CI [− 0.517, − 0.049]); however, the total effect of WOMAC on HSS-TKRES remained positive and statistically significant (β = 0.226; 95% bootstrap CI [0.092, 0.586])

Table 4.

Direct and indirect associations among variables related to HSS-TKRES in patients undergoing TKA (N = 203)

Type of Effect	Path	Standard Error (S.E.)	Standardized Estimate (β)	95% Bootstrap CI (Bias-Corrected)
Direct Effects
	WOMAC → HSS-TKRES	0.210	0.398	[0.205, 0.947]
	HADS → HSS-TKRES	0.373	−0.362	[−2.002,−0.525]
	PCS → HSS-TKRES	0.279	−0.193	[−1.133,−0.043]
Indirect Effects
Path 1	WOMAC → PCS → HSS-TKRES	0.031	−0.039	[−0.121,−0.002]
Path 2	WOMAC → HADS → HSS-TKRES	0.105	−0.101	[−0.390,−0.020]
Path 3	WOMAC → PCS → HADS → HSS-TKRES	0.042	−0.032	[−0.178,−0.001]
Total Indirect Effect	Effect of WOMAC on HSS-TKRES via all mediators	0.138	−0.172	[−0.517,−0.049]
Total Effect	Total Effect of WOMAC on HSS-TKRES	0.129	0.226	[0.092, 0.586]

TKA Total Knee Arthroplasty, HSS-TKRES Hospital for Special Surgery Total Knee Replacement Expectations Survey (Outcome Expectation), WOMAC Western Ontario and McMaster Universities Osteoarthritis Index, PCS Pain Catastrophizing Scale, HADS Hospital Anxiety and Depression Scale, CI Confidence Interval

Bootstrap confidence intervals (95% CI) were derived from 5000 bootstrap samples using the bias-corrected percentile method

Discussion

Current status of TKA patient preoperative outcome expectations

The findings of our investigation indicate that the preoperative outcome expectations of patients undergoing TKA are positioned at a moderate level, reflected by an average score of (63.58 ± 9.98), suggesting a measured and pragmatic perspective regarding their forthcoming surgical intervention. This is consistent with earlier studies indicating that moderate preoperative outcome expectations correlate with improved postoperative adjustment and satisfaction [12, 17]. In our sample, preoperative outcome expectations were statistically associated with symptom burden (WOMAC) and psychological variables (pain catastrophizing, anxiety/depression), as shown by the correlation and SEM analyses. Individuals experiencing more advanced osteoarthritis often possess elevated expectations, driven by a pronounced need for alleviation of discomfort and enhancement of functionality [29]. On the other hand, individuals who engage in catastrophic thinking regarding pain or experience anxiety and depression may possess diminished expectations, stemming from a compromised sense of agency in their recovery journey or an intensified attention to possible adverse results.

Our findings extend prior literature by showing that expectations are embedded within a network of physical and psychological factors rather than reflecting a single determinant. Consistent with previous work reporting links between expectations and postoperative outcomes [14, 15], we observed that higher anxiety/depression and pain catastrophizing were associated with lower preoperative expectations, whereas greater symptom burden (WOMAC) was associated with higher expectations. Using SEM allowed us to quantify both direct and statistical indirect associations, indicating that WOMAC was positively related to expectations overall while also showing negative indirect associations through psychological distress. This pattern highlights the importance of considering both symptom severity and psychological state when discussing expectations preoperatively.

Association between WOMAC scores and preoperative outcome expectations in patients undergoing TKA

WOMAC scores were positively associated with preoperative outcome expectations in our SEM (β = 0.398, P < 0.001). Specifically, a one-unit increase in WOMAC was associated with an average 0.398-unit increase in the expectation score. This finding is consistent with Yang et al. [16], who also reported a positive association between worse knee function and higher preoperative expectations. One possible interpretation is that patients experiencing more severe pain and functional limitation may have a stronger desire for symptom relief and quality-of-life improvement through surgery, which could translate into higher anticipated benefits [30].

Moreover, our analysis suggested that WOMAC scores may be indirectly associated with preoperative outcome expectations through psychological factors, including pain catastrophizing, anxiety, and depression. Greater pain and disability may co-occur with more catastrophic pain-related cognitions [17], and persistent discomfort and impaired mobility are often accompanied by heightened emotional distress. In turn, these psychological states may be linked to more pessimistic appraisals of surgical benefit and lower expectations.

Importantly, the possibility of reverse causation deserves explicit consideration. Given the cross-sectional design and reliance on self-reported questionnaires, it is equally plausible that preoperative outcome expectations influence how patients appraise and report their current symptoms. Patients with higher expectations may engage in hopeful thinking about the upcoming surgery and consequently minimise or downplay perceived pain and disability when completing questionnaires, which is consistent with cognitive consistency theory [31, 32]. In addition, a response shift may occur in the preoperative period, whereby patients recalibrate internal standards or reinterpret the meaning of symptoms in light of anticipated improvement [33]; such shifts have been described in patients undergoing total knee arthroplasty and may complicate the interpretation of pre–post patient-reported outcomes [34].Therefore, the observed association between WOMAC and expectations should be interpreted as correlational and potentially bidirectional, rather than evidence that greater disability necessarily causes higher expectations. From a clinical perspective, these findings support the value of addressing both symptom burden and psychological factors during preoperative consultations. Screening for distress and maladaptive pain cognitions, and providing appropriate education and support, may help patients form more realistic expectations, potentially improving satisfaction and perceived outcomes after TKA [35].

Association between pain catastrophizing, anxiety, depression, and preoperative outcome expectations in TKA patients

Our results demonstrated that PCS was negatively associated with patients’ preoperative outcome expectations. This finding suggests that individuals with higher levels of pain catastrophizing may anticipate less benefit from surgery. Patients with chronic pain and greater psychosocial distress before TKA often exhibit elevated levels of catastrophizing, which are associated with poorer perceived quality of life and reduced social functioning, and may be linked to diminished expectations regarding treatment benefit [20].We also found a significant negative association between HADS scores and preoperative outcome expectations. Higher levels of anxiety and depression were associated with lower expectations, consistent with Riddle et al. [36], who highlighted the role of anxiety and depression in psychological networks related to expectations in TKA patients. Individuals with greater emotional distress may worry that surgery will not sufficiently relieve pain or restore function, and concerns about postoperative recovery or complications may further lower anticipated benefit. Prior work has also suggested that anxiety and depression may be associated with less favourable outcomes after TKA [37]. In addition, our analysis indicated an indirect association of PCS with preoperative outcome expectations via HADS, consistent with the idea that catastrophic pain-related cognitions can contribute to anxiety and depressive symptoms by amplifying perceived threat and uncontrollability of pain [20].

As with disability, reverse causation is also plausible for psychological factors. Although greater pain catastrophizing and emotional distress were associated with lower preoperative outcome expectations, expectations may also shape psychological responses. Patients with low expectations may experience increased anxiety and depressive symptoms due to uncertainty and fear of an unsatisfactory surgical result (e.g., “what if this does not work?”), whereas those with high expectations may feel more optimistic and emotionally reassured. Moreover, stable traits such as dispositional optimism or self-efficacy—neither of which was measured in this study—could influence both expectations and the way patients perceive and report symptoms and emotional distress, acting as common antecedents. Evidence from arthroplasty research suggests that preoperative expectation-related constructs and self-efficacy are intertwined with postoperative outcomes, supporting the plausibility of such shared pathways [38]. Dispositional optimism/pessimism has also been linked to functional outcomes after TKA, further indicating that stable psychological tendencies may simultaneously shape expectations and symptom appraisal [39]. Longitudinal studies with repeated assessments before and after surgery are needed to clarify temporal ordering and test bidirectional pathways. Clinically, these findings underscore the importance of recognising and addressing preoperative psychological distress. Interventions targeting catastrophizing and anxiety/depression, together with structured expectation-setting (e.g., education focused on realistic expectations), may help patients develop more adaptive expectations and potentially improve postoperative satisfaction and perceived outcomes [40].

Limitations

Several limitations should be acknowledged. First, this study employed a cross-sectional design, with all variables assessed at a single preoperative time point. Consequently, causal relationships among knee function, psychological factors, and preoperative outcome expectations cannot be determined, and all findings should be interpreted as associations rather than evidence of causality. Second, the possibility of reverse causation and bidirectional relationships cannot be excluded. Although the structural equation model was specified based on theoretical assumptions, patients’ outcome expectations may also influence symptom perception, emotional responses, and self-reported measures. In addition, unmeasured stable traits, such as dispositional optimism or self-efficacy, may act as common antecedents that influence both expectations and how patients perceive and report symptoms and emotional distress, thereby contributing to the observed associations. Third, participants were recruited using convenience sampling from tertiary hospitals in a single region, which may limit the generalizability of the findings. Furthermore, while structural equation modelling allows examination of complex associations, the specified directional paths represent theoretical assumptions rather than empirically confirmed causal pathways. Future longitudinal studies with repeated assessments before and after surgery, as well as interventional research, are needed to clarify temporal ordering and test bidirectional and causal mechanisms.

Conclusion

In summary, this study demonstrated that knee function, pain catastrophizing, anxiety, and depression were associated with preoperative outcome expectations among patients undergoing TKA. These findings suggest that physical symptoms and psychological characteristics jointly contribute to patients’ expectations before surgery.

Given the cross-sectional design, causal relationships cannot be established. Future longitudinal studies are needed to further clarify the temporal and potentially bidirectional relationships among these variables. Clinically, comprehensive preoperative assessment and individualized expectation-focused communication may help patients develop more realistic and adaptive expectations before TKA.

Authors’ contributions

Every author has contributed to the formulation and design of the study. YC, TN, SW, and LA played pivotal roles in the study design formulation, while TN and ZY were responsible for data collection. The data analysis was conducted by YC, LA, and SW. YC and SW examined the manuscript. All authors have engaged with and endorsed the final manuscript.

Funding

This work was supported by Health Commission of Hunan Province of China (No. D202314026462 and 202114021174) and the Natural Science Foundation of Hunan Province, China (No. 2024JJ9281).

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This cross-sectional study was conducted in accordance with the principles outlined in the Declaration of Helsinki and received approval from the Medical Ethical Committee of Hunan Provincial People’s Hospital (No. 2023-34). All participants provided their informed consent before engaging in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.