Influence of mental and psychological states on surgical outcomes following open reduction and internal fixation for trimalleolar ankle fractures

Abstract

BACKGROUND

Ankle fractures, particularly trimalleolar fractures, represent some of the most severe lower extremity injuries and frequently require surgical intervention. Although open reduction and internal fixation (ORIF) reliably restores joint stability and function, postoperative outcomes exhibit considerable variability among patients. Emerging evidence indicates that preoperative psychological status may affect surgical prognosis. Nonetheless, the specific effects of anxiety and depression on outcomes following ORIF for trimalleolar fractures remain poorly understood.

AIM

To assess the impact of mental and psychological states on surgical outcomes in patients undergoing ORIF for trimalleolar fractures.

METHODS

A total of 525 patients with trimalleolar fractures who underwent surgery at Xi’an Honghui Hospital (October 2018-December 2023) were included. Patients were divided into two groups based on preoperative anxiety or depression: Group A (with anxiety/depression) and group B (without). Outcomes were assessed using Hospital Anxiety and Depression Scale, Visual Analogue Scale, American Orthopaedic Foot & Ankle Society Forefoot Score, EuroQol Five Dimensions Five Levels, and Central Sensitization Inventory. Spearman correlation analysis was used to examine the relationship between psychological status and functional outcomes.

RESULTS

Of the 525 patients, 467 completed the follow-up, yielding a follow-up rate of 88.95%. During the follow-up period, five patients developed postoperative limb swelling, and two patients experienced wound infections. Of the 467 patients, 231 (49.47%) had preoperative anxiety or depression symptoms. Postoperative assessment metrics showed marked improvement in both groups; however, the overall prognosis observed in group A remained inferior to that of group B.

CONCLUSION

ORIF for trimalleolar fractures effectively reduced pain, restored functional mobility, and improved patients’ psychological well-being. Patients presenting with anxiety or depression before surgery tended to experience less favorable prognostic outcomes.

Key Words: Trimalleolar fracture; Open reduction and internal fixation; Mental and psychological states; Prognosis; Retrospective study

Core Tip: This retrospective study of 467 patients demonstrated that preoperative anxiety or depression independently predicts poorer postoperative outcomes following open reduction and internal fixation for trimalleolar ankle fractures. Outcomes include increased pain, reduced functional recovery, heightened central sensitization, and diminished quality of life. Patients presenting with both anxiety and depression exhibited the worst prognosis, underscoring the importance of preoperative psychological assessment.

INTRODUCTION

Ankle fractures are frequent traumatic injuries, representing approximately 10% of all fractures in the body[1]. These fractures are typically caused by direct trauma. Among them, trimalleolar fractures, which involve the medial, lateral, and posterior malleoli, are among the most severe. They often result in ankle instability, chronic pain, traumatic arthritis, and other complications[2]. Open reduction and internal fixation (ORIF) has become the standard treatment for trimalleolar fractures. Compared with conservative treatments, which are associated with ankle instability and secondary arthritis[3], ORIF restores the anatomical structure of the ankle joint and significantly improves patient prognosis[4]. Despite standardized surgical treatment, some patients continue to experience persistent postoperative pain and poor functional recovery, underscoring the need for a more comprehensive understanding of the factors influencing patient prognosis.

ORIF is the primary surgical treatment for ankle fractures, designed to restore the stability and anatomical alignment of the ankle joint[5] and reduce the incidence of complications[6,7]. Egol et al[8] found that ORIF for trimalleolar fractures effectively restores the morphology and anatomical alignment of the tibia and fibula, improves ankle joint function, enhances quality of life, and reduces the incidence of ankle arthritis[9]. Research has indicated that postoperative outcomes are affected by multiple variables, including comorbid medical conditions such as diabetes[10], hypertension[11], and osteoporosis[12,13], in addition to patients’ psychological status, particularly manifestations of anxiety and depression[14]. Trimalleolar fracture is a more severe type of ankle fracture that can significantly impact a patient’s daily life, including activities such as studying, working, and daily living. This can reduce the patient’s sense of participation and increase the likelihood of negative emotions, such as anxiety and depression, which hinder recovery after surgery. According to Manjiani et al[15], individuals experiencing persistent chronic pain frequently present with psychological disturbances, particularly anxiety and depression. A strong association has been reported between pain severity and the occurrence of depressive symptoms: As the intensity of chronic pain increases, the probability of depression correspondingly rises, thereby substantially diminishing patients’ quality of life[16]. Numerous studies have shown that poor mental and psychological health can adversely affect the prognosis of patients undergoing various surgical treatments, including joint replacement[17,18], spinal surgery[19,20], and hip joint surgery[21].

Surgical treatment methods for trimalleolar fractures are well-established in clinical practice. Nevertheless, the extent to which patients’ preoperative psychological and emotional conditions influence postoperative outcomes and overall prognosis has not yet been fully elucidated. Accordingly, the present study seeks to systematically evaluate patients with trimalleolar fractures treated with ORIF, with the objective of determining the prevalence of adverse psychological states and examining their impact on surgical efficacy. The results are expected to offer a scientific foundation for optimizing surgical management, improving patient prognosis, and minimizing postoperative complications in clinical settings.

MATERIALS AND METHODS

General information

This retrospective study received approval from the Ethics Review Committee of Xi’an Honghui Hospital, No. 2025-KY-030-01. Clinical data were retrospectively obtained from patients with trimalleolar fractures who underwent ORIF in the Department of Foot and Ankle Surgery at Xi’an Honghui Hospital between October 2018 and December 2023. Given the retrospective nature of the study and the use of anonymized patient information, the Ethics Committee exempted the requirement for written informed consent.

Inclusion criteria: (1) Trimalleolar fractures confirmed through imaging examinations and injury history; (2) Patients and their families voluntarily consented to participate in this study; and (3) Patients had no significant organ dysfunction or failure.

Exclusion criteria: (1) Old, pathological, or open trimalleolar fractures, and bilateral trimalleolar fractures; (2) Pregnant patients, patients with short-term fertility intentions, and those who were lactating; (3) Patients with psychological disorders or mental illnesses; and (4) Patients with pre-existing conditions such as gout, ankle arthritis, bone tuberculosis, bone tumors, or those who had a history of long-term corticosteroid use. All surgeries were performed by the same experienced medical team of foot and ankle surgeons.

Assessment instruments and patient group classification

Following approval from the Ethics Committee, patients’ contact information was retrieved through the hospital’s medical record system, and eligible individuals were successfully invited to participate in the study. Collected data included demographic characteristics, such as age and sex, together with questionnaire responses related to subjective symptoms and functional performance. Anxiety and depression were evaluated preoperatively and again at the final follow-up using the Hospital Anxiety and Depression Scale (HADS). This instrument comprises two subscales, HADS-A for anxiety and HADS-D for depression, each containing seven items scored from 0 to 3. A total subscale score of ≥ 8 was regarded as the threshold indicating the presence of anxiety or depressive symptoms[22]. Pain intensity before surgery and at the final follow-up was assessed using the Visual Analogue Scale (VAS). The VAS consists of a 10-cm horizontal line, with 0 indicating no pain and 100 representing the most severe pain imaginable. Patients were asked to indicate their perceived pain severity on the line according to their subjective experience[23]. Functional recovery at the final follow-up was evaluated using the American Orthopaedic Foot & Ankle Society Forefoot Score (AOFAS), a quantitative tool widely used to assess foot function and monitor treatment-related improvement[24]. Postoperative quality of life was measured with the EuroQol Five Dimensions Five Levels (EQ-5D-5 L), which includes five domains: Mobility, self-care, usual daily activities, pain/discomfort, and anxiety/depression, with each dimension graded across five levels[25]. To further examine the influence of postoperative pain on prognosis, the Central Sensitization Inventory (CSI) was administered. The abbreviated nine-item version was adopted in this study, with each item scored from 0 (“never”) to 4 (“always”), yielding a maximum total score of 36. Higher scores reflected more severe symptoms of central sensitization[26]. In addition, all evaluators received standardized training to ensure the reliability and consistency of scoring. Questionnaire responses and clinical information were subsequently entered into an electronic database using a standardized protocol, and all data were anonymized to maintain accuracy and protect patient confidentiality.

Based on preoperative HADS scores, patients were categorized according to their psychological states. Patients with both subscale scores < 8 were classified as the control group (group B). Patients with a subscale score ≥ 8 were classified as having anxiety or depression (group A). For subgroup analyses, group A was further divided into three subgroups: Anxiety only (HADS-A ≥ 8, HADS-D < 8), depression only (HADS-D ≥ 8, HADS-A < 8), and both anxiety and depression (both subscales ≥ 8). This classification allowed for a more detailed assessment of the differential impact of specific psychological symptoms on surgical outcomes.

Surgical plan

A longitudinal incision measuring approximately 12 cm was created along the lateral aspect of the left malleolus. The skin, subcutaneous layers, and fascial tissues were sequentially incised, after which blunt dissection of the peroneal muscles was performed to adequately expose the distal segment of the fibula. Subsequently, the fracture margins were cleared of debris, anatomically reduced, and provisionally stabilized using a Kirschner wire. A fibular plate was applied and secured with screws, achieving satisfactory fixation. The posterior malleolus was exposed via the interval between the peroneal and extensor tendons, reduced, and fixed with cannulated screws or bone needles. C-arm fluoroscopy confirmed optimal reduction and fixation. A 6 cm curved incision was then made over the medial malleolus. After stepwise dissection, the periosteum was elevated to expose the fracture. Displaced fragments were debrided, reduced, and stabilized with cannulated screws or bone needles. A syndesmotic screw was placed when the syndesmosis was unstable. Following confirmation of stable fixation, the wounds were irrigated with hydrogen peroxide and saline, the fascial flap was trimmed, a drain was placed, and the wounds were closed in layers. The tourniquet was then released.

Statistical analysis

Statistical analyses were conducted using SPSS version 27.0 (IBM, NY, United States). The distribution of continuous variables was initially examined using the Shapiro-Wilk test (α = 0.05) to determine normality, while equality of variances was assessed using Levene’s test (α = 0.10). For variables demonstrating normal distribution, results were expressed as mean ± SD, and intergroup or intragroup comparisons were performed using independent-samples or paired t-tests, respectively. In these cases, 95% confidence intervals (CIs) and Cohen’s d effect sizes (mean difference) were also reported. For data that were not normally distributed, results were summarized using the median (interquartile range). Comparisons were conducted with the Mann-Whitney U test for independent samples or the Wilcoxon signed-rank test for paired observations. Additionally, the Hodges-Lehmann median difference, corresponding 95%CI, and rank-biserial correlation coefficient (r) were calculated to estimate effect size. When comparisons involved more than two groups, the Kruskal-Wallis H test was applied, along with appropriate measures of effect magnitude. Categorical variables were presented as n (%), and differences between groups were analyzed using the χ² test or Fisher’s exact test, as appropriate. For these analyses, risk difference or odds ratio values, together with their 95%CIs, were provided. Beyond univariate statistical testing, multivariable linear regression models were constructed to investigate the independent relationship between preoperative psychological condition (group A vs group B) and postoperative outcome indicators, including VAS, AOFAS, CSI, and EQ-5D-5 L scores. Possible confounding variables, such as age, sex, body mass index (BMI), baseline VAS score, and duration of follow-up, were incorporated into the models as covariates. Model fit was assessed using R² and F-statistics.

In the subgroup analysis involving three psychological categories (anxiety alone, depression alone, and the coexistence of both conditions), non-normally distributed continuous variables were evaluated using the Kruskal-Wallis H test. When significant overall differences were detected, subsequent pairwise comparisons were performed with the Mann-Whitney U test, applying a Bonferroni adjustment to control for the effects of multiple testing. Statistical significance for overall group comparisons was defined as P < 0.05, whereas an adjusted threshold of P < 0.017 was adopted for post-hoc pairwise analyses following the Bonferroni correction.

To enhance the interpretability of group differences regardless of sample size, corresponding effect size indices were calculated and reported. In analyses involving the independent t-test, Cohen’s d was used to quantify the magnitude of differences, where values of 0.2, 0.5, and 0.8 were interpreted as representing small, moderate, and large effect sizes, respectively. For analyses conducted with the Mann-Whitney U test, the rank-biserial correlation coefficient (r) was determined using the formula r = Z/√N. The absolute values of 0.1, 0.3, and 0.5 were considered indicative of small-, medium-, and large-sized effects, respectively[27]. For paired comparisons, the same effect size standards were applied to the corresponding test statistics.

To identify factors associated with the severity of anxiety and depression within the subgroup of patients with preoperative psychological symptoms (group A), correlation analyses were performed between anxiety/depression scores and demographic characteristics (age, sex, BMI), as well as between preoperative anxiety/depression scores and postoperative outcomes (VAS changes, CSI, EQ-5D-5 L). These analyses were conducted only in group A to explore whether specific demographic or clinical factors were associated with the severity of anxiety or depression among patients already experiencing psychological distress.

All statistical procedures were conducted in accordance with established methodological standards to maintain the validity and consistency of the findings. Graphical representations and visual data summaries were produced using GraphPad Prism 10.0 software to enhance the accuracy and comprehensibility of the presented results.

RESULTS

Imaging data of patients with trimalleolar fractures

Imaging results are presented in Figures 1 and 2, which illustrate the corresponding imaging outcomes.

Figure 1 Preoperative imaging data of patients with trimalleolar fractures. A: Anteroposterior X-ray demonstrating cortical discontinuity of the fibula and medial malleolus, with lateral displacement of the talus; B: Lateral X-ray demonstrating cortical discontinuity of the posterior malleolus, with associated talar displacement.

Figure 2 Postoperative imaging data of patients with trimalleolar fractures. A and B: Anteroposterior and lateral X-rays demonstrating restoration of normal ankle joint alignment, with stable and well-positioned internal fixation and evidence of fracture healing.

General patient conditions

This study initially enrolled 525 participants, among whom 467 patients, including 215 males and 252 females, successfully completed the scheduled follow-up assessments, resulting in an overall follow-up completion rate of 88.95%. During the follow-up period, five patients developed postoperative limb swelling, which was significantly alleviated with rehabilitation and anti-swelling medication. Additionally, two patients had wound infections, which healed following wound debridement.

Among the 467 participants, 231 individuals (79 males and 152 females) presented with symptoms of anxiety or depression prior to surgery, and their median follow-up duration was 43.00 months (32.00, 60.00). In contrast, the other 236 patients (136 males and 100 females) showed no evidence of preoperative psychological symptoms, with a median follow-up period of 43.50 months (35.00, 60.00).

No statistically meaningful differences were identified between the two groups with respect to age, BMI, or duration of follow-up (all P > 0.05). Regarding age, the median difference was -3.00 (95%CI: -3.00 to 0.00), accompanied by an almost negligible effect magnitude (r = -0.01, 95%CI: -0.01 to 0.00). In a similar manner, BMI demonstrated a mean difference of -0.38 (95%CI: -0.99 to 0.22), corresponding to a small effect size (Cohen’s d = -0.12, 95%CI: -0.30 to 0.07). The length of follow-up also did not differ significantly between groups, with a median difference of -1.00 (95%CI: -4.00 to 2.00) and a minimal effect magnitude (r = 0.03, 95%CI: -0.06 to 0.13). Conversely, statistically significant intergroup differences were identified in sex distribution, preoperative VAS scores, and baseline HADS subscales (HADS-A and HADS-D) (all P < 0.05). For preoperative pain intensity measured by VAS, the median difference reached 5.00 (95%CI: 3.00-8.00), with a small effect size (r = -0.20, 95%CI: -0.30 to -0.10). More pronounced disparities were observed in psychological assessment scores prior to surgery. Specifically, the median differences were 3.00 (95%CI: 3.00-4.00) for HADS-A and 4.00 (95%CI: 4.00-4.00) for HADS-D. These comparisons were associated with large effect sizes (r = -0.78, 95%CI: -0.83 to -0.72 for HADS-A; r = -0.80, 95%CI: -0.85 to -0.75 for HADS-D), highlighting a substantial divergence in baseline psychological status between the two groups. Detailed comparisons of baseline demographic characteristics are summarized in Table 1, while the differences in preoperative scoring outcomes are illustrated in Figure 3A.

Figure 3 Comparison of evaluation indicators between the two patient groups. A: Preoperative; B: Postoperative. VAS: Visual Analogue Scale; HADS-A: Hospital Anxiety and Depression Scale-Anxiety; HADS-D: Hospital Anxiety and Depression Scale-Depression; AOFAS: American Orthopaedic Foot & Ankle Society Forefoot Score; CSI: Central Sensitization Inventory; EQ-5D-5 L: EuroQol Five Dimensions Five Levels.

Table 1 Baseline data of patients, mean ± SD/median (interquartile range).

Variables	Group A (n = 231)		Group B (n = 236)		t/Z/χ2	P value	Group difference (95%CI)	Effect size (95%CI)
	Male	Female	Male	Female
Sex	79 (34.20)	152 (65.80)	136 (57.63)	100 (42.37)	-5.07	< 0.01
Age (years)	43.00 (32.00, 57.00)		40.50 (31.00, 59.00)		-0.18	0.86	Med diff: -3.00 (-3.00 to 0.00)	r: -0.01 (-0.01 to 0.00)
BMI (kg/m2)	23.07 ± 3.49		23.45 ± 3.13		-1.25	0.21	MD: -0.38 (-0.99 to 0.22)	Cohen’s d: -0.12 (-0.30 to 0.07)
Follow-up time (months)	43.00 (32.00, 60.00)		43.50 (35.00, 60.00)		0.70	0.48	Med diff: -1.00 (-4.00 to 2.00)	r: 0.03 (-0.06 to 0.13)
Preoperative VAS	63.00 (55.00, 70.00)		57.00 (49.00, 66.00)		-4.31	< 0.01	Med diff: 5.00 (3.00-8.00)	r: -0.20 (-0.30 to -0.10)
Preoperative HADS-A	8.00 (7.00, 9.00)		5.00 (4.00, 6.00)		-16.90	< 0.01	Med diff: 3.00 (3.00-4.00)	r: -0.78 (-0.83 to -0.72)
Preoperative HADS-D	7.00 (7.00, 9.00)		5.00 (3.00, 5.00)		-17.40	< 0.01	Med diff: 4.00 (4.00-4.00)	r: -0.80 (-0.85 to -0.75)

CI: Confidence interval; BMI: Body mass index; VAS: Visual Analogue Scale; HADS-A: Hospital Anxiety and Depression Scale-Anxiety; HADS-D: Hospital Anxiety and Depression Scale-Depression; Med diff: Median difference; MD: Mean difference; r: Rank-biserial correlation.

Of the 467 patients who completed follow-up, 231 (49.47%) met the criteria for preoperative anxiety or depression. Of these, 47 (20.35%) had anxiety only, 62 (26.84%) had depression only, and 122 (52.81%) had both anxiety and depression. The distribution of psychological symptoms is summarized in Table 2.

Table 2 Distribution of preoperative psychological symptoms in group A patients.

Psychological states	n (%)
Anxiety only (HADS-A ≥ 8, HADS-D < 8)	47 (20.35)
Depression only (HADS-D ≥ 8, HADS-A < 8)	62 (26.84)
Both anxiety and depression (both ≥ 8)	122 (52.81)
Total	231 (100)

HADS-A: Hospital Anxiety and Depression Scale-Anxiety; HADS-D: Hospital Anxiety and Depression Scale-Depression.

Multivariable regression analysis of baseline patient data

A multivariable linear regression model was constructed to determine whether preoperative psychological status served as an independent predictor of postoperative outcomes after controlling for potential confounding variables, including age, sex, BMI, baseline VAS score, and follow-up duration. As shown in Table 3, preoperative psychological state (anxiety/depression) was an independent predictor of all four postoperative outcomes: Postoperative VAS (β = 8.48, 95%CI: 5.17-11.79, P < 0.01), postoperative AOFAS (β = -3.12, 95%CI: -4.79 to -1.45, P < 0.01), postoperative CSI (β = 2.92, 95%CI: 1.89-3.95, P < 0.01), and postoperative EQ-5D-5 L (β = -3.38, 95%CI: -4.81 to -1.95, P < 0.01). These results indicate that preoperative anxiety or depression independently contributes to worse postoperative outcomes, even after accounting for baseline differences between groups.

Table 3 Multivariable linear regression analysis of factors associated with postoperative outcomes.

Outcome	Variable	β (95%CI)	Standardized β	P value
Postoperative VAS
	Group (anxiety/depression vs control)	8.48 (5.17-11.79)	0.28	< 0.01
	Preoperative VAS	0.21 (0.12-0.30)	0.22	< 0.01
	Sex (female vs male)	2.18 (-0.84 to 5.20)	0.06	0.16
	Age	-0.03 (-0.08 to 0.02)	-0.05	0.24
	BMI	0.12 (-0.15 to 0.39)	0.04	0.39
	Follow-up time	-0.01 (-0.04 to 0.02)	-0.02	0.60
Postoperative AOFAS
	Group (anxiety/depression vs control)	-3.12 (-4.79 to -1.45)	-0.19	< 0.01
	Preoperative AOFAS	0.18 (0.09-0.27)	0.16	< 0.01
	Sex (female vs male)	-1.40 (-3.13 to 0.33)	-0.07	0.11
	Age	0.01 (-0.03 to 0.05)	0.02	0.65
	BMI	-0.08 (-0.28 to 0.12)	-0.04	0.44
	Follow-up time	0.02 (-0.01 to 0.05)	0.04	0.31
Postoperative CSI
	Group (anxiety/depression vs control)	2.92 (1.89-3.95)	0.37	< 0.01
	Preoperative VAS	0.08 (0.04-0.12)	0.22	< 0.01
	Sex (female vs male)	0.43 (-0.60 to 1.46)	0.05	0.41
	Age	-0.01 (-0.04 to 0.02)	-0.04	0.55
	BMI	0.05 (-0.08 to 0.18)	0.04	0.45
	Follow-up time	-0.01 (-0.03 to 0.01)	-0.03	0.49
Postoperative EQ-5D-5 L
	Group (anxiety/depression vs control)	-3.38 (-4.81 to -1.95)	-0.25	< 0.001
	Preoperative VAS	-0.12 (-0.19 to -0.05)	-0.18	< 0.001
	Sex (female vs male)	-0.82 (-2.32 to 0.68)	-0.05	0.28
	Age	0.02 (-0.01 to 0.05)	0.05	0.21
	BMI	-0.06 (-0.22 to 0.10)	-0.04	0.46
	Follow-up time	0.01 (-0.01 to 0.03)	0.02	0.53

CI: Confidence interval; BMI: Body mass index; VAS: Visual Analogue Scale; AOFAS: American Orthopaedic Foot & Ankle Society Forefoot Score; CSI: Central Sensitization Inventory; EQ-5D-5 L: EuroQol Five Dimensions Five Levels.

Comparison of preoperative and final follow-up VAS, HADS-A, and HADS-D scores

Postoperative evaluations revealed marked reductions in VAS, HADS-A, and HADS-D scores in both groups, with all comparisons demonstrating large effect magnitudes (r values ranging from -0.59 to -0.62). These results suggest that ORIF not only contributed to meaningful pain relief and functional recovery of the foot and ankle but also produced notable improvements in patients’ psychological status following treatment for trimalleolar fractures. All observed changes reached statistical significance (P < 0.01). Detailed outcome data are summarized in Table 4.

Table 4 Comparison between the two patient groups before the operation and at the last follow-up, median (interquartile range).

Variables		Pairing		Z	P value	Group difference (95%CI)	Effect size (95%CI)
		Preoperative	Last follow-up
Group A (n = 231)	VAS; HADS-A; HADS-D	63.00 (55.00, 70.00)	19.00 (12.00, 24.00)	-13.18	< 0.01	Med diff: 44.00 (42.00-46.00)	r: -0.61 (-0.55 to -0.66)
		8.00 (7.00, 9.00)	3.00 (3.00, 5.00)	-13.20	< 0.01	Med diff: 4.00 (4.00-5.00)	r: -0.61 (-0.66 to -0.56)
		7.00 (7.00, 9.00)	5.00 (3.00, 7.00)	-13.16	< 0.01	Med diff: 4.00 (3.00-4.00)	r: -0.61 (-0.65 to -0.56)
Group B (n = 236)	VAS; HADS-A; HADS-D	57.00 (49.00, 66.00)	8.00 (5.00, 12.00)	-13.32	< 0.01	Med diff: 48.00 (47.00-50.00)	r: -0.62 (-0.66 to -0.57)
		5.00 (4.00, 6.00)	2.00 (1.00, 2.00)	-13.29	< 0.01	Med diff: 3.00 (3.00-3.00)	r: -0.62 (-0.66 to -0.58)
		5.00 (3.00, 5.00)	1.00 (1.00, 2.00)	-12.82	< 0.01	Med diff: 3.00 (3.00-3.00)	r: -0.59 (-0.63 to -0.55)

CI: Confidence interval; VAS: Visual Analogue Scale; HADS-A: Hospital Anxiety and Depression Scale-Anxiety; HADS-D: Hospital Anxiety and Depression Scale-Depression; Med diff: Median difference; r: Rank-biserial correlation.

Comparison of postoperative outcomes across psychological states

To further explore the differential impact of specific psychological symptoms, we compared postoperative outcomes among the three psychological subgroups (anxiety only, depression only, and both). As shown in Table 5, patients with both anxiety and depression had the worst outcomes across all measures, including significantly higher postoperative VAS scores (median 21.0 vs 15.0 and 18.0, P < 0.01), lower AOFAS scores (91.0 vs 95.0 and 93.0, P < 0.01), higher CSI scores (8.0 vs 5.0 and 6.0, P < 0.01), and lower EQ-5D-5 L scores (90.0 vs 94.0 and 92.0, P < 0.01) compared with patients with anxiety only or depression only. These findings suggest that the coexistence of anxiety and depression is associated with particularly poor surgical outcomes.

Table 5 Comparison of postoperative outcomes across psychological states in subgroups, median (interquartile range).

Outcome	Anxiety only (n = 47)	Depression only (n = 62)	Both (n = 122)	P value
Postoperative VAS	15.0 (10.0, 20.0)	18.0 (12.0, 23.0)	21.0 (14.0, 26.0)	< 0.01
Postoperative AOFAS	95.0 (92.0, 98.0)	93.0 (90.0, 96.0)	91.0 (88.0, 95.0)	< 0.01
Postoperative CSI	5.0 (4.0, 7.0)	6.0 (4.0, 8.0)	8.0 (6.0, 10.0)	< 0.01
Postoperative EQ-5D-5 L	94.0 (91.0, 97.0)	92.0 (89.0, 95.0)	90.0 (87.0, 93.0)	< 0.01

VAS: Visual Analogue Scale; AOFAS: American Orthopaedic Foot & Ankle Society Forefoot Score; CSI: Central Sensitization Inventory; EQ-5D-5 L: EuroQol Five Dimensions Five Levels.

Comparison of postoperative AOFAS, VAS, HADS-A, HADS-D, CSI, and EQ-5D-5 L scores

Significant intergroup differences were identified across all postoperative assessment measures (all P < 0.01). The magnitude of these differences ranged from moderate to large, with effect sizes reported as follows: Postoperative VAS (r = -0.54), HADS-A (r = -0.60), HADS-D (r = -0.61), CSI (r = -0.47), EQ-5D-5 L (r = 0.48), and AOFAS (r = -0.40). These findings indicate that preoperative psychological states have a clinically meaningful impact on postoperative recovery, with patients without anxiety or depression (group B) achieving substantially better outcomes (Table 6, Figure 3B).

Table 6 Comparison of evaluation indicators at the last follow-up between the two groups of patients, median (interquartile range).

Variables	AOFAS	VAS	HADS-A	HADS-D	CSI	EQ-5D-5 L
Group A (n = 231)	93.00 (90.00, 97.00)	19.00 (12.00, 24.00)	3.00 (3.00, 5.00)	5.00 (3.00, 7.00)	7.00 (5.00, 9.00)	92.00 (89.00, 95.00)
Group B (n = 236)	97.00 (94.00, 98.00)	8.00 (5.00, 12.00)	2.00 (1.00, 2.00)	1.00 (1.00, 2.00)	4.00 (3.00, 6.00)	97.00 (94.00, 99.00)
Z	-9.48	-12.92	-14.48	-14.98	-11.25	11.73
P value	< 0.01	< 0.01	< 0.01	< 0.01	< 0.01	< 0.01
Group difference (95%CI)	Med diff: -4.00 (-4.00 to -3.00)	Med diff: 10.00 (9.00-11.00)	Med diff: 2.00 (2.00-2.00)	Med diff: 3.00 (3.00-3.00)	Med diff: 3.00 (2.00-3.00)	Med diff: -4.00 (-5.00 to -4.00)
Effect size (95%CI)	r: -0.40 (-0.45 to -0.35)	r: -0.54 (-0.58 to -0.50)	r: -0.60 (-0.64 to -0.56)	r: -0.61 (-0.65 to -0.57)	r: -0.47 (-0.51 to -0.43)	r: 0.48 (0.43-0.53)

CI: Confidence interval; VAS: Visual Analogue Scale; HADS-A: Hospital Anxiety and Depression Scale-Anxiety; HADS-D: Hospital Anxiety and Depression Scale-Depression; AOFAS: American Orthopaedic Foot & Ankle Society Forefoot Score; CSI: Central Sensitization Inventory; EQ-5D-5 L: EuroQol Five Dimensions Five Levels; Med diff: Median difference; r: Rank-biserial correlation.

At the final follow-up assessment, statistically meaningful differences were identified between the two groups for the pain and functional subscale scores, whereas there was no significant variation detected in alignment outcomes. Regarding the pain component, both groups demonstrated a median score of 40.00; however, group A exhibited greater variability, with values ranging from 30.00 to 40.00, while group B showed a narrower distribution (40.00-40.00). This comparison yielded a moderate effect size (r = 0.36, 95%CI: 0.27-0.44; P < 0.01). For the functional subscore, group B achieved a slightly higher median score (48.00 vs 47.00), corresponding to a small effect magnitude (r = 0.13, 95%CI: 0.04-0.22; P < 0.01). In contrast, there was no statistically significant difference in alignment subscore between the two groups (r = 0.05, 95%CI: -0.04 to 0.14; P = 0.312). Detailed results for these comparisons are summarized in Table 7.

Table 7 Comparison of evaluation indicators at the last follow-up between the two patient groups, median (interquartile range).

Variables	Pain	Function	Alignment
Group A (n = 231)	40.00 (30.00, 40.00)	47.00 (44.00, 48.00)	10.00 (10.00, 20.00)
Group B (n = 236)	40.00 (40.00, 40.00)	48.00 (47.00, 50.00)	10.00 (10.00, 10.00)
Z	7.74	2.89	1.01
P value	< 0.01	< 0.01	0.31
Group difference (95%CI)	Med diff: 0.00 (-10.00 to 0.00)	Med diff: -1.0 (-2.0 to 0.0)	Med diff: 0 (0-10.0)
Effect size (95%CI)	r: 0.36 (0.27-0.44)	r: 0.13 (0.04-0.22)	r: 0.05 (-0.04 to 0.14)

CI: Confidence interval; Med diff: Median difference; r: Rank-biserial correlation.

Comparison of the differences in VAS, HADS-A, and HADS-D scores before and after surgery

Statistical analysis identified significant differences between the two groups in the magnitude of change in VAS, HADS-A, and HADS-D scores (all P < 0.01). These findings suggest that patients’ mental and psychological status plays an important role in the recovery process of foot and ankle function. Notably, the reduction in VAS scores was significantly greater in group B than in group A, with a moderate effect size (Cohen’s d = -0.42), indicating a clinically meaningful advantage in pain improvement among patients without preoperative psychological symptoms. However, the differences in HADS-A and HADS-D changes between the groups showed small to moderate effect sizes (r = -0.31 and -0.22, respectively), suggesting that while both groups experienced psychological improvement postoperatively, the magnitude of improvement was only modestly influenced by baseline psychological states. The differences in the evaluation indicators before surgery and at the final follow-up between the groups are presented in Table 8.

Table 8 Comparison of changes in each evaluation index between the two groups of patients before surgery and at the last follow-up, mean ± SD/median (interquartile range).

Variables	VAS changes	HADS-A changes	HADS-D changes
Group A (n = 231)	41.85 ± 0.91	4.00 (3.00, 5.00)	3.00 (2.00, 4.00)
Group B (n = 236)	47.43 ± 0.84	3.00 (2.00, 4.00)	3.00 (2.00, 4.00)
t/Z	-4.34	-6.87	-4.70
P value	< 0.01	< 0.01	< 0.01
Group difference (95%CI)	MD: -5.58 (-8.02 to -3.13)	Med diff: 1.00 (1.00-1.00)	Med diff: 1.00 (0.00-1.00)
Effect size (95%CI)	Cohen’s d: -0.42 (-0.60 to -0.23)	r: -0.31 (-0.35 to -0.26)	r: -0.22 (-0.26 to -0.17)

CI: Confidence interval; VAS: Visual Analogue Scale; HADS-A: Hospital Anxiety and Depression Scale-Anxiety; HADS-D: Hospital Anxiety and Depression Scale-Depression; Med diff: Median difference; r: Rank-biserial correlation; MD: Mean difference.

Correlation analysis of mental and psychological states with clinical variables

Within group A, correlation analyses indicated that neither age nor BMI was significantly associated with preoperative anxiety or depression levels. However, there was a statistically significant relationship between sex and depressive symptoms (r = -0.22, P < 0.01), suggesting that female patients tended to report higher depression scores compared with male patients. In contrast, there was no significant association observed between sex and anxiety (r = -0.11, P = 0.10). Preoperative anxiety demonstrated a weak negative correlation with the degree of postoperative pain improvement as measured by changes in VAS scores (r = -0.17, P = 0.01), indicating that higher anxiety levels before surgery were associated with less pronounced pain reduction. Conversely, preoperative depression showed no significant relationship with VAS improvement (r = -0.04, P = 0.60). Moderate to strong associations were identified between preoperative psychological status and postoperative clinical outcomes. Specifically, both preoperative anxiety and depression were positively correlated with postoperative CSI scores (r = 0.52 and r = 0.51, respectively; both P < 0.01), indicating a higher likelihood of central sensitization symptoms among patients with poorer baseline psychological status. Additionally, preoperative anxiety and depression were negatively correlated with postoperative EQ-5D-5 L scores (r = -0.48 and r = -0.57, respectively; both P < 0.01), reflecting reduced postoperative quality of life among these patients. Collectively, these findings suggest that more severe preoperative psychological distress is associated with increased central sensitization and diminished postoperative quality of life. Detailed correlation results are presented in Tables 9 and 10, and graphical representations are shown in Figure 4.

Figure 4 Correlation between surgical prognosis of group A patients and preoperative anxiety and depression. VAS: Visual Analogue Scale; CSI: Central Sensitization Inventory; EQ-5D-5 L: EuroQol Five Dimensions Five Levels.

Table 9 Correlation between age and gender of group A patients and preoperative anxiety and depression.

Variables		Anxiety	Depression
Age	r	0.00	0.02
	P value	0.98	0.74
Sex	r	-0.11	-0.22
	P value	0.10	0.00
BMI	r	0.01	-0.03
	P value	0.90	0.66

BMI: Body mass index.

Table 10 Correlation between the degree of improvement after surgery and preoperative anxiety and depression in group A patients.

Variables		Anxiety	Depression
VAS changes	r	-0.17	-0.04
	P value	0.01	0.60
CSI	r	0.52	0.51
	P value	< 0.01	< 0.01
EQ-5D-5 L	r	-0.48	-0.57
	P value	< 0.01	< 0.01

VAS: Visual Analogue Scale; CSI: Central Sensitization Inventory; EQ-5D-5 L: EuroQol Five Dimensions Five Levels.

DISCUSSION

Ankle fractures can be classified into four types based on different classification systems. According to the Lauge-Hansen classification, third-degree ankle fractures with posterior and external rotation account for 72% of all trimalleolar fractures, making them the most common type of severe ankle injury in clinical practice[28,29]. Sundlof et al[30] found that patients with fractures often experience abnormal mental and psychological states due to the pain caused by the injury, which further affects their prognosis, a conclusion consistent with that of other researchers[31].

This study identified a statistically significant difference in sex distribution between patients in group A and group B at baseline. This may be because abnormal mental and psychological states in patients with fractures are more commonly observed in females. These findings are consistent with those reported by Navarro-Gastón and Munuera-Martínez[32], who observed that female patients demonstrated a higher prevalence of anxiety and depression when experiencing ankle-related pain. This pattern may also help explain why the preoperative VAS scores in group A were higher than those observed in group B. Similarly, a study conducted by Xie et al[33] identified a positive association between pain intensity and abnormal psychological status, further supporting the relationship between pain severity and mental health disturbances. Regarding other baseline indicators, including age, BMI, follow-up duration, and preoperative AOFAS scores, no statistically significant differences were detected between the two groups. This baseline comparability strengthens the internal validity of the study and provides a reliable foundation for subsequent comparative analyses.

This study found that the ORIF technique effectively improves ankle joint function, pain levels, and mental health status in both groups. Lewis et al[34] summarized global research on ankle fracture surgeries and found that effective internal fixation for ankle fractures restores the anatomical structure and stability of the ankle joint, alleviates patient pain, and promotes recovery of ankle joint function[35]. Similarly, Pilskog et al[36] reached a similar conclusion. To further confirm the effectiveness of this surgical method, we not only evaluated ankle joint function but also analyzed patients’ mental and psychological states. The results showed significant improvements in both ankle joint function and related psychological scores in both groups post-surgery, confirming the safety and effectiveness of the ORIF technique for trimalleolar fractures.

An important finding of this study is that the coexistence of both anxiety and depression was associated with the poorest postoperative outcomes, whereas patients with anxiety alone or depression alone had intermediate outcomes. This phenomenon suggests that the severity of psychological distress may be the key determinant of poor surgical outcomes. These findings are consistent with previous studies demonstrating that psychological comorbidity exerts a cumulative negative effect on recovery after orthopedic surgery. Patients with comorbid anxiety and depression may exhibit heightened pain catastrophizing, reduced treatment adherence, and impaired engagement in postoperative rehabilitation, all of which can adversely affect functional recovery. Therefore, we suggest that when assessing the mental health of patients, if the HADS scale score is greater than or equal to 8, especially when both subscales exceed 8, greater attention should be paid to the patient’s mental health. When necessary, a corresponding psychiatric consultation can be requested to ensure optimal surgical outcomes and prognosis.

In this study, significant differences were identified between the two patient groups across several key postoperative outcome measures, including VAS scores, AOFAS scores, central sensitization (CSI), and EQ-5D-5 L scores. Furthermore, when we performed detailed comparisons of the three dimensions of the postoperative AOFAS scores, we found significant differences in terms of pain and function between the two groups. We hypothesize that the patients in group A exhibit higher sensitivity to pain after surgery due to abnormal mental and psychological states, resulting in higher VAS and central sensitization scores. This is consistent with previous studies[37-42]. The abnormal mental and psychological states, along with pain intolerance, further affect patients’ compliance, as their inability to effectively engage in ankle joint exercises post-surgery impairs ankle joint function, ultimately leading to lower AOFAS and EQ-5D-5 L scores. MacGregor et al[43] found in clinical follow-up that patients with abnormal mental and psychological states prior to surgery had lower quality of life and VAS scores after cervical disc replacement compared with patients with good mental health. Schwartz et al[44] emphasized the importance of psychological intervention for surgical patients, stating that a positive mental state can effectively improve patient prognosis and enhance quality of life.

In this study, we found that changes to the VAS score were significantly lower in group A than in group B. Although there was a difference in preoperative VAS scores between group A and group B, the difference was relatively small due to the intense pain experienced after ankle fractures. However, in terms of postoperative VAS scores, group A patients tended to exaggerate pain when observing the surgical incision and exposed bone needles, which is consistent with the findings of Lesage et al[45]. This led to a slight improvement in VAS scores, though they remained relatively high. In contrast, the postoperative VAS scores of patients in group B significantly improved and were lower. Thus, a difference in the improvement of VAS scores between the two groups was observed. Park et al[46] also found that patients who were anxious and depressed before surgery had higher pain levels following surgical intervention. Therefore, the difference in VAS scores between group A and group B is significant.

Among patients in group A, there were no statistically significant associations between age or BMI and the severity of preoperative anxiety or depression symptoms. However, female patients were more likely to experience depressive emotions, consistent with previous reports[47]. The difference in VAS values for group A patients was significantly negatively correlated with anxiety levels (r = -0.17) but not significantly correlated with depression levels. Patients with higher anxiety levels may be less adherent to postoperative immobilization protocols, resulting in excessive movement of the affected limb and consequently less improvement in pain scores. In contrast, depression primarily manifests as affective disturbance rather than behavioral changes in limb activity, which may account for its lack of significant correlation with postoperative pain outcomes.

Moreover, relevant studies have shown that patients with higher preoperative anxiety levels tend to have an increased perception of anesthesia and surgical procedures during the operation[48], which raises the likelihood of postoperative acute pain[49], further supporting this result. Additionally, this study found that postoperative central sensitization in group A patients was significantly positively correlated with preoperative anxiety and depression levels but negatively correlated with the postoperative EQ-5D-5 L score. In other words, the worse the mental and psychological states of the patients, the higher the degree of central sensitization after surgery and the lower their quality of life. Our data supports the hypothesis that patients with abnormal mental and psychological states before surgery will have a higher pain perception after surgery and exhibit poorer compliance, thereby affecting their functional rehabilitation of the ankle joint and ultimately influencing their central sensitization and quality of life. Nahle et al[50] found that patients with poor preoperative mental and psychological states often have a lower quality of life, which may be related to the amplification of pain perception, reduced daily activity, and decreased social participation due to abnormal mental and psychological states[51]. Other studies have shown that patients with poor mental and psychological states often have higher levels of central nervous system sensitivity[52]. Akeda et al[53] also found that abnormal central sensitization could amplify pain sensitivity and even cause chronic fatigue syndrome in daily life. Tanaka et al[54] further demonstrated this result. Therefore, in clinical practice, special attention should be given to the mental health assessment of patients with trimalleolar fractures to formulate individualized treatment plans, ultimately improving the clinical outcomes of surgical treatment.

This study has several limitations. First, as a retrospective observational study, causal inference is inherently limited. Second, baseline characteristics were not fully balanced between the groups; group A had a higher proportion of female patients and higher preoperative pain scores, which may independently influence postoperative outcomes. Although multivariable regression analysis was performed to adjust for these potential confounders, residual confounding cannot be excluded. Third, a loss-to-follow-up rate of approximately 11% (58/525) was observed. Given the lack of psychological assessment data for these patients at the time of dropout, we were unable to determine whether their psychological states differed systematically from those who completed follow-up, which may have introduced selection bias and potentially influenced the generalizability of our findings. Future prospective studies should implement standardized psychological screening at each follow-up time point and employ strategies such as regular patient contact and remote assessment tools to minimize attrition and enable a more comprehensive analysis of dropout-related factors. Fourth, while this study categorized patients based on HADS scores, the primary grouping strategy combined patients with anxiety only, depression only, and both conditions into a single group. Subgroup analyses revealed important differences among these categories; however, the sample sizes for subgroup comparisons were limited, and these findings should be considered exploratory. Fifth, the patients’ evaluations of their ankle function and pain are somewhat subjective. Sixth, other confounding factors, such as fracture characteristics (e.g., Lauge-Hansen classification), comorbidities (e.g., diabetes, hypertension), and socioeconomic status, were not collected and may have influenced outcomes. Future prospective studies with larger sample sizes, comprehensive data collection, and multivariable adjustment are needed to confirm the independent predictive role of specific psychological symptoms in this patient population. Seventh, the correlation analyses were exploratory and performed only within group A, limiting the generalizability of these findings to patients without preoperative psychological symptoms. The cross-sectional nature of these correlations precludes causal inferences, and the modest magnitude of some correlations (e.g., r = -0.17 for anxiety and VAS improvement) indicates that preoperative psychological states explains only a small proportion of the variance in postoperative outcomes. Future prospective studies with larger sample sizes are needed to clarify the causal relationships and mechanistic pathways linking psychological states to surgical outcomes.

CONCLUSION

ORIF effectively alleviates pain, improves functional activity, and enhances the psychological well-being of patients with trimalleolar fractures. Moreover, patients with psychological conditions before surgery often experience worse prognosis outcomes, with the coexistence of both anxiety and depression associated with the poorest outcomes. Additionally, patients with higher preoperative anxiety levels tend to exhibit higher pain scores after surgery, and those with poor preoperative mental and psychological states also experience poorer postoperative quality of life. These findings suggest that in the clinical treatment of similar patients, psychological interventions can be adopted to relieve preoperative tension and anxiety, thereby improving both surgical outcomes and functional recovery, as well as enhancing patients’ quality of life. Special attention should be given to patients with comorbid anxiety and depression, as they may benefit from more intensive perioperative psychological support.