Correlation between Noise and Anxiety/Depression in Patients with Haemophiliac Osteoarthropathy after Hip and Knee Replacement

Abstract

Objective:

This study aimed to investigate the correlation between postoperative noise exposure and anxiety/depression in patients with haemophilic osteoarthropathy undergoing hip/knee replacement.

Methods:

This retrospective study included 58 patients with haemophilic osteoarthropathy who underwent hip/knee replacement in four tertiary hospitals between 2020 and 2025. Data were collected from clinical records. Ward noise levels (daytime/nighttime) were measured on postoperative days 1–3 by using a sound level metre. Patients were divided into high-noise (≥45 dB, n = 30) and low-noise (<45 dB, n = 28) groups. The Self-Rating Anxiety Scale (SAS), Self-Rating Depression Scale (SDS), sleep quality (Pittsburgh Sleep Quality Index, PSQI) and pain (Visual Analog Scale, VAS) were assessed. Pearson’s correlation and t-tests were used for statistical analysis.

Results:

The high-noise group had significantly higher mean noise levels (52.89 ± 6.24 dB vs. 44.57 ± 5.25 dB, P < 0.001). The SAS (51.41 ± 6.37 vs. 48.84 ± 5.23, P = 0.011) and SDS scores (54.16 ± 7.48 vs. 50.31 ± 5.25, P = 0.028) were higher in the high-noise group. Noise levels were positively correlated with anxiety (r = 0.682, P < 0.001) and depression (r = 0.659, P < 0.001). The high-noise group had poorer sleep quality (PSQI: 7.21 ± 2.35 vs. 5.19 ± 1.89, P < 0.001) and higher pain scores (VAS: 5.86 ± 1.54 vs. 4.23 ± 1.27, P < 0.001).

Conclusion:

Postoperative noise exposure is significantly associated with increased anxiety, depression, poor sleep and pain in patients with haemophiliac osteoarthropathy. Reducing ward noise may enhance their psychological well-being and postoperative recovery.

KEY MESSAGES

• (1)Postoperative noise exposure is positively correlated with increased anxiety and depression in patients with haemophilic osteoarthropathy undergoing hip/knee replacement.
• (2)Higher noise levels are associated with worsened sleep quality and increased pain.
• (3)Reducing ward noise may improve psychological outcomes and postoperative recovery in this population.

INTRODUCTION

Haemophilia is a hereditary bleeding disorder caused by coagulation factor deficiency, with patients more vulnerable to environmental factors due to chronic coagulation dysfunction, repeated surgeries and prolonged recovery.[1,2] Coagulation factor therapy has improved haemostasis, enabling more frequent hip/knee arthroplasty for haemophilic osteoarthropathy, which is primarily caused by intra-articular bleeding-induced bone damage.[3,4]

Recent studies in non-haemophiliac populations have demonstrated that postoperative noise exposure ≥45 dB is significantly associated with prolonged hospital stays, increased analgesic consumption and impaired recovery.[5,6] This evidence highlights the potential influence of environmental noise on postoperative outcomes. However, its relevance to patients with haemophilia and chronic coagulation disorders has not yet been investigated. Patients with haemophilia face unique postoperative challenges, which may amplify the impact of environmental stressors like noise.

This study aims to clarify the association between postoperative noise exposure and psychological outcomes in patients with haemophilia, providing evidence for targeted environmental interventions. Patients with haemophilia face a double burden of physical and psychological stresses. Intraoperative and postoperative monitoring of coagulation indices (e.g., coagulation factor VIII/IX levels) and frequent factor supplementation create a constant reminder of their chronic condition, amplifying anxiety about potential bleeding complications. This baseline psychological vulnerability may lower their threshold for noise-induced stress because chronic worry about disease progression could synergise with environmental disruptions.[7,8,9] Prolonged bed rest due to factor therapy may enhance the auditory perception of ward noise because restricted physical activity reduces patients’ ability to distract from environmental stimuli. This investigation aims to validate this association and provide evidence-based noise-reduction strategies to improve psychological well-being and postoperative recovery.

MATERIALS AND METHODS

Study Design and Sample Screening

The sample size was determined based on the availability of patients with haemophilia who underwent joint replacement in four tertiary hospitals over a span of 5 years. Whilst no formal power calculation was performed, the sample size was consistent with those of previous retrospective studies on rare diseases.

A total of 72 patients were retrospectively identified from the medical records of four tertiary hospitals, with recruitment spanning from February 2020 to March 2025. Amongst them, 14 were excluded due to combined infectious diseases or autoimmune diseases (n = 8), other orthopaedic surgeries during the same period (n = 4) and cognitive dysfunction/psychiatric diseases (n = 2). Ultimately, 58 patients met the inclusion criteria, and they were included in the final analysis.

The inclusion criteria were as follows: (1) conforming to the haemophilia diagnosis and treatment guideline[10]; (2) Arnold–Hilgartner stage IV or V[11]; (3) treated with hip or knee joint or hip and knee arthroplasty; (4) stable vital signs in the perioperative period (in accordance with the coagulation factor deficiency of the patients before surgery, they were supplemented with corresponding coagulation factors so that their coagulation function could meet the requirements of the surgery; the intraoperative period was closely monitored for coagulation indices and timely supplementation of coagulation factors when necessary) and (5) written informed consent.

The exclusion criteria were as follows: (1) infectious or autoimmune diseases, (2) other orthopaedic surgeries or participation in other research at the same time, (3) cognitive dysfunction or psychiatric diseases and (4) fractures of other parts of the hip and knee joints.

This study was approved by the Ethics Committees of the participating hospitals [reference numbers: EC-2020-013 (YW)-01, EC-2020-051 (YW)-01, 2024LP01299, GZTCMIII-EC-2025-023], with one approval per hospital.

Statistical Note: Chi-square tests or Fisher’s exact tests were used to compare categorical variables, and t-tests and Pearson’s correlation analysis were applied for continuous variables and correlations, respectively.

Demographic and Baseline Data Collection

Demographic and baseline data were systematically collected from the hospital information system, encompassing demographic data, surgical details and medical history. The demographic data included age, gender and educational level. Surgical details included the type of procedure (hip replacement, knee replacement or simultaneous hip and knee joint replacement); operation time and intraoperative blood loss. Medical history included the duration of haemophilia and the comorbidities. The comorbidities focused on cardiovascular and metabolic comorbidities, including coronary heart disease, hypertension and diabetes mellitus, to assess perioperative risks. Hospitalisation duration (in days) was tracked to evaluate recovery trajectories.

Data were extracted by trained investigators and cross-validated for accuracy, ensuring reliability for subsequent statistical analysis.

Noise Level Detection

A precision pulse sound level metre (model: AWA5661-2; produced by Hangzhou Aihua Instrument Co., Ltd., Hangzhou, China) was used to detect noise. The equipment was calibrated with a calibration instrument certified by the Chinese Academy of Metrology before each use (calibration certificate number: CNAS-2024-001). One sound level metre was set at each patient’s bedside, 1 metre from the bed and 1.2 metre above the floor. Measurement locations were standardised across all hospital rooms to ensure consistency. In accordance with the ISO[12] 1996-1:2016 environmental noise assessment standard, three instantaneous noise tests were conducted daily at fixed time points (07:30, 14:00 and 19:00) to cover peak activity periods (e.g., morning care, midday shift changes and evening visits), each lasting for 10 min. For nighttime noise (22:01–05:59), measurements were conducted at 23:00 and 02:00 daily (to capture potential peaks from equipment alarms or urgent care), with each test lasting 10 minutes. The average of 3 consecutive days’ measurements was used for analysis. This sampling strategy is aligned with hospital operational characteristics to better reflect actual noise exposure, balancing scientific rigor with practical feasibility. All data were synchronously recorded by two independent researchers and validated for measurement consistency through intraclass correlation coefficient (0.92). Daytime noise [06:00–22:00, consistent with the World Health Organization (WHO) guidelines] denotes the average of 07:30, 14:00 and 19:00 measurements. Nighttime noise refers to the average of 23:00 and 2:00 measurements.

Observation Indices

The noise levels in the ward during daytime and nighttime in days 1–3 of postoperative period were compared. The grouping criteria refer to the WHO[13] recommended limits of noise in hospital wards: high-noise group: 3d average noise level ≥45dB; low-noise group: 3d average noise level <45 dB. The clinical data of the two groups were collected and compared.

The difference in anxiety and depression scores between the two groups was compared at postoperative day 3.

Duan and Sheng[14] explored the clinical validity of the Self-Assessment Scale for Anxiety (SAS) and Self-Depression Scale (SDS) in 2012. SAS consists of 20 questions, divided into positive and negative scoring, and uses a 1–4 level scoring method. The scores in the 20 questions are added to obtain a raw score and then multiplied by 1.25 to obtain a standard score. The normal upper limit of the standard score is 50 points. The lower the score, the better the state. A standard score of ≥50 indicates the presence of anxiety. The SAS standard score range is as follows: <50: no anxiety; 50–59: mild anxiety; 60–69: moderate anxiety; and ≥70: severe anxiety. The Cronbach’s α values for SAS and SDS were 0.82 and 0.85, respectively, with test–retest reliability coefficients of 0.79 and 0.81, respectively.

The SDS scale consists of 20 questions, divided into positive and negative scoring, and uses a 1–4 level scoring method. The scores in the 20 questions are added to obtain a rough score and then multiplied by 1.25. The integer part is taken to obtain the standard score. The normal upper limit of the standard score is 53 points. The higher the score, the more severe the depression. The anxiety (SAS) and depression (SDS) scores were assessed at admission and on postoperative day 3 by trained nurses, following the same protocol for all participants.

The Pittsburgh Sleep Quality Index (PSQI) was administered by trained nurses each morning (09:00–10:00) to assess sleep quality over the first 3 postoperative days. This seven-item scale (0–21 points, Cronbach’s α = 0.85) evaluates subjective sleep quality, latency, duration, efficiency, disturbances, hypnotic use and daytime dysfunction, with scores >7 indicating poor sleep quality.[15]

Pain intensity was assessed daily (15:00–16:00) on the first 3 postoperative days by using a 10 cm Visual Analog Scale (VAS), where 0 = ‘painless’ and 10 = ‘most severe pain’. Trained physicians guided patients to mark their pain level, with scores independently recorded by two researchers (Kappa = 0.89) and validated via 20% random sampling for consistency.[16]

Statistical Methods

SAS (version 9.4) software (SAS Institute Inc., Cary, NC, USA) was used for statistical analysis. The normality of continuous variables was assessed using Shapiro–Wilk test. All data met normality assumptions (P > 0.05), allowing for presentation as mean ± standard deviation and analysis via t-tests. Count data were expressed as an example (%). Comparisons were performed using chi-squared (χ2) test. Fisher’s exact test was employed in cases where expected cell counts were <5 or sample sizes were small (e.g., gender distribution in Table 2). Pearson’s correlation test was used to analyse the correlation between noise and anxiety/depression.

Table 2.

Comparison of baseline characteristics between high-noise and low-noise groups [cases (%), (x ± s)]

General data	High-noise group (n = 30)	Low-noise group (n = 28)	Fisher’s exact test/t	P
Age (years)	51.83 ± 5.02	52.37 ± 4.30	t = 0.439	0.663
Gender				0.936*
Male	27 (90.00)	26 (92.86)
Female	3 (10.00)	2 (7.14)
BMI (kg/m2)	22.19 ± 1.65	22.38 ± 1.17	t = 0.503	0.617
Smoking history, n (%)	11 (36.66)	9 (32.14)	0.131	0.717
Educational level
High school or technical secondary school and below	21 (70.00)	18 (64.29)	0.215	0.643
College and above	9 (30.00)	10 (35.71)
Type of surgery			0.003	0.956
Hip replacement	12 (40.00)	11 (39.29)
Knee replacement	18 (60.00)	17 (60.71)
Operation time (min)	114.35 ± 27.86	120.04 ± 33.21	t = 0.709	0.482
Intraoperative blood loss (mL)	558.74 ± 41.56	560.47 ± 48.52	t = 0.146	0.884
Underlying disease
Coronary heart disease	1 (3.33)	1 (3.57)		0.503
Hypertension	8 (26.67)	4 (14.29)		0.402
Diabetes	3 (10.00)	4 (14.29)		0.922
Length of stay (d)	23.47 ± 4.69	24.72 ± 5.18	t = 0.965	0.339

*Using Fisher’s exact test.

RESULTS

Comparison of Daytime and Nighttime Ward Noise Levels Between Groups

The high-noise group exhibited significantly higher daytime, nighttime and overall average noise levels than the low-noise group (P < 0.05, Table 1).

Table 1.

Comparison of noise levels amongst patients with different noise exposure levels (x ± s), dB

Group	n	Average daytime noise	Average nighttime noise	Overall average noise
High-noise group	30	59.37 ± 5.20	44.41 ± 4.67	51.89 ± 6.24
Low-noise group	28	43.24 ± 1.57	39.15 ± 4.38	41.20 ± 2.98
t	-	15.749	4.417	8.228
P	-	< 0.001	< 0.001	< 0.001

Comparison of Clinical Data of Patients with Different Levels of Noise Exposure

No significant difference was found in the baseline data, such as age, gender, BMI, smoking history, haemophilia course, educational level, operation type, operation time and intraoperative blood loss, between the high-noise group and the low-noise group (P > 0.05, Table 2).

Comparison of Anxiety and Depression Scores amongst Patients with Different Levels of Noise Exposure

The postoperative SAS and SDS scores were significantly higher in the high-noise group than in the low-noise group (P < 0.05). No significant differences were observed in the baseline scores (P > 0.05, Table 3).

Table 3.

Comparison of anxiety and depression scores amongst patients with different noise exposure levels (x ± s)

Group	n	SAS score		SDS score
		On admission	Days 1–3 after surgery	On admission	Days 1–3 after surgery
High-noise group	30	56.23 ± 5.44	51.41 ± 3.21	58.42 ± 5.23	54.16 ± 7.48
Low-noise group	28	57.26 ± 4.89	48.84 ± 4.23	57.22 ± 4.97	50.31 ± 5.25
T		0.756	2.617	0.894	2.254
P		0.453	0.011	0.375	0.028

Correlation between Noise Exposure and Anxiety and Depression Score

The normality of anxiety/depression scores and noise levels was confirmed via Shapiro–Wilk test (P > 0.05), justifying the use of Pearson’s correlation analysis. As shown in Table 4, the postoperative high, low and overall noise exposure levels of patients were positively correlated with anxiety and depression scores (P < 0.05).

Table 4.

Correlation between noise exposure levels and psychological scores (Pearson’s r)

Index	Overall noise level		High-noise exposure level		Low-noise exposure level
	r	P	r	P	r	P
SAS score	0.682	<0.001	0.718	<0.001	0.594	<0.001
SDS score	0.659	<0.001	0.796	<0.001	0.609	<0.001

Comparison of Sleep and Pain between the Two Groups

The baseline PSQI and VAS scores (preoperative) showed no significant difference between the two groups (P > 0.05). The PSQI and VAS scores of the high-noise group were significantly higher than those of the low-noise group (P < 0.05, Table 5).

Table 5.

Comparison of sleep and pain between two groups

Group	N	PSQI score		VAS score
		On admission	Days 1–3 after surgery	On admission	Days 1–3 after surgery
High-noise group	30	10.05 ± 2.01	7.21 ± 2.35	7.23 ± 0.89	6.86 ± 1.04
Low-noise group	28	9.63 ± 1.88	5.19 ± 1.89	7.15 ± 0.78	6.23 ± 1.07
t		0.820	3.591	0.363	2.274
P		0.416	0.001	0.718	0.027

DISCUSSION

A high-noise environment exacerbates postoperative stress via sympathetic activation, thereby increasing heart rate/blood pressure.[17,18] In addition, psychological status is a key indicator for assessing patients’ postoperative care, and severe anxiety and depression may affect patients’ sleep quality, which, in turn, affects the speed of their recovery.[19] The present study found that the anxiety and depression scores of the high-noise group were significantly higher than those of the low-noise group. This result is consistent with the findings of Zhou et al., who revealed a negative correlation between noise exposure and postoperative recovery quality by measuring the physiological parameters (such as heart rate variability and cortisol levels) and psychological states (such as anxiety and depression scores) of patients under different acoustic conditions. For example, a high-noise environment can lead to disrupted sleep and increased stress response in patients, thereby affecting their mental health recovery.[20] Biological mechanisms may involve noise-induced stimulation of the amygdala–prefrontal cortex pathway, disrupting neurotransmitter balance.[21,22] Chronic noise exposure contributes to sleep fragmentation, which is strongly associated with depressed mood.[23] However, despite the differences in scores, no significant difference was observed in the incidence of anxiety and depression between the two groups in the present study, with SAS and SDS scores of 50 and 53, respectively. These thresholds (SAS score ≥ 50 for anxiety and SDS score ≥ 53 for depression) are derived from large-scale population-based studies, which have validated their utility in identifying anxiety and depression in general populations. However, in specific patient groups (e.g., patients with haemophilic osteoarthropathy), these thresholds may not fully and accurately reflect the actual incidence due to the characteristics of the disease itself, the treatment process and the specificity of the patient’s psychological state.

Patients with haemophilia have higher baseline anxiety (38.6%) and depression (42.3%) rates,[24] with patients with arthroplasty facing intensified stress due to prolonged recovery. Noise may exacerbate this burden via a superimposed effect, as shown in prior studies.[25] Combined with the results of the present study, noise exacerbates the psychological burden in this patient population through such a superimposed effect. Thus, postoperative care for patients with haemophilia should prioritise environmental noise control alongside routine coagulation management. Whilst factors like disease knowledge, family support and postoperative pain influence anxiety/depression, Pearson’s analysis confirmed a positive correlation between noise exposure and these scores. Mechanistically, high noise increases blood pressure/heart rate, impairs organ function recovery and disrupts brain emotion regulation (e.g., amygdala–prefrontal cortex), altering neurotransmitter balance.[26] These findings suggest that reducing ward noise (≤45 dB) is a low-cost strategy to improve psychological outcomes via interventions like quiet hours or acoustic modifications. Noise affects anxiety/depression directly and indirectly through sleep disturbance and pain exacerbation, indicating multi-pathway influence on prognosis. Even in the low-noise group (average = 41.20 dB), the noise levels were positively correlated with anxiety/depression scores (r = 0.594–0.609, P < 0.001). This finding indicates that patients with haemophilia, as individuals with high stress sensitivity, have significantly reduced tolerance thresholds to noise. Environmental noise in the range of 40–45 dB, which is generally considered to have minimal impact on healthy individuals, can still cause cumulative damage to postoperative psychological recovery by activating the sympathetic nervous system, disrupting sleep continuity and amplifying disease-related anxiety. Therefore, for this vulnerable population, the goal of ward noise control should be stricter than conventional standards. This study confirmed that postoperative noise (≥45 dB) correlates with anxiety/depression, sleep disturbance and pain in patients with haemophilia.

This study has several limitations. Firstly, its retrospective design may introduce selection bias because noise exposure was not randomly assigned. Secondly, potential confounding factors (e.g., pain severity, social support and preoperative anxiety levels) were not systematically measured, which may have influenced the results. Thirdly, the short follow-up period (3 days) limited the conclusions about long-term effects. Future prospective studies with larger samples and multivariate modelling are needed to validate the findings.

Hospital noise-control policies (e.g., ≤45 dB target) and nursing interventions to reduce environmental stressors are strongly recommended. Future research could explore targeted interventions like nighttime quiet hours (22:00–06:00), acoustic panel installations or distributing noise-cancelling earplugs to validate causal relationships and optimise cost-effective strategies.

CONCLUSION

The occurrence of postoperative anxiety and depression in patients with haemophilic osteoarthropathy who underwent hip and knee replacement is closely related to noise, and reducing the noise level may help improve their psychological status.

Availability of data and materials

The original data in the study are included in the article. Further inquiries can be directed to the corresponding author.

Author contributions

WenLong Gao: Clinical research, data collection.

HongYu Zhang: Assisted in data collection and patient follow-up.

DeJun Liu: Participated in data analysis and interpretation.

Ying Wang: Literature search, manuscript editing, study supervision.

All authors are jointly responsible for the integrity of the entire work from the beginning to the publication of the article.

Ying Wang (corresponding author) had full access to all study data and final responsibility for the decision to submit for publication. WenLong Gao and HongYu Zhang performed data curation and statistical analysis; DeJun Liu contributed to imaging data interpretation; all authors approved the final manuscript.

Ethics approval and consent to participate

This study was approved by the Ethics Committees of the participating hospitals with informed consent from all participants.

This study was approved by the Ethics Committees of the following four participating hospitals, with the corresponding ethical approval numbers and full names listed as follows:

Fourth Affiliated Hospital of Harbin Medical University: EC-2020-013 (YW)-01.

Xinlin District People’s Hospital of Greater Khingan Region: EC-2020-051 (YW)-01.

Harbin First Hospital: 2024LP01299.

The Third Affiliated Hospital of Guangzhou University of Chinese Medicine: GZTCMIII-EC-2025-023.

Conflict of interests

The authors have no conflicts of interest to declare.

Funding Statement

No.